Copyright is owned by the Author of the thesis.  Permission is given for 
a copy to be downloaded by an individual for the purpose of research and 
private study only.  The thesis may not be reproduced elsewhere without 
the permission of the Author. 
 
   
 
Fitness, Fatness and Fibre Type: 
Predicting Insulin Resistance in Māori 
 
Isaac Warbrick 
BSc (Hons) 
 
 
A Thesis Presented in Partial Fulfillment of the Requirements for the 
Degree of Doctor of Philosophy 
 
Massey University, Palmerston North, New Zealand  
 
2010 
 
 
  
 
 
 
 
 
 
‘Tama tu tama ora, tama noho tama mate’ 
 
  
Acknowledgements 
 
I am grateful to many for their support and encouragement throughout this process. 
 
Firstly, to my Heavenly Father for the energy and knowledge that he has blessed me with 
through these years, without which, none of this could have been achieved. 
 
To my wife Rachel, who has lovingly and patiently endured sharing her husband’s mental 
focus with a doctoral thesis. I am grateful for her support and love throughout this process. 
 
To my sons, Hosea and Joseph, who inspire me to keep aiming higher and to leave a legacy 
worthy of following.  
 
To my parents and family, who always encourage and support me in any endeavour I pursue. 
 
To my supervisor, Associate Professor Stephen Stannard, who has patiently worked beside 
me, and provided an example of hard work and intelligence worthy of emulation, as well as 
Professor Chris Cunningham, for his wisdom and academic experience. 
 
To the participants of my studies, for their patience and willingness to be involved in this 
research. 
 
To the Health Research Council of New Zealand (HRC), for their financial support of this 
project. 
 
And lastly, to all those who have made a contribution to my life during this time.  
 
E mihi atu ki a koutou katoa. 
 
  i
Table of Contents  
 
List of Tables  ........................................................................................................................ viii 
List of Figures  ........................................................................................................................... x 
List of Abbreviations .................................................................................................................. xi 
Explanation of Māori terms .................................................................................................... xiii 
1 Introduction ................................................................................................................ 1 
2 Literature Review .................................................................................................... 3 
2.1 Overview of Literature review ...................................................................... 3 
2.2 Diabetes and Insulin resistance in New Zealand and Globally .................. 4 
2.2.1 Definitions of diabetes & diagnostic criteria .................................................... 4 
2.2.2 Prevalence, incidence and mortality ................................................................. 6 
2.2.3 Prevalence of obesity ........................................................................................ 8 
2.2.4 Mortality ........................................................................................................... 9 
2.2.5 Inequalities in diabetes prevalence ................................................................. 11 
2.2.5.1 Age ................................................................................................................ 11 
2.2.5.2 Socio-economic status, ................................................................................. 11 
2.2.5.3 Ethnicity ........................................................................................................ 13 
2.2.6 The cost of diabetes ........................................................................................ 15 
2.2.7 Summary ......................................................................................................... 16 
2.3 Diabetes and Insulin Resistance: A metabolic disorder ............................ 17 
2.3.1 Glucose Transport and Insulin signalling ....................................................... 17 
2.3.2 Insulin resistance and its role in type-2 diabetes ............................................ 18 
2.3.3 Techniques to measure whole-body insulin sensitivity .................................. 20 
2.3.4 Summary ......................................................................................................... 21 
  ii
2.4 Body Composition and Diabetes ................................................................. 22 
2.4.1 Obesity and its role in insulin resistance ........................................................ 22 
2.4.2 Techniques used to measure body composition ............................................. 27 
2.4.2.1 BMI ............................................................................................................... 28 
2.4.2.2 Underwater weighing (Hydrodensitometry) ................................................. 33 
2.4.2.3 Bioelectrical Impedance ............................................................................... 34 
2.4.2.4 Waist circumference ..................................................................................... 37 
2.4.2.5 Skinfold thickness ......................................................................................... 38 
2.4.2.6 Dual energy x-ray absorptiometry ............................................................... 39 
2.4.2.7 Māori and body composition ........................................................................ 40 
2.4.3 Summary ......................................................................................................... 41 
2.5 Exercise and Diabetes ................................................................................... 42 
2.5.1 Fuel selection during endurance and high intensity exercise ......................... 42 
2.5.2 Effects of an acute exercise bout on glucose transport ................................... 43 
2.5.3 Effects of Chronic exercise on glucose transport ........................................... 45 
2.5.4 Aerobic Capacity, Oxidative Capacity and Insulin Resistance ...................... 46 
2.5.5 Endurance training & IMTG accumulation: A paradox ................................. 47 
2.5.6 Exercise physiology research with Māori ...................................................... 49 
2.5.7 Summary ......................................................................................................... 50 
2.6 Muscle Fibre type and Diabetes .................................................................. 51 
2.6.1 A history of fibre type and how it’s measured ............................................... 51 
2.6.2 Metabolic and functional characteristics of fibre type ................................... 55 
2.6.3 Fibre type and exercise ................................................................................... 57 
2.6.4 Fibre type and insulin resistance .................................................................... 59 
2.6.5 Fibre type proportion: Genetically or environmentally determined? ............. 60 
  iii 
2.6.6 Summary ......................................................................................................... 62 
2.7 The influence of environment and genes on diabetes prevalence ............ 64 
2.7.1 Ethnicity and health research .......................................................................... 64 
2.7.2 Environment vs. genetics ................................................................................ 65 
2.7.3 Genetic adaptation and ethnic health disparities ............................................ 67 
2.7.4 Summary ......................................................................................................... 70 
3 The Interface between Exercise Science and Māori Health ....... 71 
3.1 Introduction .................................................................................................. 71 
3.2 The Scientific Approach ............................................................................... 72 
3.3 The Holistic Approach – The Māori Approach ......................................... 74 
3.4 Exercise Science – The Bridge ..................................................................... 75 
3.5 What is happening in New Zealand? .......................................................... 78 
3.6 Implications – The Interface ........................................................................ 78 
3.7 My Research .................................................................................................. 80 
3.8 Limitations of the methodology ................................................................... 81 
3.9 Discussion ...................................................................................................... 82 
4 Hypotheses and Research Design ................................................................ 84 
4.1 Research Design ............................................................................................ 84 
4.2 Hypotheses ..................................................................................................... 87 
5 Study 1: Aerobic fitness and adiposity as predictors of fasting insulin 
concentration in healthy Māori men. Is the relationship consistent over 
time? ................................................................................................................ 88 
5.1 Introduction .................................................................................................. 88 
5.2 Methods ......................................................................................................... 91 
  iv
5.2.1 Study Design .................................................................................................. 91 
5.2.2 Participants ..................................................................................................... 91 
5.2.3 Protocol ........................................................................................................... 93 
5.2.3.1 Blood biochemistry ....................................................................................... 93 
5.2.3.2 Maximal oxygen uptake ................................................................................ 94 
5.2.3.3 Body composition ......................................................................................... 95 
5.2.4 Statistical Analyses ......................................................................................... 97 
5.3 Results ............................................................................................................ 98 
5.3.1 Body composition and aerobic capacity ......................................................... 98 
5.3.2 Blood biochemistry ...................................................................................... 100 
5.3.3 Multiple regression analysis ......................................................................... 102 
5.4 Discussion .................................................................................................... 104 
5.4.1 Chapter Summary ......................................................................................... 109 
6 Study 2: Aerobic capacity and adiposity as predictors of fasting insulin 
concentration: A comparison between glucose tolerant Māori and non-
Māori men. ................................................................................................... 110 
6.1 Introduction ................................................................................................ 110 
6.2 Methods ....................................................................................................... 113 
6.2.1 Study Design ................................................................................................ 113 
6.2.2 Participants ................................................................................................... 113 
6.2.3 Statistical Analyses ....................................................................................... 114 
6.3 Results .......................................................................................................... 114 
6.3.1 Body Composition ........................................................................................ 115 
6.3.2 Aerobic capacity ........................................................................................... 115 
6.3.3 Blood Biochemistry ...................................................................................... 116 
  v
6.3.4 Statistical Analysis ....................................................................................... 116 
6.4 Discussion .................................................................................................... 118 
6.4.1 Chapter Summary ......................................................................................... 122 
7 Study 3: Bioelectrical impedance: A valid method of measuring body 
composition in Māori, when compared to hydrodensitometry. .............. 123 
7.1 Introduction ................................................................................................ 123 
7.2 Methods ....................................................................................................... 126 
7.2.1 Participants ................................................................................................... 126 
7.2.2 Protocol ......................................................................................................... 126 
7.2.2.1 Hydrodensitometry ..................................................................................... 126 
7.2.2.2 Bioelectrical impedance ............................................................................. 126 
7.2.3 Statistical Analyses ....................................................................................... 127 
7.3 Results .......................................................................................................... 128 
7.3.1 Body composition ......................................................................................... 128 
7.3.2 Statistical Analyses ....................................................................................... 128 
7.4 Discussion .................................................................................................... 130 
7.4.1 Chapter Summary ......................................................................................... 137 
8 Study 4: The repeatability of percutaneous electrically stimulated twitch 
characteristics in healthy men .................................................................... 138 
8.1 Introduction ................................................................................................ 138 
8.2 Methods ....................................................................................................... 142 
8.2.1 Participants ................................................................................................... 142 
8.2.2 Protocol ......................................................................................................... 143 
8.2.3 Analysis of twitch ......................................................................................... 144 
8.2.4 Statistical Analyses ....................................................................................... 145 
  vi
8.3 Results .......................................................................................................... 146 
8.3.1 Reliability of twitch characteristics .............................................................. 146 
8.4 Discussion .................................................................................................... 148 
8.4.1 Chapter Summary ......................................................................................... 151 
9 Study 5: Measuring muscle contractile properties to predict insulin sensitivity in 
healthy men .................................................................................................. 152 
9.1 Introduction ................................................................................................ 152 
9.2 Methods ....................................................................................................... 155 
9.2.1 Participants ................................................................................................... 155 
9.2.2 Protocol ......................................................................................................... 155 
9.2.2.1 Electrically evoked twitch test .................................................................... 155 
9.2.2.2 Fatigue Test ................................................................................................ 156 
9.2.3 Statistical analyses ........................................................................................ 157 
9.3 Results .......................................................................................................... 158 
9.3.1 Blood Biochemistry ...................................................................................... 158 
9.3.2 Body Composition and Aerobic Capacity .................................................... 158 
9.3.3 Twitch Test ................................................................................................... 159 
9.3.3.1 Multiple regression analysis ....................................................................... 159 
9.3.3.2 Bivariate Correlation .................................................................................. 159 
9.3.4 Fatigue Test .................................................................................................. 159 
9.3.4.1 Multiple regression analysis ....................................................................... 159 
9.3.4.2 Bivariate Correlation ................................................................................. 159 
9.4 Discussion .................................................................................................... 161 
9.4.1 Chapter Summary ......................................................................................... 164 
10 Conclusions .............................................................................................................. 166 
  vii
10.1 Summary of findings .................................................................................. 166 
10.2 Applicability ................................................................................................ 167 
10.3 Implications ................................................................................................. 169 
10.4 Future Directions ........................................................................................ 172 
References  ....................................................................................................................... 175 
Appendices ............................................................................................................................ I 
10.5 Appendix A ....................................................................................................... I 
10.6 Appendix B .................................................................................................. VII 
10.7 Appendix C ................................................................................................. XIII 
10.8 Appendix D .................................................................................................. XV 
10.9 Appendix E ................................................................................................. XIX 
10.10 Appendix F ............................................................................................... XXII 
 
  viii 
List of Tables 
 
Table 2-1 – Age adjusted prevalence of diabetes within New Zealand by year ........................ 7 
Table 2-2 – Metabolic, structural & functional characteristics of different muscle fibre types 
(Adapted from Wilmore & Costill, 2005 & Zierath & Hawley, 2004) .................................... 57 
Table 5-1 – Participant physical characteristics....................................................................... 92 
Table 5-2 – Body composition and aerobic capacity ............................................................... 99 
Table 5-3 – Blood Biochemistry within group ...................................................................... 101 
Table 5-4 – Multiple regression analysis for combined results ............................................. 103 
Table 6-1 – Participant physical characteristics..................................................................... 114 
Table 6-2 – Body composition (percent body fat and total lean body mass) ........................ 115 
Table 6-3 – Aerobic capacity (absolute and relative to lean body mass) .............................. 115 
Table 6-4 – Blood Biochemistry ............................................................................................ 116 
Table 6-5 – Simultaneous multiple linear regression analysis............................................... 117 
Table 7-1 – Participant physical characteristics..................................................................... 126 
Table 7-2 – BMI and percentage body fat from underwater weighing and BIA ................... 128 
Table 8-1 – Participant physical characteristics..................................................................... 142 
Table 9-1 – Twitch speed, aerobic capacity and fatigability of varying muscle fibre types . 153 
Table 9-2 – Participant Physical characteristics .................................................................... 155 
Table 9-3 – Blood biochemistry ............................................................................................ 158 
Table 9-4 – Percent body fat, lean body mass and VO2maxLBM ......................................... 158 
Table 9-5 – Simultaneous multiple regression analysis ......................................................... 160 
Table 0-1 – Participant physical characteristics.......................................................................... I 
Table 0-2 – Blood Pathology (Sydney) .................................................................................... II 
Table 0-3 – Blood Pathology (Wellington) ............................................................................. III 
  ix
Table 0-4 – Aerobic capacity ................................................................................................... IV 
Table 0-5 – Multiple regression analysis (Sydney) .................................................................. V 
Table 0-6 – Multiple regression analysis (Wellington only) ................................................... VI 
Table 0-7 – Participant physical characteristics (non-Māori) ................................................ VII 
Table 0-8 – Participant physical characteristics (Māori) ...................................................... VIII 
Table 0-9 – Blood Pathology (Māori) ...................................................................................... IX 
Table 0-10 – Blood Pathology (non-Māori) ............................................................................. X 
Table 0-11 – Aerobic capacity (Māori) ................................................................................... XI 
Table 0-12 – Aerobic capacity (non-Māori) .......................................................................... XII 
Table 0-13 – non-Māori participants .................................................................................... XIII 
Table 0-14 – Māori participants ............................................................................................ XIV 
Table 0-15 – Participant physical characteristcs .................................................................... XV 
Table 0-16 – T50 & TPF ....................................................................................................... XVI 
Table 0-17 – T50 & TPF (continued) ................................................................................. XVII 
Table 0-18 – ANOVA table for TPF & T50 ...................................................................... XVIII 
Table 0-19 – Participant physical characteristics.................................................................. XIX 
Table 0-20 – Blood pathology ............................................................................................... XX 
Table 0-21 – Aerobic capacity, fatigue test and twitch data ................................................. XXI 
  x
List of Figures 
 
Figure 2 - Development of type-1 and type-2 diabetes mellitus ............................................. 19 
Figure 3 – Relative contribution of substrate utilized at different exercise intensities (L.J.C 
van Loon, 2004) ........................................................................................................................ 44 
Figure 4 – Group mean of lean body mass (LBM) in Sydney and Wellington study ............. 99 
Figure 5 – Group mean for VO2maxLBM in Sydney and Wellington study ........................ 100 
Figure 6 – Group mean for HOMA-IR in Sydney and Wellington study ............................. 101 
Figure 7 – Correlation between underwater weighing and BIA (entire cohort) .................... 129 
Figure 8 – Correlation between underwater weighing and BIA (Māori cohort only) ........... 129 
Figure 9 – Correlation between underwater weighing and BIA (non-Māori cohort only) .... 130 
Figure 10 - Correlation between underwater weighing and BIA (entire cohort) 95% 
confidence intervals added ..................................................................................................... 132 
Figure 11 – The relationship between body fatness and the degree to which BIA 
under/overestimates percent body fat ..................................................................................... 133 
Figure 12 – Computer analysis of electrically evoked twitch ............................................... 144 
Figure 13 – Repeatability of the T50 measure for each participant, and the variation between 
participants (each colour represents a single participant) ....................................................... 147 
Figure 14 – Repeatability of the TPF measure for each participant, and the variation between 
participants (each colour represents a single participant) ....................................................... 147 
Figure 16 – Computer Analysis of Tdiff by subtracting T60 from Fmax ............................. 157 
  xi
List of Abbreviations 
 
ADA – The American Diabetes Association 
BMI – Body Mass Index 
CO2 – Carbon dioxide 
CRP – C - reactive protein 
CVD – Cardiovascular disease 
FFA – Free fatty acid 
Gluc2hr – Glucose measurement 2 hours post-glucose load 
GlucFast – Fasting glucose 
HAD – 3-hydroxyacyl-CoA-dehydrogenase 
HDL – High-density lipoprotein 
HOMA-IR – Homeostasis model assessment – insulin reistance 
IDDM – Insulin dependant diabetes mellitus  
IMTG – Intra myocellular triacylglycerol 
Ins2hr – Insulin measurement 2 hours post-glucose load 
InsFast – Fasting Insulin 
IR – Insulin resistance 
LBM – Lean body mass 
LDL – Low-density lipoprotein 
MoH – Ministry of Health 
NIDDM – Non-insulin dependant diabetes 
NZDep – New Zealand Index of Deprivation 
NZEO – New Zealanders of European origin 
NZHS – New Zealand Health Survey 
O2 – Oxygen 
  xii
OGTT – Oral glucose tolerance test 
PFK – Phosphofructokinase 
PHO – Primary health organisation 
SDH – Succinate dehydrogenase 
SPARC – Sport and Recreation New Zealand 
T50 – Time to 50% of peak force (See Study 4 and 5)  
TPF – Time to peak force  
VO2max – Maximal aerobic capacity 
WHO – World Health Organization 
  xiii 
Explanation of Māori terms 
 
Hapu – clan, tribe, sub-tribe 
Hauora – wellness, health 
He Korowai Oranga – Māori health strategy developed to set the direction for Māori health 
development in New Zealand 
Hinengaro – mind, thoughts: In the context of Māori health, hinengaro represents mental 
health and wellbeing 
Iwi – tribe, nation, people; Often refers to a large group of people descended from a common 
ancestor  
Kapahaka – Māori cultural performance group, performing Māori dances, haka, waiata 
(songs) etc. in a group  
Karakia – chant, prayer 
Kaupapa Māori research – research methodology based on Māori ideology, values  
Rangatahi – younger generation, youth 
Tamariki – children 
Tapu – sacred, set apart, forbidden 
Te Whare Tapa Whā – Māori health model with the four aspects of health represented as the 
four walls of a house; developed by Mason Durie 
Tikanga – correct procedure, customs, habits, method 
Tinana – physical body: In context of Māori health, tinana represents physical wellbeing 
Tino rangatiratanga – The idea of self-determination, self-governance, Māori sovereignty. 
Wairua – spirit, soul: represents spiritual wellbeing in the context of Māori health 
Whakatātaka – The Māori health action plan 
Whānau – extended family, family group  
Whanaungatanga – relationship, kinship, sense of family connection 
Chapter 1  Introduction 
 1
1 Introduction 
The prevalence of type-2 diabetes has increased dramatically in recent years, in New Zealand 
and around the world. As the global burden of diabetes increases, certain ethnic groups seem 
to be hit hardest when compared with those of European descent. This is true in New Zealand 
where incidence, prevalence, death rates and complications associated with diabetes are all 
greater in Māori when compared to New Zealanders of European origin (NZEO). (Joshy & 
Simmons, 2006; Ministry of Health, 2008).  
 
Insulin resistance, a metabolic disorder which reduces disposal of circulating glucose, has 
been identified as the underlying cause for type-2 diabetes. Increased adiposity (obesity) and 
an inability to oxidize lipid appears to disrupt the cellular response to insulin, leading to 
chronic hyperglycaemia, β-cell destruction, and ultimately diabetes.  
 
While ethnicity is related to many socioeconomic factors that may contribute to increased 
prevalence of obesity and therefore diabetes, many ethnic groups appear to be more ‘insulin 
resistant’ even when socioeconomic factors are taken into account. This leads to the question 
of whether Māori and other high-risk ethnic groups have a genetic predisposition to diabetes. 
 
Along with increased adiposity, a decreased oxidative capacity (aerobic fitness of skeletal 
muscle) (C. R. Bruce et al., 2003), as well as a higher percentage of type II (fast twitch) 
muscle fibre, are associated with insulin resistance and the obesity that leads to it. Nearly all 
of the studies observing these relationships have been done in Europeans however. 
 
If identified early, insulin resistance can be treated before the disorder progresses to β-cell 
dysfunction and diabetes. Because use of health services is also low in Māori, it can be 
Chapter 1  Introduction 
 2
difficult to identify those most at risk of insulin resistance. Thus, development of non-invasive 
methods to predict risk of metabolic disorder could prove valuable in the identification and 
prevention of type-2 diabetes in Māori. 
 
Because of cultural barriers, participation of Māori in physiological research has been low, 
making it difficult to understand the physiological reasons, if any, behind ethnic health 
disparities in New Zealand. In order to overcome these barriers, the approach to research and 
intervention must also be led by Māori, and employ culturally appropriate methods. 
Chapter 2  Literature Review 
 3
2 Literature Review 
2.1 Overview of Literature review 
The aim of the first section of this review is to describe the incidence and prevalence of type-2 
diabetes, specifically in New Zealand and within Māori. This background will include a brief 
definition and the diagnostic criteria for diabetes, discuss the inequalities regarding diabetes 
that are seen within the population, and present what is being done to lessen the burden of 
diabetes within New Zealand. It will also briefly address the issue of insulin resistance, 
although it will be discussed in more detail later. 
 
The remainder of the literature review discusses the physiological mechanisms that lead to 
insulin resistance focusing on disruptions in lipid and glucose metabolism in skeletal muscle, 
and the effect of exercise on these metabolic processes. The review will also discuss possible 
physiological explanations for the disparities seen between different ethnicities in obesity and 
insulin resistance, and methods used to predict the disorder.  
 
Chapter 2  Literature Review 
 4
2.2 Diabetes and Insulin resistance in New Zealand and Globally 
2.2.1 Definitions of diabetes & diagnostic criteria 
Diabetes mellitus is considered a metabolic disorder, characterised by chronic hyperglycaemia 
and metabolic disturbances as a result of disrupted insulin secretion and/or insulin action 
(Alberti, Zimmet, & World Health Organisation, 1998). The word diabetes is derived from a 
Greek word meaning ‘to pass’ or ‘urine’, while mellitus is a Latin derivative meaning ‘honey’ 
or ‘sweet’; The combination of these words was given due to the glucose-rich urine passed by 
those with the disorder ("Stedman's medical dictionary", 2000). Diabetes is classified into a 
number of categories, including type-1, type-2 and gestational diabetes; this review will focus 
on the most common of the classifications; Type-2 diabetes, previously known as ‘non-insulin 
dependent diabetes mellitus’ (NIDDM) and ‘adult onset diabetes’.   
 
Type-1 diabetes, previously identified as ‘insulin dependant diabetes mellitus’ and ‘juvenile 
onset diabetes’, is a result of disrupted insulin secretion due to an initial disorder within 
pancreatic β-cells. This disorder can occur in non-obese individuals and most often presents in 
children. Although the cause of the initial β-cell dysfunction in type-1 diabetes is unknown, it 
is thought to be genetically linked (Atkinson & Maclaren, 1994). Because dysfunction of β-
cells disrupts insulin secretion, blood glucose levels remain chronically high signalling the 
diagnosis of diabetes. The anti-lipolytic role of insulin is also absent, so ketosis, and loss of 
body weight, are symptoms upon presentation (Alberti, Zimmet, & World Health 
Organisation, 1998). 
 
Type-2 diabetes is also characterised by decreased insulin secretion and hyperglycaemia, but 
the β-cell disorder that leads to the diabetic state stems initially from a condition known as 
Chapter 2  Literature Review 
 5
insulin resistance. In this case, blood glucose levels become chronically elevated similar to 
type-1 diabetes, but initially the glucose level remains high regardless of normally sufficient 
insulin secretion. In response to this glucose intolerance, β-cells compensate by secreting 
more insulin, and the hyperglycaemia is controlled depending on the β-cells ability to 
maintain the elevated levels of insulin needed (Reaven, 1988). Thus, in type-2 diabetes, rather 
than the disorder initiating from disrupted insulin secretion, elevated blood glucose levels 
remain high due to myocytes and adipocytes ‘resisting’ the insulin signal; hence the term 
‘insulin resistance’. Although the precise mechanisms responsible for insulin resistance are 
still not fully known, the disorder is broadly a result of obesity (B. B. Kahn & Flier, 2000; 
Reaven, 1988), and can be avoided with lifestyle change (Kelly, 2000).  
 
Since French physician Lancereaux first made the distinction between ‘fat’ and ‘thin’ diabetes 
in the late 1800s (cited in Gale, 2001), the definition of diabetes has been altered many times. 
During a meeting between The World Health Organisation (WHO) and The American 
Diabetes Association (ADA), it was recommended that the terms ‘insulin dependant diabetes 
mellitus’ and ‘non insulin dependant diabetes mellitus’ and their acronyms ‘IDDM’ and 
‘NIDDM’ be no longer used. This recommendation was made because the previously 
described terms caused confusion and often led to patients being classified according to their 
treatment rather than pathogenesis (Alberti, Zimmet, & World Health Organisation, 1998). 
Similarly, the terms ‘juvenile’ and ‘adult’ onset diabetes are no longer used because children 
are presenting with type-2 diabetes in increasing numbers and type-1 can be first diagnosed in 
adulthood. 
  
Along with its definition, the diagnostic criteria for diabetes have also changed many times, 
and variation in criteria between different health organisations has led to confusion regarding 
Chapter 2  Literature Review 
 6
its diagnosis. While symptoms that prompt diagnosis include an increased thirst and urine 
volume, recurrent infections and unexplained weight loss, clinical diagnosis of diabetes is 
often made by measuring fasting glucose concentration and performing an oral glucose 
tolerance test (OGTT). With an OGTT, blood glucose levels are measured while fasted and 
then at incremental periods for 2 hours after an oral glucose load of 75g (Matsuda & 
DeFronzo, 1997). The rationale behind the OGTT is that those who are insulin sensitive will 
return to fasting (or near fasting) glucose levels within 2 hours of the glucose load, while a 
relatively high blood glucose level, 2 hours after the glucose load, is indicative of glucose 
intolerance and diabetes. In 1998 the WHO lowered their diagnostic criteria for diabetes from 
a fasting plasma glucose concentration of 7.8 mmol L-1 and above, to the currently accepted 
value of 7.0 mmol L-1 and above, and/or a plasma concentration of 11.1 mmol L-1 and above 
for a 2-hour post glucose load (Alberti, Zimmet, & World Health Organisation, 1998).  
 
2.2.2 Prevalence, incidence and mortality  
In 1998, WHO predicted that non-communicable diseases, such as type-2 diabetes, would 
become more prevalent in developing countries due to adopted Western lifestyles and the risk 
factors, such as smoking, high energy diets and lack of exercise, that accompany such a 
lifestyle (WHO, 1998). Since then diabetes has become a global epidemic, increasing in 
prevalence and incidence in most of the developed world. One study estimated the global 
prevalence of diabetes at 2.8% in 2000, and predicted that this would increase to 4.4% in 2030 
(Wild, Roglic, Green, Sicree, & King, 2004). In New Zealand the burden of diabetes has 
become an area of concern as well, with national diabetes prevalence exceeding global 
estimates. The 1996/97 New Zealand Health Survey (NZHS) showed that 1 in 27 (3.7%) 
adults in New Zealand, were diagnosed with diabetes (Ministry of Health, 1999b). Ten years 
later, the 2006/07 survey showed that prevalence of diagnosed diabetes had increased to 
Chapter 2  Literature Review 
 7
approximately 5.0% (1 in 20 adults) (Ministry of Health, 2008). What’s more, the burden of 
this disorder will likely be compounded in years to come, as the population of the groups that 
seem to be most at risk of diabetes (Māori, Pacific Islanders, Indian and Asian) are increasing 
in numbers dramatically (Moore & Lunt, 2000). As is the case in most countries, type-2 
diabetes accounts for the majority of diabetes cases in New Zealand (Ministry of Health, 
2004b, , 2008). 
 
Table 2-1 – Age adjusted prevalence of diabetes within New Zealand by year 
Year Prevalence 
NZHS 1996/97 3.7% 
NZHS 2002/03 4.3% 
NZHS 2006/07 5.0% 
 
The prevalence of diabetes differs between ethnicities in New Zealand similar to other 
countries. The 1996/97 New Zealand Health Survey (NZHS) showed the prevalence of 
diabetes at 3.1% in Europeans compared with 8.3% of the Māori population (Ministry of 
Health, 1999b). The age standardized rate ratios in the 2006/07 NZHS (where 1.00 equals the 
national average) were 1.74 for Māori men and 1.61 in Māori women, compared to just 0.76 
in European men and 0.82 in European women (Ministry of Health, 2008). Another study 
done in South Auckland found that prevalence of the metabolic syndrome and known diabetes 
was greater than 50% in the Māori and Pacific people aged 40 years and older that were 
tested, much higher than observed in European groups (David Simmons & Thompson, 2004).  
 
Chapter 2  Literature Review 
 8
2.2.3 Prevalence of obesity  
As mentioned earlier, obesity, as a result of chronic caloric excess i.e. relatively low energy 
expenditure for energy intake, has been identified as the key factor leading to insulin 
resistance, and is therefore associated with type-2 diabetes (Hohepa, Schofield, & Kolt, 2004). 
A population-based study in the US showed that obesity was the most important risk factor 
for insulin resistance in adolescents, independent of sex, age or ethnicity (J. M. Lee, 
Okumura, Davis, Herman, & Gurney, 2006). Others have shown that ‘Western’ dietary 
patterns are associated with obesity and a higher risk of insulin resistance, while ‘healthy’ 
dietary patterns on the other hand, are associated with lower risk of insulin resistance 
(Esmaillzadeh et al., 2007). Accordingly, studies in Europe have revealed that reduction of 
weight within a population is associated with a similar reduction in diabetes incidence 
(Astrup, 2001). Intervention programs in the United States, which aimed at reducing weight 
with physical exercise and healthy diet, also reduced the risk of diabetes in those who 
participated (Hamman et al., 2006).  
 
These studies portray the close association between obesity and diabetes prevalence. It is 
therefore no surprise that the prevalence of obesity in New Zealand is increasing at a rate 
similar to the increased prevalence of insulin resistance and diabetes. Results of the 
2002/2003 New Zealand Health Survey (NZHS) showed that 35.2% of adult New Zealanders 
(15 years plus) were considered overweight, and 20.9% were obese (Parnell, Scragg, Wilson, 
Schaaf, & Fitzgerald, 2003). Four years later, the 2006/2007 NZHS showed that the 
prevalence of overweight adults in New Zealand had increased to 36.2%, and obesity had 
increased to 26.5% (Ministry of Health, 2008).  
 
Chapter 2  Literature Review 
 9
Notably, the rates of participation in sporting clubs and organised teams has fallen 
considerably (Bauman et al., 2003). Not only is participation in sport and leisure decreasing 
but advances in technology promote a sedentary lifestyle and decrease the need for physical 
labour. This decrease in daily physical activity coupled with less costly, readily available 
high-energy foods, are the primary cause for the growing prevalence of obesity and the 
accompanying co-morbidity of type-2 diabetes. 
 
2.2.4 Mortality 
Since it is the complications of diabetes that cause death and not diabetes per se, coding of 
diabetes-related mortality statistics can be flawed (Joshy & Simmons, 2006). In 2000, 3% of 
all deaths in New Zealand were attributed to diabetes (Ministry of Health, 2004a). 
Accordingly, a report prepared by the New Zealand Ministry of Health  showed that 8% of all 
New Zealand deaths in 1996/97, were associated with physical inactivity (Ministry of Health, 
1999b) making it the second highest preventable risk factor among New Zealanders (behind 
tobacco smoking).   
 
Just as many illnesses are disproportionately higher in Māori than NZEO, so too is the general 
mortality rate compared to that in non-Māori (N. Pearce, Pomare, Marshall, & Borman, 
1993). In 2004, the overall death rate was 7.0 (deaths per 1000) for all New Zealanders and 
4.3 for Māori. When adjusted for age however, age standardized death rates for Māori, who 
are younger than the general NZ population, were 7.6 per 1000 compared with 4.5 per 1000 in 
the general New Zealand population respectively (Statistics New Zealand, 2006a, , 2006c). In 
2005/07, life expectancy for Māori was 70 and 75 years for males and females respectively, 
compared with 79 years in non-Māori males and 83 in non-Māori females (Statistics New 
Zealand, 2008).  
Chapter 2  Literature Review 
 10
 
Similarly, ethnic disparities exist in diabetes-related mortality. In the United States, the risk of 
diabetes-related death in Indigenous Americans was 2.37 times higher than Caucasian 
Americans. However, even higher still was diabetes-related mortality in Māori which was 
5.81 times higher than European New Zealanders (Bramley, Hebert, Tuzzio, & Chassin, 
2005). In fact, this same study showed that, in regard to many health indicators, the disparity 
between indigenous and non-indigenous groups in New Zealand i.e. Māori and non-Māori, 
was greater than the disparity between indigenous and non-indigenous groups in the US.  
 
The reason for the higher mortality rate in Māori, even compared to other indigenous groups, 
is not clear, though Moore and Lunt speculate that delivery of medical services, and the 
absence of a Māori focus, may contribute (Moore & Lunt, 2000). Interestingly, results of the 
2006/07 NZ Health Survey showed that Māori were significantly less likely to report that a) 
their health provider had treated them with respect and dignity ‘all of the time’, b) their health 
care provider listened carefully to what they had to say ‘all the time’, and c) that their health 
care provider had discussed their health care as much as they wanted ‘all of the time’, when 
compared to the total population (Ministry of Health, 2008). In context with ethnic health 
disparities, use of medical services is partly reflected in the high prevalence of undiagnosed 
diabetes in Māori. One study showed that 7% of Māori respondents were undiagnosed 
diabetics, while 15.5% had impaired glucose tolerance (Joshy & Simmons, 2006). Another 
study found that in a cohort of Māori where 7.1% (n=247) were known diabetics, an 
additional 3.6% were undiagnosed diabetics (Tipene-Leach et al., 2004). This result is 
probably modest considering the response rate in this study (49%). Furthermore, 37% of those 
that did respond were insulin resistant. 
 
Chapter 2  Literature Review 
 11
2.2.5 Inequalities in diabetes prevalence 
2.2.5.1  Age 
Epidemiological studies show that diabetes prevalence increases with age (Ministry of Health, 
2008). Nevertheless, where type-2 diabetes was once a disease common only among older age 
groups, it is now becoming a serious problem among adolescents. This increase in the 
incidence of type-2 diabetes in adolescents and children, is directly tied to the increasing 
prevalence of childhood obesity (Rocchini, 2002). Hotu et al. observed a 6-fold increase in 
adolescents with type-2 diabetes attending the Adolescent Diabetes Service in New Zealand 
between 1996 and 2002 (Hotu, Carter, Watson, Cutfield, & Cundy, 2004). Studies also 
indicate that the onset age of diabetes in Māori is almost 10 years earlier than Europeans 
(Joshy & Simmons, 2006). Accordingly, the median age for Māori is considerably lower than 
the median for New Zealand (22.7 years compared with 35.9 years) (Statistics New Zealand, 
2006c), which may partly explain the earlier onset age of diabetes in Māori. Alternatively, 
Moore and Lunt suggest that the earlier onset age of diabetes in Māori, could be due to a 
longer lifetime exposure to conditions causing hyperglycaemia; namely caloric imbalance 
(Moore & Lunt, 2000) (See section2.2.5.2). 
  
2.2.5.2  Socio-economic status, 
The prevalence of diabetes in those that live in the most deprived quintile of the NZDep2006 
(quintile 5), is more than twice as high as those in the least deprived (quintile 1) (Ministry of 
Health, 2008). Accordingly, obesity is also highest in the most deprived quintile (Ministry of 
Health, 2008). A number of epidemiological findings may help to explain this association. 
Firstly, the 2006/07 NZHS showed that the time children spent watching television, as well as 
their ‘fizzy drink’ and ‘fast food’ consumption, were higher in areas of high neighbourhood 
deprivation (quintile 5) than in areas of low deprivation (Ministry of Health, 2008). Fittingly, 
Chapter 2  Literature Review 
 12
insulin resistance and diabetes have been shown to be associated with time spent watching 
television (Hu et al., 2001; Hu, Li, Colditz, Willett, & Manson, 2003), high consumption of 
‘soft drink’ (Schulze et al., 2004) and ‘fast food’ (Pereira et al., 2005), and a high fat/high 
sugar (high energy) diet (Gittelsohn et al., 1998; Ministry of Health, 2008).  
 
Another possible explanation is that physical activity and participation in sports and leisure 
may be perceived by the socioeconomically deprived as costly, discouraging those with low 
incomes from participating (Estabrooks, Lee, & Gyurcsik, 2003). A report by the Ministry of 
Health also suggests that those who are least deprived are better able to respond to social 
pressures associated with weight gain (Tobias, Paul, & Yea, 2006). 
 
These trends have major implications for Māori who are proportionately over represented in 
the most deprived quintile, having an annual income approximately 20% lower than NZEO 
(Statistics New Zealand, 2006b). Accordingly, Māori children spend more time watching TV, 
and consume more ‘fizzy drink’ and ‘fast food’ than the general NZ population (Ministry of 
Health, 2008). What’s more, hazardous drinking patterns are significantly higher in Māori 
adults  which compounds the ‘high-risk’ health behaviours of Māori considering the relatively 
high calorie density of alcohol and its association with obesity (P. M. Suter, Hasler, & Vetter, 
1997). Interestingly however, Māori are reportedly as active as, or more active than NZEO 
(Ministry of Health, 2008; SPARC., 2004). Another interesting finding is that the relationship 
between obesity and socioeconomic status in Māori men is counterintuitive; that is, elevated 
markers of overweight and obesity are associated with higher income and education (Tobias, 
Paul, & Yea, 2006). This suggests that the effect of socioeconomic status on obesity is 
modified by ethnicity and culture. One possible explanation for this may be the difference in 
cultural perceptions of body image and preferred body shape (Tobias, Paul, & Yea, 2006). 
Chapter 2  Literature Review 
 13
 
2.2.5.3  Ethnicity 
Diabetes research comparing ethnic groups in New Zealand paints a grave portrait for the 
future of Māori health if this epidemic is allowed to continue on its current path. Among 
Māori, prevalence of known diabetes, as well as diabetes risk factors such as obesity and low 
HDL are higher than the general New Zealand population (Ministry of Health, 2008; David 
Simmons & Thompson, 2004). The prevalence of insulin resistance is also much higher in 
Māori even when BMI and body fat percentage is similar to European New Zealanders (K. 
McAuley, Williams, Mann, Goulding, & Murphy, 2002). Though studies show Māori to have 
higher levels of fat-free mass and lower body fat percentage than Europeans of the same BMI 
(Swinburn, Ley, Carmichael, & Palnk, 1999), Māori appear to be at greater risk of obesity 
related diseases at a lower BMI. This indicates that obesity related health risk differs between 
ethnic groups, with Māori being much more sensitive to extra body fatness (adiposity). One 
study by Simmons et al. challenged these findings and showed that when adjusted for obesity, 
Māori and other Polynesian groups seem to have relatively normal insulin sensitivity and do 
not have the usual marker of hyperinsulinaemia found in other high-risk ethnic groups (D. 
Simmons, Thompson, & Volklander, 2001). This finding would suggest that the high 
prevalence of diabetes is due to increased obesity in Māori and not necessarily an ethnic 
susceptibility to risk factors such as insulin resistance. 
 
Although much debate surrounds the use of current BMI standards in Māori and other ethnic 
groups (see section 2.4.2), Māori appear much more obese than Europeans, when defined by 
current BMI standards (Moore & Lunt, 2000). The 2006/07 NZ Health Survey showed that 
prevalence of obesity within Māori was 41.7% compared with just 24.3% in Europeans 
(Ministry of Health, 2008). Pacific Islanders were even higher at 63.7%. In accordance with a 
Chapter 2  Literature Review 
 14
higher prevalence of obesity and diabetes, Māori diets are higher in fat than the general 
population (Ministry of Health, 1999a; Parnell, Scragg, Wilson, Schaaf, & Fitzgerald, 2003).  
 
Not only is diabetes more prevalent among Māori, but Māori have more than double the risk 
of the disease progressing into associated complications such as blindness and renal failure 
(Joshy & Simmons, 2006). The fact that smoking has been associated with the progression of 
diabetes complications as well, makes the current situation even worse, as the MoH have 
shown that almost half of Māori adults are smokers, compared to 23% of European adults 
(Ministry of Health, 1999b). Correspondingly, Māori and Pacific people have poorer medium 
term metabolic control (HbA1c > 8%) than Europeans (40%, 51% and 23% respectively) 
(Joshy & Simmons, 2006).  
 
Similar to New Zealand, ethnic disparities in diabetes prevalence can be seen around the 
world. For example, one study from a multicultural community in Hawaii showed that even 
when energy intake, physical activity, BMI and waist-hip ratio were accounted for, prevalence 
of type 2 diabetes and glucose intolerance was significantly higher among the Hawaiians, 
Phillipinos and Japanese when compared with those who identified as Caucasian (Grandinetti 
et al., 2007). Likewise, the prevalence of diabetes among some indigenous groups in Canada 
is more than 3 times that of the general population (Dannenbaum, Kuzmina, Lejeune, Torrie, 
& Gangbe, 2008), while the Pima Indians of Arizona are believed to have the highest 
prevalence of diabetes of any group in the world (Pratley, 2007).  
 
A number of theories which have been postulated to explain these ethnic disparities in relation 
to genetic makeup, will be discussed later in this review (See section 2.7). On the other hand, 
environmental factors such as lifestyle play a large part in ethnic health disparities. 
Chapter 2  Literature Review 
 15
Accordingly, Zimmet (Zimmet, 1979) observed that prevalence of diabetes was low in 
aboriginal people of Australia and the Pacific that continued to live a traditional, physically 
active lifestyle, but high in those who adopted a sedentary, westernized way of living. 
Similarly, O’Dea (O'Dea, 1984) showed that the metabolic abnormalities associated with 
type-2 diabetes in a group of diabetic Australian Aborigines were greatly improved, and in 
some cases normalized, with just 7 weeks of living a traditional lifestyle. One implication of 
these findings is that intense exercise is not necessary for the maintenance of health in Māori, 
but that the everyday activities of a traditional lifestyle, such as farming, cultivation of land, 
fishing and physical labour, in conjunction with a healthy diet, may suffice.  
 
Lastly, studies show that Māori are less likely to make use of government-based health 
services (Moore & Lunt, 2000). For example, the results of the Get Checked programme in 
2004 showed that 63% of European diabetics and 92% of Pacific diabetics had a free annual 
check while only 27% of Māori with diabetes took advantage of this free check (Joshy & 
Simmons, 2006). Whether this is due to expense of health care, accessibility of the service 
and/or cultural perception is not entirely clear, however, this has been discussed in more detail 
in section 2.2.4.  
 
2.2.6 The cost of diabetes 
The estimated cost of obesity related health care alone in New Zealand in 2002 was NZ$303 
million per year (Ministry of Health, 2003), however, more recent costs would no doubt be 
greater. SPARC estimated the direct cost of physical inactivity in New Zealand to be NZ $180 
million per year, compared to Australia, with an estimated cost of AUS $400 million per year 
(SPARC., 2004), proportionally lower considering the relatively larger population. Other 
figures show that the economic burden of Type-2 diabetes in New Zealand during 2001 was 
Chapter 2  Literature Review 
 16
close to NZ$400 million which is predicted to rise to NZ$1 billion by 2021 
(PricewaterhouseCoopers, 2001). Much of these costs are due to lifestyle choices and could 
be greatly reduced with relatively small lifestyle changes. In fact, one study estimated that an 
increase in physical activity as small as 5% across the population would result in savings of 
NZ$25 million per year (Bauman, 1997), though this figure may be much greater as it does 
not take into account the direct effect of physical activity upon diabetes cost. What’s more, 
lifestyle change for those most at risk of insulin resistance and diabetes, is more cost effective 
at a personal level when compared with pharmaceutical interventions (Herman et al., 2005). 
 
2.2.7 Summary 
Population data paints a grave portrait for Māori who, when compared to NZEO, are worse 
effected by diabetes at all stages. Type-2 diabetes prevention begins with the promotion of 
healthy behaviours; a balanced diet along with adequate and consistent physical activity to 
ensure maintenance of a healthy body composition (Young, Reading, Elias, & O'Neil, 2000). 
Thorough research and appropriate interventions will come at a price, but considering 
diabetes is a preventable disease, and the potentially vast economic burden stemming from 
diabetes in coming years, it would be a wise public health investment.  
Chapter 2  Literature Review 
 17
2.3 Diabetes and Insulin Resistance: A metabolic disorder 
2.3.1 Glucose Transport and Insulin signalling 
After a carbohydrate-containing meal, complex sugars (polysaccharides) are broken down in 
the gut into simple sugars (monosaccharide) such as glucose (Crabtree & Garlick, 1993). 
Following absorption into the bloodstream, blood glucose concentration becomes elevated 
and in response β-cells in the pancreas secrete insulin, the hormone responsible for signalling 
the uptake of glucose into adipocytes (fat cells) and myocytes (muscle cells) (Mathers & 
Wolever, 2002; Wilcox, 2005). Insulin is then detected by receptor proteins on the cell 
membrane of these tissues, setting off a signalling cascade which facilitates the translocation 
of GLUT4, a transport protein which allows the transport of glucose across the cell membrane 
and into the cell (Bryant, Govers, & James, 2002). After glucose enters muscle cells it is 
quickly phosphorylated, then either stored as glycogen or used as a substrate for glycolysis. 
Adipose tissue also has a role in glucose disposal and responds to insulin by storing excess 
glucose as fat (Flatt, 1995; Winegrad & Renold, 1958). Although other glucose transporters 
exist, such as GLUT1 and GLUT5, GLUT4 is the most important glucose transporter in 
skeletal muscle and adipose tissue.   
 
Along with signalling the uptake of glucose into myocytes, insulin also signals a reduction in 
the rate of lipolysis and increases the rate of esterification of free fatty acids (FFA) for storage 
as triglycerides in adipocytes (Campbell, Carlson, Hill, & Nurjhan, 2006). Similarly the liver 
responds to insulin by taking up glucose to be stored as glycogen, at the same time inhibiting 
gluconeogenesis and glucose output (Michael et al., 2000; Saltiel & Kahn, 2001). Thus, 
skeletal muscle, adipose tissue and the liver all contribute and play a major role in the 
regulation of lipid and glucose metabolism. Nevertheless, skeletal muscle is responsible for 
Chapter 2  Literature Review 
 18
up to 75% of glucose disposal following a meal (Baron, Brechtel, Wallace, & Edelman, 
1988). 
 
2.3.2 Insulin resistance and its role in type-2 diabetes 
In insulin resistant individuals the insulin-signalling pathway in the insulin-sensitive tissues 
are disrupted and insulin-mediated uptake of glucose by myocytes and adipocytes is impaired. 
Initially, insulin is secreted in response to increasing blood glucose as expected, but the 
signalling cascade within the cell, which normally results in GLUT-4 translocation to the cell 
surface, is blunted (Garvey et al., 1998; Roden et al., 1996), reducing sensitivity to the insulin 
signal, ‘insulin sensitivity’, within these myocytes and adipocytes.  
 
In response to the resultant hyperglycaemia, β-cells in the pancreas compensate by secreting 
more insulin to maintain control of blood glucose levels (Warram, Martin, Krolewski, 
Soeldner, & Kahn, 1990). This hyperinsulinaemia, if left untreated, worsens to the point 
where the pancreas cannot secrete enough insulin to maintain normal blood glucose levels, 
and blood glucose levels remain chronically elevated (Weir & Bonner-Weir, 2004). If β-cells 
cannot cope with the chronically elevated blood glucose levels and the continual resistance to 
secreted insulin, β-cell dysfunction can occur and blood glucose levels eventually reach the 
point at which diabetes is diagnosed i.e. a fasting blood glucose concentration of ≥7.0 mmol/l 
(Alberti, Zimmet, & World Health Organisation, 1998; Reaven, 1988). On the other hand, all 
persons with insulin resistance do not necessarily go on to develop hyperglycaemia. Large 
ranges exist in fasting insulin concentrations before hyperglycaemia occurs (Hollenbeck & 
Reaven, 1987), suggesting that some individuals possess a greater ‘compensatory’ ability to 
secrete insulin in response to hyperglycaemia. 
 
Chapter 2  Literature Review 
 19
Figure 1 - Development of type-1 and type-2 diabetes mellitus 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
It is believed that dysregulation of lipid metabolism, as a consequence of excess fatty acid 
(FA) availability and intracellular lipid accumulation, somehow interferes with the signalling 
cascade and translocation of GLUT4 in skeletal muscle, causing insulin resistance in 
myocytes (Boden & Shulman, 2002; Lewis, Carpentier, Adeli, & Giacca, 2002).  What’s 
more, when the liver becomes resistant to insulin, hepatic glucose release is no longer 
inhibited, increasing blood glucose further and compounding the hyperglycaemic state 
(Boden, 1997) (See Figure 1). These points highlight the importance of all three tissues 
(adipose, liver and muscle) in the development of insulin resistance and type-2 diabetes; as 
well as highlighting the complexity of the disorder which involves many different tissues and 
INSULIN RESISTANCE 
Obesity, Sedentary Lifestyle, 
Poor Nutrition 
Genetic Factors 
Hyperinsulinaemia 
Β-cell failure 
Chronic Hyperglycaemia 
TYPE-1 DIABETES  TYPE-2 DIABETES 
Genetic Factors, Disease 
Insufficient insulin 
secretion 
Type-1 diabetes 
initiated by genetic 
factors and/or 
disease 
Hepatic Glucose release 
Hepatic Insulin resistance 
Chapter 2  Literature Review 
 20
physiological processes. The pathophysiology of insulin resistance will be discussed further in 
section 2.4.1. 
 
2.3.3 Techniques to measure whole-body insulin sensitivity 
The ‘gold standard’ method for measurement of whole-body insulin sensitivity is the glucose 
clamp technique (Ferrannini & Mari, 1998). The glucose clamp, or hyperinsulinaemic-
euglycemic clamp, measures the amount of glucose that is needed to balance an artificial 
increase in insulin, without causing hypoglycaemia. Insulin is first infused into a peripheral 
vein followed by infusions of glucose at increasing rates, to compensate for the increased 
insulin and maintain blood sugar levels at 5 – 5.5 mmol/l. Glucose levels in the blood are 
monitored every 5 to 10 minutes for around 2 hours until a steady state of infusion is reached 
(DeFronzo, 1979). This usually occurs during the last 30 minutes of the test, where the ratio 
of glucose infusion to insulin concentration during this time is then calculated to determine 
insulin sensitivity (Wallace & Matthews, 2002). However, because this technique involves 
intra-venous (IV) infusion of insulin, frequent blood samples over a 2-3 hour period, and 
continuous infusions of glucose, the glucose clamp technique is not practical for large scale 
population studies. 
 
A less costly and less arduous means of estimating insulin sensitivity is by using the glucose 
tolerance test (GTT). This is performed either by ingesting (oral glucose tolerance test, 
OGTT) or infusing (intravenous glucose tolerance test, IGTT) glucose, and then measuring 
the blood glucose and insulin response during the next two to three hours, while there is 
restoration of normal glycaemia. Mathematical models can then be applied to this data to infer 
insulin action. Perhaps the most popular model to date is the Homeostasis Assessment Model 
(HOMA-IR) which was first described in 1985 by Mathews et al. (Matthews et al., 1985). 
Chapter 2  Literature Review 
 21
HOMA-IR can be used to infer insulin sensitivity and β-cell function in the basal state, 
requiring only fasted insulin and glucose measurements. Thus, it is a simple and practical 
method of estimating insulin sensitivity and β-cell function in population studies and clinical 
practice. Nevertheless, some have shown that measurement of fasting insulin alone is just as 
effective at predicting insulin sensitivity as an OGTT (Stumvoll et al., 2000). In fact, studies 
in Māori have shown that HOMA-IR is no better than measuring fasting insulin alone, at 
predicting insulin sensitivity in this ethnic group (Bell, McAuley, Mann, Murphy, & 
Williams, 2004). Therefore, a fasting blood sample, and analysis of insulin, may be all that is 
required to screen for insulin resistance in Māori and others. 
 
2.3.4 Summary 
Type II diabetes is a disorder initially resulting from resistance to insulin, within insulin 
sensitive tissue. While the exact mechanisms which blunt the insulin response are not entirely 
understood, it appears that high concentrations of circulating and intracellular lipid, associated 
with obesity, interfere with the insulin signalling cascade, preventing GLUT-4 translocation 
and thus uptake of glucose from the blood.  
Chapter 2  Literature Review 
 22
2.4 Body Composition and Diabetes 
2.4.1 Obesity and its role in insulin resistance 
Mechanisms behind the relationship between body fatness and insulin sensitivity are still not 
fully understood. Nevertheless, obesity is associated with increased circulating free fatty acids 
(FFA) and intramyocellular triacylglycerols (IMTG or IMCL) concentrations (Goodpaster, 
Theriault, Watkins, & Kelley, 2000). This increase in FFA and IMTG is also associated with 
insulin resistance in skeletal muscle (Jacob et al., 1999; D. A. Pan et al., 1997). Because 
muscle is the most important insulin sensitive tissue (to glucose disposal), changes in muscle 
insulin sensitivity affect whole body insulin sensitivity.  
 
Still, resistance to insulin in the liver also plays a role in whole body insulin sensitivity. As 
discussed earlier, under normal circumstances, insulin inhibits gluconeogenesis within the 
liver (Saltiel & Kahn, 2001). On the other hand, increased FFA has been shown to reduce the 
inhibiting effect of insulin on hepatic glucose release, adding to the hyperglycaemic condition 
in insulin resistant individuals (Boden, 1997). Interestingly, elevated lipid availability 
increases triglyceride accumulation in the liver (Oakes et al., 1994), similar to how IMTG 
accumulates in skeletal muscle, and this accumulation appears to be associated with hepatic 
insulin resistance (Seppälä-Lindroos et al., 2002). What’s more, extraction of insulin, which is 
largely controlled by the liver, is reduced during hepatic insulin resistance, exacerbating the 
compensatory hyperinsulinema in response to whole body insulin resistance (Polonsky et al., 
1988).  
 
Because insulin resistance can occur within liver, muscle and adipose tissue, there is debate 
regarding the primary site of insulin resistance (Caldwell, Ikura, Iezzoni, & Liu, 2007). Some 
Chapter 2  Literature Review 
 23
propose that excess adiposity causes insulin resistance within adipocytes primarily, which 
then initiates whole-body insulin resistance (Bergman & Mittelman, 1998). In a normal state, 
insulin secretion suppresses lipolysis (break down of fat) within adipocytes, facilitating 
storage of triglycerides into adipose tissue. When this tissue becomes insulin resistant 
however, lipolysis continues which increases circulating FFA (Boden & Shulman, 2002). As 
discussed previously, increased concentration of circulating FFA is associated with insulin 
resistance. However, it appears that IMTG accumulation has an even stronger relationship 
with insulin sensitivity, suggesting that increased FFA drives accumulation of IMTG, rather 
than effecting glucose uptake in skeletal muscle directly (Shulman, 2000). Certainly, 
situations which encourage increased FFA, also increase concentrations of IMTG in healthy 
persons, though with a delay (Stannard et al., 2002). On the other hand, general weight loss is 
associated with a decrease in IMTG (Goodpaster, Theriault, Watkins, & Kelley, 2000) and 
improvements in insulin sensitivity. However, Klein et al. found that liposuction decreased 
weight and body fat percentage but did not improve insulin sensitivity (Klein et al., 2004), 
suggesting that a decrease in adipose tissue alone is not enough to improve insulin sensitivity, 
but must also be accompanied by a negative energy balance. Although IMTG appears to be a 
marker of insulin resistance in healthy people, this was not observed in Māori (Stannard, 
Holdaway, Sachinwalla, & Cunningham, 2007), suggesting that the relationship between 
IMTG content and insulin sensitivity may be effected by ethnicity. It is unlikely however, that 
IMTG accumulation directly impairs insulin action. Rather, other related lipid compounds that 
build up within muscle cells, as IMTG accumulates, alter the insulin signalling pathway (Ellis 
et al., 2000).  
 
One theory suggests that increased fat oxidation, from an oversupply of fat, competes with 
glucose as a substrate (Randle, Garland, Hales, & Newsholme, 1963; Shulman, 2000).  
Chapter 2  Literature Review 
 24
Randle et al. hypothesized that FFA induces insulin resistance by increasing the oxidation of 
fat relative to carbohydrate (Boden & Shulman, 2002; Randle, Garland, Hales, & Newsholme, 
1963). They showed that an increase in fatty acids increased intramitochondrial acetyl 
CoA/CoA and NADH/NAD+ ratio causing inactivation of pyruvate dehydrogenase. This leads 
to an increase of intracellular citrate, inhibiting phosphofructokinase (PFK), an important 
enzyme which controls the rate of glycolysis. This would then lead to accumulation of 
glucose-6-phosphate (G-6-P) and inhibition of hexokinase II activity, therefore increasing 
intracellular glucose and reducing glucose uptake. 
 
Others challenge this theory, suggesting that glucose transport is the primary factor effecting 
insulin sensitivity while enzyme activities, such as hexokinase II in glycogen synthesis, are 
secondary (Boden, Chen, Ruiz, White, & Rossetti, 1994; Dresner et al., 1999; Roden et al., 
1996). These studies found that in healthy persons, the reduction in muscle glycogen synthesis 
associated with elevated FFA, was preceded by a fall in intramuscular G-6-P, rather than an 
increase as would be expected with Randle’s theory, decreasing intramuscular glucose. These 
observations indicate that an increase in plasma fatty acid concentrations, inhibit glucose 
transport across the cell membrane, which in turn reduces the rate of phosphorylation (rather 
than mitochondrial ratios). This then results in reduced glycogen synthesis and glucose 
oxidation. Shulman suggests that fatty acid accumulation interferes with an earlier step in 
insulin stimulation of GLUT4 directly or the signalling pathways that signal GLUT4 
translocation (Shulman, 2000). A series of later studies by this group confirm these findings 
(Boden & Shulman, 2002). As discussed previously, rather than FFA directly inhibiting 
glucose uptake, it is thought that metabolites of IMTG directly interfere with one or more 
stages of GLUT4 translocation. For example, long-chain fatty acyl CoAs, the activated form 
of intracellular FFA, and 1,2-Diacylglycerol (DAG), another intermediate of lipid 
Chapter 2  Literature Review 
 25
metabolism,  have been shown to disrupt insulin signalling through activation of protein 
kinase C (PKC) which is associated with glucose disruption (Hegarty, Furler, Ye, Cooney, & 
Kraegen, 2003). Another lipid intermediate, ceramide, has been shown to directly inhibit the 
phosphorylation of insulin signalling molecules in muscle and adipose cells (Hajduch et al., 
2001; Summers, Garza, Zhou, & Birnbaum, 1998). Nevertheless, the precise mechanism/s 
responsible for disturbed glucose transport is not entirely clear. 
 
Interestingly, fasting studies have shown that extended periods of fasting result in elevated 
FFA concentrations, IMTG accumulation and impaired glucose uptake in healthy persons, 
similar to the insulin resistance seen in obese subjects (N A Johnson et al., 2006; Stannard et 
al., 2002). These findings suggest that, in an evolutionary context, IMTG accumulation and 
insulin resistance may have ensured an intra muscular fuel supply when food was scarce, 
whilst at the same time preventing muscle from taking up glucose to preserve it for the brain 
(Neel, 1962; Stannard et al., 2002).  
 
To further compound the pathological effects of obesity, it appears that FFAs may have a 
direct effect upon secretion of insulin and β-cell dysfunction as well, independent of insulin 
resistance and compensatory hyperinsulinemia (Boden & Shulman, 2002; Unger, 1995; Unger 
& Zhou, 2001). Crespin et al. showed that FFA infusion stimulated secretion of insulin in 
dogs (Crespin, Greenough Iii, & Steinberg, 1969). Furthermore, larger increases in FFA 
infusion resulted in correspondingly higher insulin levels. It also appears that elevated FFA is 
associated with increased insulin secretion in humans (Boden, Chen, Rosner, & Barton, 
1995). These observations suggest that in obese, insulin resistant individuals, elevated FFA 
may stimulate secretion of insulin directly, placing further strain on already overcompensating 
Chapter 2  Literature Review 
 26
β-cells. If this is the case, obesity has a multifaceted effect upon type-2 diabetes; through 
disruption of glucose disposal as well as dysfunction of insulin secretion. 
 
While there is a considerable amount of research regarding insulin resistance and its relation 
to body composition and/or aerobic fitness, these studies have been done almost entirely with 
European/Caucasian people (Cauza et al., 2005; Goodpaster, He, Watkins, & Kelley, 2001; 
Jacob et al., 1999; Kasa-Vubu, Lee, Rosenthal, Singer, & Halter, 2005; Thamer et al., 2003; 
Uusitupa et al., 2003). To date it appears that only one study exists on the relationship 
between IMTG and insulin resistance using a Māori cohort (Stannard, Holdaway, 
Sachinwalla, & Cunningham, 2007). Unfortunately this cross-sectional study only involved a 
relatively small cohort of convenience, and thus cannot be confidently generalised to the 
wider Māori population. Nevertheless, this study showed that body fatness and maximal 
aerobic capacity (VO2max) expressed per kg lean body mass, were both strong predictors of 
fasting insulin concentration in apparently healthy, young Māori men. 
 
In addition, most studies which have identified the relationship between adiposity and insulin 
sensitivity, have used less-direct methods to assess adiposity, such as BMI or skin-fold 
thickness (Bogardus, Lillioja, Mott, Hollenbeck, & Reaven, 1985). Surprisingly few have 
utilized hydrodensitometry; the ‘gold standard’ method of body composition assessment. On 
the other hand, body fat distribution, namely central adiposity, appears to be more closely 
associated to insulin resistance and other obesity related illnesses than total fat per se (Blair, 
Habicht, Sims, Sylwester, & Abraham, 1984). Following on from this, visceral fat appears to 
be more closely associated with insulin resistance than subcutaneous fat, fat in the extremities 
and lower body fat (Abate, Garg, Peshock, Stray-Gundersen, & Grundy, 1995; Björntorp, 
1991; Ross, Aru, Freeman, Hudson, & Janssen, 2002). Others suggest that subcutaneous 
Chapter 2  Literature Review 
 27
abdominal adiposity alone is more closely associated with insulin sensitivity (Abate, Garg, 
Peshock, Stray-Gundersen, & Grundy, 1995; Goodpaster, Thaete, Simoneau, & Kelley, 
1997). On the other hand, Fabbrini et al. (Fabbrini et al., 2009) reveal that intrahepatic fat is 
an independent indicator of ‘multi-organ’ insulin resistance and that visceral fat is only 
associated with insulin sensitivity because of the association between visceral fat and 
intrahepatic fat. Indeed, it is suggested that the anatomical location of visceral fat in the 
abdomen allows easy transport of FFA to the liver (Bergman et al., 2006). Regardless of 
whether visceral/abdominal fat effects insulin sensitivity directly or indirectly, measuring total 
body fat may actually be less effective than a simple measure of waist circumference. In light 
of this, one might suppose that the higher risk of obesity-related illness seen in Māori and 
Pacific Islanders, at a relatively low BMI, may be due to a genetic predisposition to harmful 
fat storage. However, the literature suggests that this is not the case (E. Rush et al., 2004; E. 
C. Rush, Freitas, & Plank, 2009). Nevertheless, waist circumference will be discussed in more 
detail in section 2.4.2.4. 
 
Given the importance of body composition in the prediction of insulin action, techniques used 
to assess body composition must be accurate. What’s more, body composition (P. 
Deurenberg, Yap, & van Staveren, 1998; Luke et al., 1997; Swinburn, Ley, Carmichael, & 
Palnk, 1999) as well as body fat distribution (Raji, Seely, Arky, & Simonson, 2001) differ 
between ethnic groups, so it is important that research in this area be ethnicity specific to 
better understand diabetes risk in those groups at greatest risk of metabolic disorders.  
 
2.4.2 Techniques used to measure body composition 
During the last 2 decades, literature regarding anthropometric techniques has become 
abundant and modern, more complex techniques have been compared to less intricate 
Chapter 2  Literature Review 
 28
techniques of the past. Body composition techniques rely on reference data regarding the 
composition of the human body. This reference data is based primarily on the chemical 
analysis of just a few human cadavers. Because only a limited amount of cadavers have 
actually been analyzed in the past century (Clarys, Martin, & Drinkwater, 1984), assumptions 
that are used in body composition models are based on reference data that is not necessarily 
indicative of the population being measured (Heyward & Wagner, 2004). 
 
2.4.2.1 BMI 
The body mass index (BMI) was first developed to meet the needs for screening those at high 
risk of obesity related illnesses such as hypertension, heart disease and diabetes, and is based 
on a simple measurement of an individual’s weight in relation to their height (BMI = 
weight/height2). 
 
In the early 90’s ‘The Expert Committee on Clinical Guidelines for Overweight in Adolescent 
Preventive Services’ was established in the United States to develop specific criteria to 
identify adolescents most at risk of obesity and its co-morbidities (Himes & Dietz, 1994). 
This committee proposed that a BMI above the 85th percentile, or 25 kg/m2, was a 
representation of overweight, while those above the 95th percentile, or 30kg/m2, were 
considered obese. Three years later the International Obesity Task Force (IOTF) held a 
workshop in Dublin to explore the strengths and limitations of existing approaches to measure 
childhood obesity, including the previously mentioned BMI cut-off points (Dietz & Bellizzi, 
1999). Although a number of limitations were identified, it was concluded from this workshop 
that BMI was a reasonable way to assess obesity and overweight in adolescents. Similar BMI 
standards of 25 kg/m2 and 30kg/m2, representing overweight and obesity respectively, have 
been accepted as appropriate cut-off points for health screening in adults (D. Gallagher et al., 
Chapter 2  Literature Review 
 29
2000). The reference data for these cut-off points is derived from studies showing that BMI is 
associated with risk of diabetes, cardiovascular disease (CVD) and mortality (Calle, Thun, 
Petrelli, Rodriguez, & Heath, 1999; I. M. Lee, Manson, Hennekens, & Paffenbarger Jr, 1993; 
Lohman, 1981b; WHO, 1995). BMI has also been shown to correlate well with body fatness 
(Paul Deurenberg, Weststrate, A., & Seidell, 1991). There are obvious problems that arise 
from adopting a single set of rules for BMI cut-off points however, as reference data is 
derived from populations that are not representative of all groups. Therefore, what may be 
applicable in a predominantly Caucasian group may not be an appropriate measure of health 
risk in populations without Caucasian heritage.  
 
Because BMI is inexpensive, easy to calculate [weight (kg)/height2 (m)] and non-invasive, it 
has become a practical and popular tool in the assessment of obesity-related disease risk in 
populations. However, because it is excess adiposity and not body weight per se that causes 
obesity related illness (D. Gallagher et al., 2000; Janssen, Katzmarzyk, & Ross, 2004), relying 
solely on BMI as a measurement of obesity, has its problems (Prentice & Jebb, 2001). 
Research shows that BMI and its relation to adiposity varies among ethnic groups (P. 
Deurenberg, Yap, & van Staveren, 1998; Prentice & Jebb, 2001; Seidell, 2000; Swinburn, 
Ley, Carmichael, & Palnk, 1999), between age groups (D. Gallagher et al., 2000; Mei et al., 
2002) and differs with gender (D. Gallagher et al., 1996). These variations lead to the issue of 
appropriate BMI values that can accurately express overweight and obesity, most commonly 
known in the literature as BMI cut-off points. For example, when using the currently accepted 
BMI cut-off points, 7 time Mr. Olympia Arnold Schwarzenegger, who at the apex of his 
bodybuilding career was branded by the Guinness book of records as “the most perfectly 
developed man in the history of the world”, would at the same time be considered well in the 
obese range with a BMI of 33.4. Similarly, world renowned rugby winger Jonah Lomu, who 
Chapter 2  Literature Review 
 30
could run 100 metres in 11 seconds in his prime, would have carried the label of obese, with a 
BMI of 31. On the other hand, a study by O’Dea et al. (O’Dea, White, & Sinclair, 1988) 
revealed that Australian Aborigine adults living a traditionally oriented lifestyle all displayed 
a BMI categorised as ‘underweight’ by current BMI standards (13.4-19.3 kg/m2), but showed 
no biochemical evidence of malnutrition or poor health. Although these examples do not 
represent the body composition of an average NZ man or women, their examples outline a 
need for specificity in BMI standards. 
 
Much of the recent literature regarding the body mass index deals with the association of BMI 
to adiposity and how it differs between populations. Because public health strategies rely on 
accurate representation of health trends within populations, and considering the extensive 
research that relies on BMI use, it is important that the effectiveness of BMI as a tool in 
predicting obesity related illness and appropriate cut-off points is re-evaluated. This is 
especially important for Māori, who are at greater than average risk from obesity-related 
morbidity and mortality. 
 
BMI & Age 
 In his studies of ageing and its effect on body composition, S. H. Cohn showed that a typical 
healthy male will have an increase of body fat from 15% to 29% from the age of 25 to 70, 
indicating an increase in the ratio of fat to lean body mass as ageing occurs (Cohn, 
1987(citation)). Accordingly, a review by Serdula et al. showed that obese children were at 
least twice as likely to become obese adults when compared with non-obese children, and that 
the degree of obesity in childhood increased the risk of adult obesity (Serdula et al., 1993). 
BMI has also been shown to increase with age, but BMI as a representation of body fat, 
underestimates the magnitude of the increase of adipose tissue associated with ageing, and 
Chapter 2  Literature Review 
 31
therefore fails to detect the age-associated decrease of lean mass (Prentice & Jebb, 2001). 
What’s more, the validity of employing BMI to determine obesity in children has been 
questioned (Paul Deurenberg, Weststrate, A., & Seidell, 1991). These findings highlight the 
need for age-specific BMI cut-off points. In fact, a publication by Deurenberg et al. outlines 
the effectiveness of age-specific formulas, showing that with such formulas BMI becomes an 
accurate predictor of body fat stores with a prediction error similar to skin-fold thickness and 
bioelectrical impedance measurements (Paul Deurenberg, Weststrate, A., & Seidell, 1991). 
 
BMI & Ethnicity 
Most ethnic comparative studies of body composition show differences in the relationships 
between BMI, percentage body fat, and health risks between ethnic groups (P. Deurenberg, 
Yap, & van Staveren, 1998; Luke et al., 1997; Prentice & Jebb, 2001), while some have 
shown no difference (D. Gallagher et al., 1996). The international cut-off points set by the 
WHO in 1993 (WHO, 1995) come from definitions of overweight and obesity derived mainly 
from data collected by Lohman et al. (Lohman, 1981a). More recently researchers have found 
that some ethnic groups such as Asians and Indians have a higher percentage of body fat than 
Caucasians when controlled for BMI, age and sex (WHO, 2004). Polynesians and African 
Americans on the other hand, are characterised by a lower body fat percentage than 
Caucasians (P. Deurenberg, Yap, & van Staveren, 1998; Swinburn, Ley, Carmichael, & 
Palnk, 1999) because of higher proportions of fat-free mass (Swinburn, Ley, Carmichael, & 
Palnk, 1999). Prentice and Jebb calculate that in order to have the same body fat proportion as 
a Caucasian, a Polynesian would need to have a BMI 5kgm-2 higher than a Caucasian of the 
same age and gender (Prentice & Jebb, 2001). In Asians and Indians however, the risk of 
obesity related illnesses such as diabetes seem to increase rapidly at a BMI that falls in the 
acceptable range for Europeans (Kosaka, Kuzuya, Yoshinaga, & Hagura, 1996; Prentice & 
Chapter 2  Literature Review 
 32
Jebb, 2001). Because it is excess adipose tissue and not excess weight that causes obesity 
related health risk (P. Deurenberg, Yap, & van Staveren, 1998), BMI scores that are 
unadjusted for ethnic differences may lead to an inaccurate diagnosis of risk. The implication 
is that those with relatively lower BMI may be at risk of obesity related illness at a BMI lower 
than the cut-off, and therefore are ignored in the screening process as they fall within a 
healthy range. Those with a high BMI due to greater fat-free mass, such as Polynesians and 
African Americans, would be wrongly categorized as obese. BMI cut-off points representing 
overweight and obesity should therefore be population specific (P. Deurenberg, Yap, & van 
Staveren, 1998). 
 
To complicate this issue, differences in the BMI/body fat relationship have been observed 
between populations with the same or similar ethnic background. Luke et al. (Luke et al., 
1997) studied the relationship between BMI and body fat in three different black populations 
from Nigeria, Jamaica and the US, who all shared a common West African ancestry. They 
found that at any given BMI, the percentage of body fat differed significantly between the 
three populations. For example, males at a BMI of 25 had a body fat percentage of 25.8% for 
the US sample, while Jamaicans had 22.2% and Nigerian men just 16.4%. Deurenberg et al. 
also observed differences between the Americans and Europeans within a ‘white’ cohort of 
their study (P. Deurenberg, Yap, & van Staveren, 1998). These findings indicate that 
environmental factors along with ethnic make-up may influence the relationship between BMI 
and body fat percentage. More importantly, the difference in obesity-related disease risk 
between these genetically close groups, that are geographically disparate, requires 
investigation to identify important environmental factors. Luke et al. suggests that differences 
in energy intake and expenditure (the availability of food and the amount of physical activity) 
Chapter 2  Literature Review 
 33
between these populations are the main environmental factors influencing these differences 
(Luke et al., 1997). 
 
BMI & body composition 
BMI is one of a number of ways to assess obesity, and studies of its effectiveness compared to 
other methods of measurement are well reported in current literature. BMI alone does not 
measure body composition or the amount of adipose tissue (Prentice & Jebb, 2001). Body 
composition is also the reason for the obvious differences in the BMI/body fat relationship 
between populations. In those that are highly muscular, whether from resistance training or a 
natural disposition, BMI gives an overestimation of body fat (Prentice & Jebb, 2001). Some 
populations such as Polynesians also appear to have a higher bone mineral density (Swinburn, 
Ley, Carmichael, & Palnk, 1999), while the body fat of populations with naturally small 
frames is underestimated by BMI because of a lower level of muscle mass (P. Deurenberg, 
Yap, & van Staveren, 1998; Swinburn, Ley, Carmichael, & Palnk, 1999). 
 
2.4.2.2 Underwater weighing (Hydrodensitometry) 
Underwater weighing, also known as hydrostatic weighing or hydrodensitometry, has been 
accepted as an effective method of measuring body composition (Heyward & Wagner, 2004; 
Nord & Payne, 1995; van der Ploeg, Gunn, Withers, Modra, & Crockett, 2000). This method 
uses Archimedes principle to calculate body density from values of an individual’s mass in air 
and their mass underwater, taking into account the water temperature (van der Ploeg, Gunn, 
Withers, Modra, & Crockett, 2000). Good practice also measures the volume of gas in the 
respiratory system while an individual is fully submerged in water (known as residual lung 
volume), because this alters buoyancy. Values from assumed density of fat and fat-free mass 
are then used to estimate body fat percentage according to algorithims based on cadaver 
Chapter 2  Literature Review 
 34
studies of body density (Brozek, Grande, Anderson, & Keys, 1963; Norton, 1996; Siri, 1961). 
Because underwater weighing is the benchmark that other methods of body composition 
assessment have been compared to, it is difficult to verify the validity of underwater 
weighing. What’s more, more modern techniques of body composition appear to exceed 
hydrodensitometry in its precision of body composition assessment (Heymsfield, Wang, 
Heshka, Kehayias, & Pierson, 1989) (See section 2.4.2.6).  
 
The main problem with this method is the need for accurate measures of residual lung volume 
which can be difficult to obtain and uncomfortable for participants (Wilmore, 1969; Wilmore, 
Vodak, Parr, Girandola, & Billing, 1980). Calculation of body fat percentage from 
individuals, whose bone density or total body hydration deviates from the standard by 
assumption of the four compartment model, will also result in errors (Nord & Payne, 1995) 
because of the relatively small cadaver-based data sets from which the density equations are 
drawn. 
 
Underwater weighing can be very difficult to perform, for both the participant being measured 
and for those administering the test. It requires a pool or tub big enough to fully submerge a 
person, comfortably warm water, and easy access in and out of the pool. This is difficult for 
old, large and/or physically incapable persons. Furthermore, it can be quite time consuming, 
because a number of measures are needed, as well as calibration, so it is not practical for use 
in population studies or in clinical assessment. 
 
2.4.2.3 Bioelectrical Impedance 
Bioelectrical impedance (BIA) machines have become a popular tool for measuring body 
composition. BIA is quick, non-invasive and recent models are simple to use. BIA does not 
Chapter 2  Literature Review 
 35
measure body fat directly, but measures total body water and segmental water content through 
the impedance of tissues when a small electrical current is passed through the body. Based on 
the impedance of this current and certain assumptions regarding water content in FFM (~73% 
water), body fat and fat-free mass can then be estimated using knowledge from prior 
experiments associating impedance with body composition (Lukaski, 1987; National 
Institutes of Health Technology, 1996). Fat-free mass contains a greater proportion of 
electrolytes and therefore conducts electricity better than fat which has a relatively low water 
content (Khaled et al., 1988). This implies that the impedance of the electrical current as it 
passes through the body is related to the amount of fat it has to pass through. This method 
relies on calculations involving the length and cross sectional area of the body measured 
(Ursula G. Kyle et al., 2004). There are obvious problems with this however as the length and 
cross sectional area of the human body is much more complex than a basic cylinder. 
 
Since BIA produces data quickly and is relatively inexpensive, it is a practical tool for 
obtaining body composition data in clinical screening and population research. The reliability 
of BIA is a major concern however. Individuals measured with BIA have recorded variations 
in body fat percentage of up to 10% because of differences in machines and the 
methodologies used by manufacturers (National Institutes of Health Technology, 1996). It 
also seems that like BMI, BIA is limited by its need for age, gender and ethnicity-specific 
equations due to the differences between populations (National Institutes of Health 
Technology, 1996; Swinburn, Ley, Carmichael, & Palnk, 1999). It is important that the most 
appropriate prediction equation is used for each individual. 
 
The body of literature regarding bioelectrical impedance is massive, and most of the research 
is focused on developing the BIA methods and equations in order to produce more valid 
Chapter 2  Literature Review 
 36
measurements of body composition, when compared to more direct measurements such as 
hydrodensitometry and DEXA. While the skill of the technician taking the BIA measurement 
is not a major source of error, there are many other possible sources of error when using the 
BIA method. In the late 80’s, Caton et al. found that body composition measurements 
obtained from BIA were significantly affected by changes in skin temperature when ambient 
air temperature was changed (Caton, MolÉ, Adams, & Heustis, 1988). In warmer conditions, 
fat mass was significantly lower than in the cooler conditions. Another study 10 years later 
also repeated this observation and found that skin hydration (similar to sweating) also affected 
impedance (Cornish, Thomas, & Ward, 1998). These findings highlight the importance of 
standardized environmental conditions when performing BIA measures. 
 
Because BIA is a measure of electrical impedance of body tissues, variation in bone and 
muscular densities between individuals can affect the validity of BIA. In fact, BIA was found 
to overestimate fatness in lean males and underestimate it in overweight subjects, when 
compared with hydrostatic weighing (Heath, Adams, Daines, & Hunt, 1998; Segal, Gutin, 
Presta, Wang, & Van Itallie, 1985). What’s more, not only do Māori and Polynesians have a 
greater muscle mass when compared with Europeans of the same BMI, but their bone density 
is also greater (Reid, Mackie, & Ibbertson, 1986; Swinburn, Ley, Carmichael, & Palnk, 1999). 
Durenberg et al. also showed that differences in body build, including arm and leg length in 
proportion to the trunk, was a major cause of the errors in BIA results between ethnic groups 
(P. Deurenberg, Deurenberg-Yap, & Schouten, 2002). They also showed however, that when 
the differences in body build were accounted for in the BIA equations relating impedance to 
body fatness, the error disappeared. Though there are many studies on the different 
confounders that effect BIA measurements, only a few studies exist regarding the affect of 
Chapter 2  Literature Review 
 37
ethnicity on the validity of BIA methods. Therefore, appropriate equations are needed for 
Māori, as well as other ethnic groups. 
 
Theoretically, over hydration should give relatively higher electrical resistance measurements, 
while dehydration will give relatively lower resistance levels (Khaled et al., 1988). A review 
by Kyle et al. (U. G. Kyle et al., 2004) revealed that BIA results are affected most by whether 
subjects were in the fasted or fed state. For example Gallagher et al. (M. R. Gallagher, 
Walker, & O'Dea, 1998) showed that measuring BIA 3-5hrs post-consumption of food 
significantly underestimated body fat when compared to fasting measurements. Kushner et al. 
(Kushner, Gudivaka, & Schoeller, 1996) showed similar effects of eating, but also reported 
that the effect of eating was not apparent until one hour after food consumption. 
 
Thus, for the most accurate and reliable results, development of universally standardized 
protocols are needed for BIA (U. G. Kyle et al., 2004). This means that BIA measurements 
should be taken at the same time of the day, in a fasted state, with controls to reduce the 
effects of other confounders such as hydration status and exercise. 
 
2.4.2.4 Waist circumference 
Measurement of abdominal adiposity is a good, independent predictor of chronic diseases 
such as CVD and diabetes (Bell, McAuley, Mann, Murphy, & Williams, 2004; Lean, Han, & 
Seidell, 1998; Zhu et al., 2002). In fact, measurement of waist circumference (WC), is a better 
predictor of obesity-related illness than BMI alone (Janssen, Katzmarzyk, & Ross, 2004; Zhu 
et al., 2002). In a study of over 182,000 people from 63 different countries, Balkau et al. 
showed that WC was a stronger predictor of CVD and diabetes than BMI (Balkau et al., 
2007). What’s more, body fat distribution i.e. a greater proportion of fat in the abdominal 
Chapter 2  Literature Review 
 38
area, is a more important indicator of cardio-vascular (Blair, Habicht, Sims, Sylwester, & 
Abraham, 1984) and metabolic health (Ross, Aru, Freeman, Hudson, & Janssen, 2002), than 
total body fat per se. Because of the close association between central adiposity and insulin 
resistance, WC could become a practical method of assessing risk of insulin resistance and 
diabetes. .  
 
Body fat distribution and the relationship between abdominal adiposity and obesity-related 
co-morbidities differ with ethnicity. One study by Raji et al. showed that Asian Indians had 
greater abdominal fat when compared to Caucasians of a similar BMI (Raji, Seely, Arky, & 
Simonson, 2001). Accordingly, at a comparable BMI and age, the Asian Indian group also 
demonstrated an increase of markers associated with insulin resistance and CVD. Similarly, 
Okosun et al. reported that the value of WC as a predictor of diabetes and CVD, differed 
between ethnicities in a cohort of American Caucasian, African American and Mexican 
American people (Okosun, Liao, Rotimi, Choi, & Cooper, 2000). Because of these 
differences, ethnic specific criteria is required in the use of WC as a measure of adiposity 
(Misra, Wasir, & Vikram, 2005; Okosun, Tedders, Choi, & Dever, 2000). Nevertheless, 
regardless of the differences in the relationship between WC and health risk, WC has still 
proven to be a reliable predictor of obesity-related co-morbidities in many high risk ethnic 
groups, including Mexican Americans (M. Wei, Gaskill, Haffner, & Stern, 1997), African 
Americans (Okosun, Cooper, Rotimi, Osotimehin, & Forrester, 1998) and Māori (Bell, 
McAuley, Mann, Murphy, & Williams, 2004). 
 
2.4.2.5 Skinfold thickness 
Another way of measuring excess adiposity is through skin-fold measurements which 
measure thickness of subcutaneous adipose tissue from various sites on the body. The sum of 
Chapter 2  Literature Review 
 39
the skin-folds taken from the sites mentioned above may then be used on their own or put into 
an equation to estimate fat stores, based on reference data (usually obtained from 
hydrodensitometry). There are over a hundred such population-specific equations in use, but 
generalized equations have been developed also (Heyward & Wagner, 2004). This type of 
measurement is inexpensive and easily measured, though skill is required for reliability 
(Heyward & Wagner, 2004). Measuring skin-fold thickness is invasive however, and 
concerns about the reliability of this method arise because skin-fold thickness measurements 
are often only taken from a few body sites (Lohman, 1981b(Citation); Mei et al., 2002). 
Intertechnician reliability is also a cause for error as improper location and measurement of 
skin-fold sites between technicians, as well as choice of an appropriate equation, can cause 
significant differences in %BF interpretation (Lohman, Pollock, Slaughter, Brandon, & 
Boileau, 1984) The use of reporting body fatness from skin-fold measurements also has 
questionable validity since it completely ignores variations in the subcutaneous/visceral fat 
relationships which exist (Norton, 1996).  
 
2.4.2.6 Dual energy x-ray absorptiometry 
A more sophisticated way of measuring body composition is through dual energy x-ray 
absorptiometry (DEXA or DXA). This method distinguishes, then quantifies fat mass, lean 
mass and total-body bone mineral, through the use of two x-rays of differing energy levels 
(Mazess, Barden, Bisek, & Hanson, 1990; Svendsen, Haarbo, Hassager, & Christiansen, 
1993). It does this by measuring the different attenuation by tissues (muscle, fat, bone etc.) of 
emitted photons at two energy levels; these measurements are then used to calculate 
proportions of different body tissues (Heymsfield, Wang, Heshka, Kehayias, & Pierson, 
1989). DXA has proven to be a very valuable method of measuring body composition as it is 
able to differentiate between fat and fat-free mass, in regions as well as total body (Mei et al., 
Chapter 2  Literature Review 
 40
2002; Svendsen, Haarbo, Hassager, & Christiansen, 1993). In fact, some results would 
suggest that DXA is more precise than hydrodensitometry (Heymsfield, Wang, Heshka, 
Kehayias, & Pierson, 1989), highlighting its use as a method of measuring small changes in 
body composition. However, the use of DXA as a measure of body composition in clinical 
screening at a population level is not practical, as it is time consuming and expensive, limiting 
its use primarily to research settings (Prentice & Jebb, 2001). 
 
2.4.2.7 Māori and body composition 
BMI has been adopted by clinicians as a practical way of determining body composition in 
New Zealand. As previously mentioned however, the problem with relying solely on BMI as 
a measure of obesity in this priority group is that Māori and Pacific people have a different 
BMI/body fat ratio to NZEO. With this in mind, Swinburn et al. has recommended a BMI of 
26-32 as overweight and 32 as obese in Māori and Polynesians, for clinical and 
epidemiological purposes (Swinburn, Ley, Carmichael, & Palnk, 1999). At the standard BMI 
cut-off for overweight (25kg/m2) Polynesians have more fat-free mass and less fat than 
Caucasians (Prentice & Jebb, 2001; Swinburn, Ley, Carmichael, & Palnk, 1999), suggesting 
the need for Māori/Polynesian-specific equations. Other studies suggest that BMI is not an 
adequate measure of fatness in a mixed European, Māori and Pacific population and other 
methods of assessment such as BIA (E. C. Rush, Puniani, Valencia, Davies, & Plank, 2003) 
or waist circumference (Bell, McAuley, Mann, Murphy, & Williams, 2004) would be more 
appropriate. On the other hand, it has been argued that  cut-off points should reflect the risk of 
co-morbidities associated with a given BMI, rather than an individual’s lean body mass (K. 
McAuley, Williams, Mann, Goulding, & Murphy, 2002). In their study on a group of Māori 
and non-Māori women, McAuley et al. showed that Māori women were more insulin resistant 
than Europeans even when BMI levels were similar.  
Chapter 2  Literature Review 
 41
2.4.3 Summary 
Increased body fat is identified as a major cause of insulin resistance. Fat is an important 
substrate during exercise, although in sedentary individuals excess fat from energy rich meals 
is not used and fat is stored in adipose tissue and muscle cells. A high concentration of IMTG 
is associated with diabetes, but can also be seen in healthy endurance trained individuals. This 
suggests that the turnover of intramuscular lipids, rather than the concentration per se, affects 
the insulin signalling pathway. Because of body fat’s relationship to insulin resistance, 
obtaining accurate measures of body composition is important in predicting many health risks. 
Of the methods of measuring or estimating body fatness discussed above, only DEXA and 
underwater weighing have been shown to be reliable and accurate measures.  However, these 
are techniques which require expensive equipment and trained technicians to obtain the 
measurements. Simple and quick means of accurately measuring body composition must be 
sought if large scale studies are to become more valid. 
 
Chapter 2  Literature Review 
 42
2.5 Exercise and Diabetes 
2.5.1 Fuel selection during endurance and high intensity exercise 
The increasing energy requirement during exercise, due to increased skeletal muscle 
metabolism, is predominantly fuelled by oxidation of lipids and carbohydrate (CHO). The 
total fuel requirement is determined by the absolute work rate and duration of exercise, while 
the proportion of fuel used, i.e. the ratio of fat to CHO, is determined by the relative intensity 
of the exercise for a particular duration (Brooks & Mercier, 1994). Generally speaking, during 
low intensity exercise, particularly as duration increases, oxidation of fat provides most of the 
energy for contraction, although CHO still contributes in part. At the onset of exercise, and 
particularly at higher intensities, glycolytic flux predominates and fatty acid oxidation plays a 
more minor role (Hawley, 2001). 
 
Circulating glucose is an important substrate for contracting muscle. It has been repeatedly 
shown that if circulating glucose concentrations can be supported or even increased in the 
latter stages of prolonged exercise, then muscle performance similarly is maintained (Coggan 
& Coyle, 1987; Coyle et al., 1983). The rate of blood glucose utilization by contracting 
muscle, is controlled by various factors including glycogen status (Fell, Terblanche, Ivy, 
Young, & Holloszy, 1982), lipid availability (Ferrannini, Barrett, Bevilacqua, & DeFronzo, 
1983), muscle temperature (Febbraio, Snow, Stathis, Hargreaves, & Carey, 1994), exercise 
duration (Ahlborg, Felig, Hagenfeldt, Hendler, & Wahren, 1974), and exercise intensity 
(Hawley, 2001). Glucose uptake into muscle fibre increases as a function of contraction 
(Ploug, Galbo, & Richter, 1984), whilst an exercise-induced increase in blood flow increases 
substrate availability (Goodyear & Kahn, 1998). In recovery (at rest), glucose uptake 
increases to replenish muscle glycogen stores, but some continuing oxidation takes place. At 
Chapter 2  Literature Review 
 43
rest, following recovery and in the fed or postprandial state, it is not usually availability of 
glucose during exercise that limits oxidation of glucose, but the process of glucose transport 
into the cell. As touched upon earlier, GLUT4, the main transport protein responsible for 
glucose transport, is not stimulated to the cell surface (to enable glucose uptake) by insulin 
secretion alone but GLUT4 translocation is also independently stimulated by contraction 
(Ploug, Galbo, & Richter, 1984). For these reasons, exercise and adequate physical activity 
are important in maintaining glucose transport and preventing type-2 diabetes (J. Eriksson, 
Taimela, & Koivisto, 1997; Ivy, 1997). 
 
2.5.2 Effects of an acute exercise bout on glucose transport 
A single bout of exercise improves insulin-mediated glucose disposal in insulin resistant, 
diabetic and even insulin sensitive subjects (J. Eriksson, Taimela, & Koivisto, 1997; E. 
Henriksen, 2002). The insulin sensitive response to acute exercise vanishes within a few days 
however, emphasizing the need for chronic exercise (J. Eriksson, Taimela, & Koivisto, 1997; 
Richter, Garetto, Goodman, & Ruderman, 1982). 
 
A single bout of exercise also reduces body fat, albeit in small amounts, through increased 
oxidation of intramuscular lipid, and mobilization of fatty acids from adipose tissue. At 
exercise intensities below 75% of maximal workload, utilization of fat as a substrate is at its 
highest (See Figure 2). Past this point, the relative contribution of glycogen and glucose 
increases, while fat oxidation decreases (van Loon, Greenhaff, Constantin-Teodosiu, Saris, & 
Wagenmakers, 2001). This is likely the reason why sub-maximal exercise has become the 
recommended form of exercise for those with insulin resistance or at risk of the disorder. 
 
Chapter 2  Literature Review 
 44
Figure 2 – Relative contribution of substrate utilized at different exercise intensities (L.J.C 
van Loon, 2004)  
 
However, because GLUT4 translocation responds to muscle contraction independent of 
insulin sensitivity, other types of exercise, such as resistance training, have proven effective in 
improving glucose tolerance through greater expression and increased translocation of 
GLUT4 transport proteins (Holten et al., 2004). Host et al. (Host, Hansen, Nolte, Chen, & 
Holloszy, 1998) found that the increase of GLUT4 seen after training in animals, reversed 
after just 2 days of detraining. Thus, in order for exercise to be effective in maintaining 
glucose tolerance, exercise must be performed regularly. Interestingly, the period just after 
exercise is also associated with increased insulin sensitivity within muscle cells (Richter, 
Chapter 2  Literature Review 
 45
Mikines, Galbo, & Kiens, 1989), which re-enforces the potential benefits of exercise in 
improving and maintaining insulin sensitivity. 
2.5.3 Effects of Chronic exercise on glucose transport 
While the relationship between physical activity and insulin resistance is not entirely 
understood, many studies have showed that progression from a glucose intolerant state to 
diabetes, is prevented by physical exercise (K. F. Eriksson & Lindgärde, 1991; Lindstrom et 
al., 2003; X. R. Pan et al., 1997; Tuomilehto et al., 2001). In some cases, exercise and a 
healthy diet have even led to remission of diabetes (K. F. Eriksson & Lindgärde, 1991). 
Furthermore, type-2 diabetics who are physically fit live longer than diabetics who are unfit 
(Ming Wei, Gibbons, Kampert, Nichaman, & Blair, 2000). Physical activity not only reduces 
adiposity, but improves insulin sensitivity independent of weight loss (Duncan et al., 2003; 
Katzmarzyk, Church, Janssen, Ross, & Blair, 2005; S. Lee et al., 2005; S. J. Lee et al., 2005), 
suggesting that diabetes develops not just because of obesity, but from physical inactivity. 
This leads to the idea that an overweight person who is physically fit is protected from insulin 
resistance, while an unfit lean person may be at higher risk of insulin resistance and diabetes. 
 
Aerobic fitness is largely a function of the oxidative capacity of myocytes; the ability of 
muscle to oxidize substrate. Although there are many mechanisms underlying the increase in 
physical fitness that results from training, classic studies by Holloszy (J. O. Holloszy, 1967), 
as well as Gollnick and King (Cited in J O  Holloszy & Coyle, 1984), revealed that exercise 
training increased mitochondrial content, density and function within muscle fibres of rats. 
Later, studies by Hoppeler et al. and Morgan et al. (Cited in J O  Holloszy & Coyle, 1984) 
confirmed these findings in humans. Conversely, studies have shown that mitochondrial 
density and content are reduced in obese (Toledo, Watkins, & Kelley, 2006) and type-2 
diabetic patients (D. E. Kelley, He, Menshikova, & Ritov, 2002). However, exercise increases 
Chapter 2  Literature Review 
 46
mitochondrial function and content in these groups as well (Toledo et al., 2007; Toledo, 
Watkins, & Kelley, 2006). These findings highlight the role of mitochondrial content and 
density as determinants of aerobic fitness, and their importance in regulating metabolic health. 
What’s more, oxidative capacity also mediates the rate of lipid oxidation which is associated 
with an individual’s ability to maintain or lose weight (David E. Kelley, Goodpaster, Wing, & 
Simoneau, 1999).  
 
Interestingly, ageing is also associated with a diminished oxidative capacity (Conley, Jubrias, 
& Esselman, 2000; Papa, 1996) and insulin sensitivity (Fink, Kolterman, Griffin, & Olefsky, 
1983). On the other hand, Rimbert et al. showed that aerobic capacity is not diminished with 
age, but is associated with the physical inactivity that accompanies aging (Rimbert et al., 
2004). This observation implies that a reduction of physical activity in later years, rather than 
age itself, as the name ‘adult-onset diabetes’ suggests, explains in part, the association 
between age and insulin resistance. 
 
2.5.4 Aerobic Capacity, Oxidative Capacity and Insulin Resistance 
Chronic endurance exercise also increases the oxidative capacity of skeletal muscle and 
therefore its ability to use fat as a substrate. Aerobic fitness is also closely associated with 
insulin sensitivity. For example, oxidative enzyme activity is closely related to insulin 
sensitivity (Rimbert et al., 2004; J.-A. Simoneau & Kelley, 1997). Some even suggest that 
oxidative capacity is a better predictor of insulin resistance than adiposity (C. R. Bruce et al., 
2003). Although measuring oxidative enzyme activity is a relatively accurate representation 
of a muscle’s aerobic fitness, the problem with this method is that invasive muscle biopsies 
are required. A test of maximal oxygen consumption (VO2max) on the other hand, is less 
invasive and the only discomfort felt is the discomfort associated with intense exercise. 
Chapter 2  Literature Review 
 47
However, maximal aerobic capacity is probably limited by O2 delivery to the working muscle, 
rather than O2 utilization (González-Alonso & Calbet, 2003). Nevertheless, a high VO2max 
requires a good oxidative capacity (Bassett Jr & Howley, 2000). 
 
Literature observing the relationship between insulin sensitivity and VO2max is limited, 
though one study found that VO2max was a better predictor of insulin resistance than BMI or 
even adiposity in a group of post-pubertal adolescent females (Kasa-Vubu, Lee, Rosenthal, 
Singer, & Halter, 2005). Likewise, Stannard et al. revealed a similar relationship between 
VO2max and insulin sensitivity in a cohort of healthy Māori men (Stannard, Holdaway, 
Sachinwalla, & Cunningham, 2007). 
 
There are problems using maximal oxygen consumption as a measure of aerobic fitness 
however. When used in research, VO2max is most often expressed relative to total body 
weight (ml.kg total body mass -1.min-1), which includes fat mass. Because physical fitness is a 
risk factor independent of fatness, and because skeletal muscle (and not adipose tissue) is 
responsible for the increase in oxygen consumption during exercise, it seems more sensible to 
express fitness in terms of lean body mass (VO2maxLBM) according to the methodology of 
two previous studies i.e. (ml.kg lean body mass -1.min-1) (Goodpaster, Wolfe, & Kelley, 2002; 
Stannard, Holdaway, Sachinwalla, & Cunningham, 2007). 
 
2.5.5 Endurance training & IMTG accumulation: A paradox 
As discussed earlier, resting IMTG levels are elevated in sedentary obese individuals and 
those with insulin resistance. A paradox exists regarding this association however as 
endurance trained athletes also have elevated levels of IMTG even though their insulin 
sensitivity is high (Hoppeler et al., 1985; Morgan, Short, & Cobb, 1969). Schrauwen-
Chapter 2  Literature Review 
 48
Hinderling et al. (Schrauwen-Hinderling et al., 2003) observed that IMTG content increased 
in untrained healthy males after just 2 weeks of endurance training, but that this increase was 
not associated with a reduction of insulin sensitivity. The authors suggest that 1) increased 
IMTG accumulation is amongst the earliest responses to training, 2) The presence of IMTG 
alone does not have detrimental effects on insulin sensitivity, and 3) that IMTG contributes to 
the fuel used during prolonged exercise. This is further supported by Goodpaster et al. 
(Goodpaster, He, Watkins, & Kelley, 2001) who showed that in sedentary individuals, IMTG 
content was inversely associated with insulin sensitivity, while endurance-trained subjects 
were insulin sensitive despite having higher IMTG content than the sedentary group. Thamer 
et al. (Thamer et al., 2003) further suggest that the correlation between IMTG and insulin 
resistance is modified by the extent of aerobic fitness, and that interpretation of IMTG 
measurements and how they relate to insulin resistance must include measurements of aerobic 
fitness. 
 
The difference between the obese and endurance trained individual is that the endurance 
trained person has muscle with a higher fat oxidation capacity and is frequently involved in 
exercise, both of which are conditions enabling turnover of IMTG. Stannard and Johnson 
suggest that in a context of exercise, the ability to store large amounts of IMTG aids in 
increasing the availability of lipid as a  substrate during prolonged bouts of exercise, or at the 
onset of exercise when alternative intramuscular substrate, glycogen, has not been replenished 
(Stannard & Johnson, 2004). 
 
In contrast, the skeletal muscle of sedentary individuals does not have a well developed 
capacity for fat oxidation, and is usually not involved in regular acute exercise bouts. 
Furthermore, even when they do exercise, fat oxidation is lower than in non-obese individuals 
Chapter 2  Literature Review 
 49
for the same relative intensities (Blaak, 2007). Blaak et al. (Blaak et al., 2000) showed that 
50% of fatty acids taken up by non-diabetic muscle were directly oxidised during β-
adrenergic stimulation, but in diabetic muscle, which had impaired FFA uptake, no oxidation 
was detected. In addition, Goodpaster et al. (Goodpaster, Katsiaras, & Kelley, 2003) observed 
that in obese individuals, improvement in insulin sensitivity resulting from a combined diet 
and exercise intervention, was associated with an increased reliance on fat oxidation during 
fasted conditions. Thus, it is believed that an imbalance between uptake and oxidation may 
lead to accumulation of lipid intermediates in skeletal muscle, thereby affecting the insulin 
signalling pathway and leading to insulin resistance (Blaak, 2007; Moro, Bajpeyi, & Smith, 
2008). In summary, the ‘paradox’ surrounding IMTG and how increased storage of IMTG 
leads to insulin resistance in obese individuals but not the endurance trained, suggests it is the 
oxidation of lipid, and therefore aerobic fitness and participation in regular exercise bouts, 
that dictate insulin sensitivity (Goodpaster, He, Watkins, & Kelley, 2001). 
 
2.5.6 Exercise physiology research with Māori 
To the authors knowledge, only one study exists which investigates the relationship between 
aerobic capacity and insulin sensitivity in Māori (Stannard, Holdaway, Sachinwalla, & 
Cunningham, 2007). This study showed that the aerobic capacity, or VO2maxLBM (mL.kg-1 
LBM.min-1) in Māori men, was able to help predict insulin sensitivity as measured by 
HOMA-IR. Only one other published study has made the comparison between insulin 
sensitivity and aerobic fitness, using VO2maxLBM. This study was done in a cohort of 
Caucasian men, and no observed relationship between VO2maxLBM and insulin sensitivity in 
this cohort was apparent (Coon, Bleecker, Drinkwater, Meyers, & Goldberg, 1989). 
 
Chapter 2  Literature Review 
 50
2.5.7 Summary 
A single exercise bout stimulates uptake of glucose and FFA, and if the exercise is of a 
sufficient duration, the turnover of IMTG. This reduction in intramuscular lipid lowers the 
risk of insulin resistance and increases insulin sensitivity. Muscle contraction also increases 
expression of glucose transporters and their translocation, improving glucose tolerance, 
independent of insulin action. Because of this, the demand on the pancreas and β-cells to 
secrete insulin is reduced with exercise, further reducing the risk of diabetes and β-cell failure. 
Chronic exercise improves the muscle cells ability to maintain healthy metabolic function 
(flexibility), by increasing mitochondria content, which in turn improves oxidative capacity 
and enhances skeletal muscle’s ability to utilize lipid as a substrate (J. O. Holloszy & Booth, 
1976). A higher aerobic capacity has also been shown to independently predict insulin 
sensitivity in Māori (Stannard, Holdaway, Sachinwalla, & Cunningham, 2007). Thus, regular 
and consistent exercise is likely an important means of maintaining glucose disposal in a 
healthy state, and reducing hyperglycaemia in the glucose intolerant.  
Chapter 2  Literature Review 
 51
2.6 Muscle Fibre type and Diabetes 
2.6.1 A history of fibre type and how it’s measured 
In the late 19th century, French anatomist Louis Antoine Ranvier observed that some muscle 
in rabbits differed in colour and that the redder muscle contracted at a slower rate than the 
muscle with a more pale coloration (Needham, 1926). From this simple observation, the basis 
of skeletal muscle fibre type study was formed and the understanding that muscle was a 
heterogeneous tissue was established. Just shortly after Ranvier’s experiments, Grutzner 
(cited in Needham, 1926) proposed that all muscle contained a mixture of two different fibres 
(red and white) and that the response of a muscle to stimulation (whether fast or slow) 
depended on the proportion of these two fibres within the muscle.  These early in-vivo studies 
were done with animals, until the advancement of surgical techniques made it possible to take 
samples of human muscle by biopsy (Zierath & Hawley, 2004). In the 1960’s, the needle 
biopsy technique was re-introduced to the field of physiology (Bergström & Hultman, 1966), 
enabling physiologists to obtain human muscle tissue while subjects were awake. Using this 
procedure, samples of human muscle became more readily available, allowing further 
development in the study of fibre-type. 
 
In the late 1960’s and early 1970’s, studies by Edstrom and Kugelberg (Edström & 
Kugelberg, 1968), as well as Burke et al. (Burke, Levine, Zajac Iii, Tsairis, & Engel, 1971) 
showed a relationship between physiological parameters and histochemistry within motor 
units. These studies paved the way for histological and biochemical analysis of muscle 
samples, leading to current classifications based on histochemical, biochemical, 
morphological or physiological characteristics (Scott, Stevens, & Binder-Macleod, 2001). 
 
Chapter 2  Literature Review 
 52
Since these early studies, many techniques have been developed to classify fibre type, making 
the interpretation of fibre type data more difficult, as classifications from different techniques 
do not always agree (Staron, 1997). At present, the most common way of identifying muscle 
fibres in humans is by dividing them into different categories according to the myosin heavy 
chain (MHC) isoform found within the fibre. Because myosin isoform content in a given 
muscle is quite stable and relatively unaffected by contraction, it has been identified as a 
suitable basis for fibre-type classification (Astrand, Rodahl, Dahl, & Stromme, 2003). 
Furthermore, measurement of MHC isoforms is probably the best method of fibre typing 
currently available, because MHC isoforms are identified within single muscle fibres by 
electrophoresis, meaning the analysis is quantitative.  
 
Another way muscle fibres are often identified is through histochemical staining to measure 
ATPase activity. The activity of ATPase, the enzyme responsible for breaking down ATP, is 
known to be associated with myosin isoform content (Billeter, Heizmann, Howald, & Jenny, 
1981) and the speed of muscle shortening (Barany, 1967). When histochemically analyzed, 
muscle fibres with high ATPase activity are identified as type II muscle fibres, while those 
fibres with low ATPase activity are identified as type I fibres. Differing pH sensitivities 
within type II fibres can then be used to separate type II fibres into two subgroups; type IIA 
and type IIX (also commonly known as IIB) (Brooke & Kaiser, 1970 - citation only; Rosen, 
1969). The problem with measuring ATPase activity with histochemical staining is that it is 
based on qualitative analysis of the different staining intensities, so that any given fibre could 
be classified differently by different researchers (Scott, Stevens, & Binder-Macleod, 2001).  
 
Presently there are three common fibre type divisions in humans based on MHC content; type 
I – slow oxidative, type IIA – fast oxidative and type IIX– fast glycolytic. There are 
Chapter 2  Literature Review 
 53
limitations in trying to define muscle fibre types however, as functional properties vary within 
groups of muscle fibres that have the same fibre type designation (Astrand, Rodahl, Dahl, & 
Stromme, 2003). This is largely due to the ‘plasticity’ of muscle fibres; their ability to 
transition from fast to slower fibre types and vice versa, in response to certain stimulus which 
will be discussed later. Because of this ‘plasticity’, many sub-categories of fibre type, those 
that are transitioning between ‘pure’ fibre types, and therefore express two MHC isoforms, 
have been identified within human muscle (Pette, 2001). However, this review will only deal 
with the 3 most commonly used classifications of fibre type, identified above.  
 
The needle biopsy technique also has many limitations. Within humans, fibre type varies 
between different muscles as well as within a given muscle (Blomstrand & Ekblom, 1982; J. 
A. Simoneau, Lortie, Boulay, Thibault, & Bouchard, 2008). Because biopsies are usually 
taken from only a few sites, this inter and intra-muscular variation is often not accounted for, 
and muscle samples are falsely considered as representative of the fibre type proportion of the 
whole muscle (Lexell, Taylor, & Sjostrom, 1985). Blomstrand and Ekblom reported that fibre 
type percentage varied between two biopsies taken from the same leg  by 6% for type I fibres, 
4% for type IIA and 5% for type IIX (Blomstrand & Ekblom, 1982). When the samples were 
taken from both legs this variation increased to 12%, 7% and 7% respectively. Furthermore, it 
appears that fibre type distribution varies depending on the depth of the muscle sample, with a 
greater proportion of type II fibres close to the surface and type I fibres in deeper regions of 
the muscle (Lexell, Henriksson-Larsen, & Sjostrom, 1983). The varying distribution of fibre 
type within whole muscle raises concern regarding the repeatability of obtaining muscle 
samples by biopsy, and the biopsy technique’s reliability in obtaining muscle samples to 
identify those with a certain ‘fibre type proportion’. What’s more, the muscle biopsy 
Chapter 2  Literature Review 
 54
technique is invasive and requires the skill of a trained physician with specialized equipment 
(Bergström, 1975), making it impractical for many research and clinical settings. 
 
In accordance with early observations of different fibre types having different functional 
properties (Needham, 1926), more recent studies have shown, using non-invasive techniques, 
that contractile function of muscle, such as time to peak torque and fatigability measured 
externally, is related to fibre type proportion in-vivo (Ivy, Withers, Brose, Maxwell, & Costill, 
1981; MacIntosh, Herzog, Suter, Wiley, & Sokolosky, 1993; Sadoyama, Masuda, Miyata, & 
Katsuta, 1988; E. Suter, Herzog, Sokolosky, Wiley, & Macintosh, 1993; Thorstensson & 
Karlsson, 1976). To the author’s knowledge, these techniques have not yet been used as a 
surrogate measure of fibre type proportion in larger studies. Nevertheless, the relationships 
observed between histo/biochemically determined fibre type and functional characteristics, 
indicate that methods which measure muscle contractile properties externally, may provide a 
non-invasive way to estimate fibre type proportion in future studies. Therefore, further 
development of these techniques is required.  
 
One method of modelling muscle contraction which has been used a great deal in the clinical 
setting is electrical stimulation (ES). ES enables muscle to contract independent of the central 
nervous system and is therefore useful for maintenance of strength, muscle mass and physical 
activity levels during muscle immobilization/paralysis (Delitto & Snyder-Mackler, 1990). 
This technique is useful in a research setting as the magnitude of muscle contraction can be 
controlled with ES, and psychological factors such as motivation are bypassed. For example, 
electrical stimulation has been used to study the role of the nervous system in force 
production. By administering an electrical impulse to a maximal voluntary contraction, and 
observing that greater force was produced with the addition of electrical stimulation, early 
Chapter 2  Literature Review 
 55
researchers showed that neural factors limited maximal force production rather than the 
muscle structure (Shield & Zhou, 2004). The most important factor to consider when 
administering percutaneous ES is the location of electrode on the skin. Placement of electrode 
pads should ensure that antagonist muscles are not being stimulated at the same time as 
agonists as incorrect placement of electrodes can alter force measurements (Shield & Zhou, 
2004). Nevertheless, apart from error associated with electrode placement, technical error is 
small, as the administration, the timing and magnitude of the twitch, is controlled 
electronically. 
 
Although the ES technique has been used extensively in a research setting as a substitute for 
voluntary exercise, its use in stimulating muscle to measure contractile properties has hardly 
been employed. The only published study I know of which has employed ES to measure 
muscle contractile properties is one by Blimkie et al. (Blimkie, Sale, & Bar-Or, 1990). Thus, 
although validity and limitations have been discussed in regard to the use of electrical 
stimulation, these discussions have focused on the validity of ES as a tool for simulating 
exercise rather than creating twitches for measurement of contractile properties. Therefore, 
the validity of ES for use in this way, and its limitations, has not been addressed. 
 
2.6.2 Metabolic and functional characteristics of fibre type 
A muscle’s contractile properties are determined largely by the myosin heavy chain (MHC) 
protein that is expressed in the muscle fibres within (Bottinelli, Betto, Schiaffino, & Reggiani, 
1994). However, it seems that the innervation of muscle fibres may be the primary 
determinant of MHC, and thus fibre type. A classic study by Buller et al. (Buller, Eccles, & 
Eccles, 1960) showed that slow-twitch muscles become faster when reinnervated by the motor 
neuron that originally innervated a fast muscle, and vice versa. This suggests that the 
Chapter 2  Literature Review 
 56
contractile characteristics of a muscle are ultimately the result of the activity of the motor 
neuron which innervates it and that any changes in structure, phenotype or contractile protein 
content, are secondary functions in response to the innervation of the fibre.  
 
Enzyme activities also differ between fibre types. Essen et al. found that activity of SDH 
(succinate dehydrogenase) and HAD (3-hydroxyacyl-CoA-dehydrogenase), important 
oxidative enzymes, are higher in slow-twitch fibres, while activity of PFK 
(phosphofructokinase), an important glycolytic enzyme, is higher in fast-twitch fibre (Essén, 
Jansson, Henriksson, Taylor, & Saltin, 1975). Others have reported a similar relationship 
between fibre type and glycolytic/oxidative enzyme activity (He, Watkins, & Kelley, 2001). 
These relationships are not surprising, considering the recruitment of these different fibres 
during exercise. In accordance with Henneman’s ‘size principle’ (cited in Enoka & Stuart, 
1984), slow-twitch fibres, which are part of a relatively small motor unit, are recruited during 
low-intensity activity and can maintain contraction (albeit at low-intensities) for prolonged 
periods. Accordingly, slow-twitch fibres require a greater oxidative capacity to provide fuel 
during prolonged exercise. Fast-twitch fibres on the other hand, are part of a larger motor unit 
and are therefore recruited during more intense bouts of exercise that require maximal 
contractions fuelled mainly by the glycolytic system.  
 
It appears that glucose transport is fibre type specific as well. Henriksen et al. showed that 
expression of GLUT4 transport protein was greatest in the slow-twitch fibres of rats (E. J. 
Henriksen et al., 1990). Similarly, Daugaard et al. (Daugaard et al., 2000) demonstrated that 
in humans GLUT4 expression was slightly higher (P<0.05) in slow-twitch than in fast-twitch 
fibres, both before and after a 2 week training protocol. 
 
Chapter 2  Literature Review 
 57
Table 2-2 – Metabolic, structural & functional characteristics of different muscle fibre types 
(Adapted from Wilmore & Costill, 2005 & Zierath & Hawley, 2004) 
 
 
2.6.3 Fibre type and exercise 
As can be seen in the table above, type II fibres have a lower oxidative capacity than type I 
fibres. This is important to note as it is thought that the oxidative capacity of skeletal muscle 
is related directly to its ability oxidize fat (David E. Kelley, Goodpaster, Wing, & Simoneau, 
1999), and therefore take up glucose in response to insulin. If this is the case, an individual 
with a greater proportion of type II fibres in their muscles would be expected to be at greater 
Characteristic Slow-twitch Fast-twitch  
(oxidative) 
Fast-twitch 
(glycolytic) 
MHC MHC I MHC IIA MHC IIX 
Type of myosin ATPase Slow Fast Fast 
Time to peak tension High Low Low 
Fibres per motor neuron Few Many Many 
Motor unit force Low High High 
Aerobic capacity (oxidative) High Moderate Low 
Anaerobic capacity (glycolytic) Low High High 
Mitochondrial density High Moderate Low 
GLUT4 Content High Medium Low 
Cross sectional area Small Medium Large 
Motor neuron size Small Large Large 
Nerve conduction velocity Slow Fast Fast 
Fatigue resistance High Moderate Low 
Insulin sensitivity High Medium Low 
Capillary density High Moderate Low 
Chapter 2  Literature Review 
 58
risk of lipid accumulation (obesity) and whole body insulin resistance, than one with a high 
proportion of type I fibres.  
 
Although exercise is well known to increase insulin sensitivity, Cortez et al. (Cortez, Torgan, 
Brozinick, & Ivy, 1991) showed that the effect of exercise on insulin sensitivity is only 
significant within the fibres recruited during the exercise. Their studies on obese Zucker rats 
revealed that with low-intensity training, glucose transport only improved in fast-twitch red 
fibres (type IIA). On the other hand, glucose transport in fast-twitch white fibres (type IIX) 
only improved after high-intensity training. Accordingly, Daugaard et al. (Daugaard et al., 
2000) showed that after endurance-type training in humans, GLUT4 expression only 
increased in slow-twitch fibres. What’s more, Koopman et al. (Koopman et al., 2006) found 
that IMTG only decreased in type-1 muscle fibres after resistance training in untrained 
healthy males, but that the decrease was closely related to fibre-type specific oxidative 
capacity. On the other hand, this same study showed that the greatest decrease in glycogen 
content was in type IIX fibres. These findings all suggest that the increased GLUT4 content 
and increased glycogen and IMTG turnover associated with exercise training, is more a 
function of the oxidative capacity and activity level of the muscle fibre, rather than the fibre 
type per se (Daugaard & Richter, 2001).  
 
As discussed earlier, the recruitment of type II fibres increase as increased spatial recruitment 
is required, such as occurs during high intensity exercise. Type I fibres are recruited first to 
provide the majority of tension during endurance type exercise. Hence, the intensity of 
exercise prescribed to the glucose intolerant individual, should be carefully considered. At 
present, biopsies are required to obtain a direct measure of fibre type proportion, but this 
procedure is very invasive and not practical for large studies. Thus, developing a less invasive 
Chapter 2  Literature Review 
 59
way of estimating muscle composition could provide a means of identifying those who are 
most at risk of obesity and metabolic disorders, and allow prescription of fibre-type specific 
exercise. 
 
2.6.4 Fibre type and insulin resistance 
As previously discussed, insulin-stimulated glucose uptake and metabolism within skeletal 
muscle is influenced by fibre type (Hickey, Weidner et al., 1995). Accordingly, fibre type 
proportion has been related to whole body insulin sensitivity and obesity. Lillioja et al. 
(Lillioja et al., 1987a) found that a higher proportion of type IIX (fast-twitch B) fibres is 
associated with many risk factors of type-2 diabetes, while insulin sensitivity correlates 
positively with the percentage of type-1 fibres. Likewise, Wade et al. (Wade, Marbut, & 
Round, 1990) found an inverse relationship between fatness and proportion of slow-twitch 
fibres in a group of untrained males. Other studies have also shown increased proportions of 
type IIX fibres in type-2 diabetics (Hickey, Carey et al., 1995; Marin, Andersson, 
Krotkiewski, & Bjorntorp, 1994). Tanner et al. (Tanner et al., 2002) showed that morbidly 
obese women with a higher proportion of type-1 fibre, had a greater capacity for weight loss 
12 months after a surgical weight loss intervention, when compared with those with more 
type-2 fibres. This finding is further supported by a study by Wade et al. (Wade, Marbut, & 
Round, 1990) who found that untrained men with a greater proportion of slow-twitch fibre, 
had a greater ability to oxidize fat during exercise, when compared to those with more fast-
twitch fibres. Along with the fact that slow-twitch fibres have a greater capillarization and 
mitochondrial content than fast- twitch fibres (Ingjer, 1979), these findings are consistent with 
the idea that impaired insulin action in skeletal muscle is a function of an inability to properly 
oxidize lipids (Goodpaster, Katsiaras, & Kelley, 2003).  
 
Chapter 2  Literature Review 
 60
It has been reported that the balance between oxidative and glycolytic enzymes in the muscle 
of obese individuals is related to insulin resistance (J.-A. Simoneau & Kelley, 1997). IMTG 
content  is also strongly related to insulin resistance (Shulman, 2000). Fittingly, He et al. 
showed that oxidative enzyme activity was reduced in all fibre types of obese and diabetic 
individuals when compared with lean subjects (He, Watkins, & Kelley, 2001). They also 
demonstrated that the ratio of oxidative enzyme activity to lipid content within all fibre types 
was significantly lower in obese and type-2 diabetic individuals. Though type I fibres have a 
higher concentration of IMTG, this fibre type’s greater oxidative capacity and ability to 
mobilize fatty acids allows for a greater turnover of the accumulated lipid (Blaak, 2007; 
Moro, Bajpeyi, & Smith, 2008). This suggests furthermore, that the rate of turnover, rather 
than the concentration of IMTG, determines insulin sensitivity. Considering that fibre-type is 
related to the ability to utilize fat during exercise, an individual’s muscle composition may 
play a role in determining the effectiveness of exercise interventions in reducing adiposity, 
and improving insulin sensitivity. 
 
2.6.5 Fibre type proportion: Genetically or environmentally determined? 
When observing a variety of sporting and athletic events at the elite level, it is difficult not to 
notice an ethnic bias towards success in particular events. It is common to see an African-
American running the 100m sprint at the Olympics, while Caucasians usually find success in 
long-distance events (Cited in Suminski, Mattern, & Devor, 2002). On the other hand it is not 
uncommon to see an African Marathon runner take a gold medal, or a European dominate the 
power lifting. It is often remarked that some ethnicities are genetically predisposed to excel in 
different athletic events (Hunter, 1998). Alternatively factors like socio-economic status or 
exposure to certain sports may also explain these ethnic biases. 
 
Chapter 2  Literature Review 
 61
From their classic studies which investigated fibre type proportion in monozygous and 
dizygous twins, Komi et al. (Komi et al., 1977) were the first to provide evidence that fibre 
type proportion is mostly determined by genetic influences. In accordance with these findings, 
Ama et al. showed that a cohort of Black men had a lower percentage of type-1 fibres when 
compared to Caucasian men (Ama et al., 1986). However, the differences in fibre type 
proportion between the two ethnic groups in this study, was only slightly larger than the 
sampling error which occurs when using the needle biopsy to collect muscle tissue. However, 
a study in the US found that obese African-American woman had a lower percentage of type-
1 fibres than obese Caucasian women (Tanner et al., 2002),  while others have showed in a 
cohort of sub-elite endurance runners, that African men had lower proportions of type I 
muscle fibre than European men (Kohn, Essen-Gustavsson, & Myburgh, 2007). Others 
challenge this idea however, suggesting that environmental factors explain a large part of the 
variation (J. A. Simoneau & Bouchard, 1995). 
 
Simoneau and Bouchard (J. A. Simoneau & Bouchard, 1995) suggest that fibre type 
proportion can be partly explained by inherited factors (45%) and partly by environmental 
factors (40%). Elite athletes involved in endurance exercise appear to have a high proportion 
of slow-twitch fibres, while sprinters have more fast-twitch fibres (Costill et al., 1976). While 
this seems to imply a fibre-type change with training, selection into events by individuals 
possessing a natural endowment can occur (P. D. Gollnick, Armstrong, Saubert, Piehl, & 
Saltin, 1972), which may in part explain the ethnic grouping observed in certain athletic 
events. 
 
Interestingly, studies have shown that muscle fibre proportions can change with training. 
Adams et al. (Adams, Hather, Baldwin, & Dudley, 1993) found that resistance training led to 
Chapter 2  Literature Review 
 62
a decrease in type IIB fibres and an increase in type IIA fibres in a cohort of healthy men. 
Staron et al. (Staron et al., 1991) also saw an increase of type IIA fibres and a decrease in type 
IIB fibres after resistance training in a cohort of women that had detrained previous to the 
study. Both of these studies also showed that detraining causes the conversion in reverse 
(from IIA to IIB). Similar to resistance training, it appears that the shift from IIB to IIA occurs 
with endurance training also (Schantz, Billeter, Henriksson, & Jansson, 1982). A reduction in 
the proportion of type I fibres is seen in the denervated muscle of paraplegics, but can be 
reversed with muscle contraction by way of electrical stimulation (Martin, Stein, Hoeppner, & 
Reid, 1992; Rochester et al., 1995). All of these findings suggest that chronic contractile 
activity induces a conversion from faster, fatigable muscle fibres, to slower, fatigue resistant 
fibres. Detraining or denervation on the other hand, induces the conversion in the opposite 
direction, from slow to fast fibres. Nevertheless, it appears that in normal individuals fibre 
type conversion in response to training or detraining, is limited to the conversion from IIA to 
IIX and vice versa (P. Andersen & Henriksson, 1977; Houmard et al., 1993). While, studies in 
rats have revealed that electrical stimulation within denervated fast-twitch muscle fibres, can 
‘transform’ these fibres into slow-twitch muscle (Windisch, Gundersen, Szabolcs, Gruber, & 
Lømo, 1998), no conclusive research exists to show conversion from type IIA to type I fibres 
in humans (Daugaard & Richter, 2001). Some even propose that exercise increases the 
oxidative capacity of muscle fibres rather than actually changing fibre type composition (P. D. 
Gollnick, Armstrong, Saubert, Piehl, & Saltin, 1972).  
 
2.6.6 Summary 
In summary, though it was previously thought that fibre type was solely genetically 
determined, it has more recently been shown that conversion between fibre types does exist, 
although to a small degree, with environmental stimulus. Ethnic differences in fibre type 
Chapter 2  Literature Review 
 63
proportion also exist, but caution must be used in interpreting the results of these studies as 
confounding factors that also relate to ethnicity, such as socioeconomic status, may explain 
part of the variation. It appears that chronic exercise induces a switch from faster, less insulin 
sensitive muscle to slower, more insulin sensitive fibres. Inactivity on the other hand, causes a 
shift from slower to faster muscle fibres. If this is so, the physiological benefits of exercise 
would be two-fold for those at risk of insulin resistance; first, from reductions in lipid content 
and improved oxidative capacity of individual muscle fibres, and second from an increased 
proportion of the slower, more insulin sensitive fibre types. Genetics, rather than determining 
a set fibre type proportion, may instead provide the limits of fibre type conversion in response 
to stimulus. Because of the invasive nature of the muscle biopsy, research on fibre type 
proportion has been limited in humans. Less invasive procedures, which measure muscle 
contractile properties externally, have produced results that are closely related to the fibre 
type proportions obtained through biopsy (E. Suter, Herzog, Sokolosky, Wiley, & Macintosh, 
1993). However, it appears that these methods have not been utilized as a substitute to the 
biopsy technique in research, or to identify those at risk of insulin sensitivity. Exercise 
induced improvements in glucose tolerance are specific to the fibre types recruited, thus 
developing a non-invasive means to measure fibre type proportion could be valuable in 
prescription of fibre-type specific exercise. 
 
  
Chapter 2  Literature Review 
 64
2.7  The influence of environment and genes on diabetes 
prevalence 
2.7.1 Ethnicity and health research 
While obesity, by way of caloric imbalance, has been identified as the major factor associated 
with insulin resistance, it appears that some populations are still predisposed to this condition 
when body fatness is taken into account (Scragg, Baker, Metcalf, & Dryson, 1991; David 
Simmons & Thompson, 2004). Because most data revealing an ethnic disparity in health is 
based on national surveys or community questionnaires where ethnicity is self-identified, 
there are problems when interpreting this data in a genetic or physiological context. 
Accordingly, there is much debate surrounding the collection and analysis of ethnicity data in 
New Zealand and around the world.  
 
First of all, because the meaning of words such as ‘race’, ‘ethnicity’ and ‘culture’ are often 
confused in the literature (D. R. Thomas, 2001), interpretation and application of such data is 
difficult. For example, of the 80 participants that were categorised as Māori in the Auckland 
Region Coronary or Stroke Study (ARCOS) only 50% (40) were classified as Māori in the 
death registration data (Graham, Jackson, Beaglehole, & De Boer, 1989). The reason for this 
discrepancy was that participants in the ARCOS study were classified by “cultural 
affiliation”, while ethnicity in the death registration was classified according to “biological 
origin – half or more Māori blood” (D. R. Thomas, 2001). Additionally, Thomas  points out 
that changes in the census question regarding an individual’s ethnicity, which happened from 
1981 to 1996, may not have been taken into account when interpreting health data over this 
time period (D. R. Thomas, 2001). Secondly, because the concepts of race and ethnicity are 
more representative of socio-cultural identification, rather than a common genetic ancestry 
Chapter 2  Literature Review 
 65
(Rebbeck & Sankar, 2005), the application of ethnicity data, in identifying physiological 
bases for health disparities, has been deemed invalid by some (Goodman, 2000). 
 
In spite of this, there is clear evidence that self-identified ethnicity is closely associated with 
genetic variability (Tang et al., 2005), suggesting that while ethnicity may be more closely 
linked to socio-cultural factors, there is still a genetic basis to ethnicity. Nevertheless, 
methods of differentiating groups according to genetic ancestry are limited. In fact, only one 
study to my knowledge, has observed disparities in type-2 diabetes prevalence according to 
genetic ancestry (Gardner et al., 1984), and this study did show that type-2 diabetes did, in 
fact, correlate with genetic ancestry. However, the method used in this study to quantify 
ancestry was skin colouration, which is not necessarily an accurate representation of ancestry 
(Parra, Kittles, & Shriver, 2004).  
 
Self-identified ethnicity is a complex concept which is difficult to interpret in the context of 
physiological research. Therefore, further research and discussion is required to ensure that 
ethnicity data is collected, analysed and interpreted appropriately when applied to health, 
medicine and scientific research. 
 
2.7.2 Environment vs. genetics 
Because problems exist in the interpretation of ethnicity data, it is difficult to understand the 
relative contribution of environmental and genetic influences in ethnic health disparities. For 
example, Diamond (Diamond, 2003) described how the lowest prevalence of diabetes is in 
rural third world areas where food is scarce and people are forced to expend energy to obtain 
food. Whereas, prevalence is highest in rural areas where food is easily obtained and a 
Chapter 2  Literature Review 
 66
sedentary lifestyle has become the norm, emphasizing the role of environmental factors on 
ethnic disparities in obesity and diabetes.  
 
On the other hand, Diamond points out that prevalence of diabetes is relatively low among 
European groups living urban lifestyles and highest among Nauru islanders, Pima Indians of 
Arizona and Wanigela people in Papua New Guinea, all of which have only (relatively) 
recently adopted a ‘Western’ way of life. Thus, although Māori and other ‘high-risk’ ethnic 
groups appear to have an abnormal predisposition toward obesity and its co-morbidities, it is 
actually European groups who are unique, in that the prevalence of obesity and diabetes is 
relatively low even in a similar environment.  
 
From the literature, a number of physiological characteristics can be identified which may be 
responsible, in part, for the disparity in metabolic control that can be seen between Europeans 
and other ethnic groups, such as NZEO and Māori in New Zealand. Physiological factors 
associated with insulin resistance include unfavourable body fat distribution (S. E. Kahn et al., 
2001), a decreased fat-storage capacity in adipose tissue (Lewis, Carpentier, Adeli, & Giacca, 
2002), a greater tendency for fat accumulation in muscle (D. A. Pan et al., 1997) and liver 
tissue (Seppälä-Lindroos et al., 2002), a greater proportion of fast-twitch muscle fibre (Lillioja 
et al., 1987a), reduced mitochondrial content/function (D. E. Kelley, He, Menshikova, & 
Ritov, 2002), and diminished insulin secretion (β-cell function) (Hollenbeck & Reaven, 
1987).  
 
Although literature is abundant which supports the stance that environmental factors 
determine ethnic health disparities, the purpose of this part of the review is to focus on 
physiological explanations for ethnic health disparities. 
Chapter 2  Literature Review 
 67
2.7.3 Genetic adaptation and ethnic health disparities 
In view of the physiological characteristics associated with lipid accumulation and insulin 
resistance, many authors have sought to explain predispositions to obesity and poor metabolic 
health, in the context of genetic adaptation to environmental stresses. For example, Diamond 
reasons that, in response to gradually increasing food supply in Europe between the fifteenth 
and eighteenth century, Europeans may have developed a protective adaptation to the effects 
of caloric imbalance (Diamond, 2003). Alternatively, Fridlyand and Philipson (Fridlyand & 
Philipson, 2006) propose that the physiological characteristics which appear to protect 
Europeans from unfavourable lipid accumulation and insulin resistance, can be explained by, 
what they term, ‘cold climate genes’. Their theory is based on the thought that as humans 
migrated to colder areas in the North, the metabolic functions associated with thermogenesis, 
adapted to aid survival in cold climates. Thermogenesis is largely associated with the basal 
metabolic rate (BMR), which is determined by metabolic processes in the mitochondria of 
skeletal muscle. In context of obesity and insulin resistance, basal metabolic processes are 
responsible for almost all expenditure of energy for an individual living a sedentary lifestyle. 
The implication is that those with an increased metabolic rate, by way of a genetic adaptation 
to cold climate, would have an advantage in reducing adiposity over others living a sedentary 
lifestyle, through an increased ability to oxidize excess substrate. Fridlyand and Philipson 
point out that while the ancestors of Europeans, who lived in many far North areas, would 
have benefited from such an adaptation, the adaptation would not have occurred in the 
ancestors of Africans, South East Asians and Pacific Islanders, who originate from equatorial 
areas. In connection with this, African American women were found to have a lower RMR 
than Caucasian women, even when lean body mass was accounted for (Forman, Miller, 
Szymanski, & Fernhall, 1998). Likewise, a study by Rush et al. (E. C. Rush, Plank, & 
Robinson, 1997) revealed that the resting metabolic rate of young Māori and Polynesian 
Chapter 2  Literature Review 
 68
woman was lower than in young Caucasian women. These authors also propose that this 
decreased resting metabolic rate may predispose Māori/Polynesians to obesity and the onset 
of diabetes. Fridlyand and Philipson observe that an increase in the number of mitochondria, 
or increased mitochondrial activity, could be the mechanism that adapts in a cold climate, 
increasing basal metabolic rate and heat production (Fridlyand & Philipson, 2006). Others 
have also suggested that physiological adaption to cold climate is associated with increased 
mitochondrial content or activity (Lowell & Spiegelman, 2000). Interestingly, mitochondrial 
content is also associated with aerobic capacity and fibre type, both of which are associated 
with insulin sensitivity. 
 
As discussed previously, fibre type proportion differs between ethnicities and a greater 
proportion of fast-twitch muscle fibre is associated with increased risk of insulin sensitivity 
and diabetes. It is interesting to note that in many instances the same ethnic groups which 
have been identified as predominantly ‘fast twitch’ in physiological studies, are also those 
who are identified as metabolically disadvantaged in epidemiological studies. For instance 
Tanner et al. found that obese African American women possessed a greater proportion of 
fast-twitch fibre when compared to obese Caucasian men (Tanner et al., 2002). What’s more, 
Ama et al. found that sedentary Black men had a greater proportion of fast twitch fibre 
compared with Caucasian men (Ama et al., 1986). In accordance with these findings, the 
1999/2002 National Health and Nutrition Examination Survey (NHANES) showed that 11% 
of African-Americans were diabetic compared with just 5.2% of Caucasian Americans 
(Cowie et al., 2006). Likewise, diabetes is more than twice as prevalent in Māori than in 
NZEO (when adjusted for age) (Ministry of Health, 2004b, , 2008). Thus, it is possible that a 
genetically determined fibre type proportion may be an underlying factor in predisposing, or 
protecting, certain ethnic groups from the detrimental effects of a sedentary lifestyle. 
Chapter 2  Literature Review 
 69
Nevertheless, no studies to date have investigated fibre type proportion in Māori, so the 
application of this theory to Māori is limited to assumptions gained from fibre-type studies 
done in other ethnic groups who share a disproportionate prevalence of obesity and diabetes.  
 
Similar to the theories previously mentioned, Neel’s ‘thrifty gene’ theory also seeks to explain 
ethnic disparities in insulin resistance as a result of survival adaptations. Neel’s theory 
however, is based on the metabolic adaptations geared toward survival from starvation, rather 
than cold climate. This theory suggests that insulin resistance is a natural response to food 
scarcity, which preserves blood glucose for brain function (Neel, 1962; Stannard & Johnson, 
2004). When combined with excessive feeding and a sedentary lifestyle however, over supply 
of substrate and energy imbalance makes this mechanism for survival a destructive condition. 
 
In recent years, many more theories have been postulated such as  the ‘adipose tissue 
overflow hypothesis’ (Sniderman, Bhopal, Prabhakaran, Sarrafzadegan, & Tchernof, 2007) 
which proposes that some ethnic groups, in this case South-East Asians, may have a reduced 
capacity to store fat in subcutaneous adipose tissue, leading to a greater accumulation of 
abdominal fat. Thus, because a greater distribution of fat in the abdominal area is associated 
with a higher-risk of metabolic disorder (S. E. Kahn et al., 2001; Ross, Aru, Freeman, 
Hudson, & Janssen, 2002), and because Asian populations seem to have greater abdominal 
adiposity compared with Caucasians of the same BMI (Wang et al., 1994), the effect of 
caloric imbalance on insulin sensitivity is greater in this ethnic group than those who are less 
prone to central adipose storage. Similar to the ‘cold climate’ theory discussed previously, 
Sniderman et al. suggest that the reduced capacity to store fat in subcutaneous adipose tissue 
could be an adaptive response to climatic influences. However, ‘adipose tissue overflow’ may 
not be a valid explanation in Māori and Polynesians, who are more muscular than Asians and 
Chapter 2  Literature Review 
 70
Caucasians of the same BMI  (P. Deurenberg, Yap, & van Staveren, 1998; Swinburn, Ley, 
Carmichael, & Palnk, 1999), and have a relatively lower abdominal distribution of fat (E. 
Rush et al., 2004).  
 
Although none of the theories discussed have been investigated in Māori, these theories 
warrant further investigation regarding the contribution of environmental and genetic factors 
to the development of obesity and its associated metabolic disorders. When taking into 
account the theories presented, it is possible that Māori, who likely descend from ancestors 
that lived close to the equator, and also faced periods of starvation, have a double genetic 
disadvantage when faced with a lifestyle of caloric imbalance by having ‘thrifty genes’, 
without the advantage of ‘cold climate genes’. 
 
2.7.4 Summary 
A number of theories have been developed to explain the genetic adaptations responsible for 
ethnic disparities in metabolic control seen in developed countries. When taking these theories 
into account, Māori are genetically disadvantaged in a modern, sedentary lifestyle. It appears 
that a genetically determined fibre type proportion may also increase the risk of obesity and 
therefore insulin resistance and diabetes. While some studies have shown fibre type to be 
genetically determined, it appears that environmental factors, such as training, may be just as 
important in determining fibre type proportion. A greater understanding of fibre type, and 
whether metabolic processes differ between ethnic groups, may shed some light on why the 
disparities exist in obesity, insulin resistance and diabetes. Because ethnicity is a social 
construct and not necessarily biological, developing methods to measure someone’s ancestry 
could also prove valuable. 
Chapter 3  Methodology 
 71
3 The Interface between Exercise Science and 
Māori Health 
3.1 Introduction 
When the World Health Organisation (WHO) was first established, it defined health as a state 
of complete physical, mental and social wellbeing and not merely the absence of disease or 
infirmity (WHO, 1946). Similarly, a Māori view of health is holistic and focuses on wellbeing 
rather than just absence of illness (M. H. Durie, 1985). Past experience however, suggests that 
the focus of mainstream health systems and health research seems to be on the treatment of 
illness (Booth, Gordon, Carlson, & Hamilton, 2000). Furthermore, indigenous views are often 
overlooked as health is approached in a Western manner (Sachdev, 1998). Because we view 
the world differently, a Māori approach to health and research (along with all other aspects of 
life) differs from that of the ‘norm’ taken by non-Māori. Cultural sensitivity is important and 
can become an ethical issue when assumptions are made that certain requirements for research 
and interventions which are appropriate for non-Māori must also be appropriate for Māori. 
Because the Māori view of hauora (often incorrectly used synonymously as a translation of 
the English word for ‘health’) differs from a non-Māori view of health, the measures of 
success for hauora also differ from those expressed for the idea of non-Māori health. Hauora 
includes a spiritual and holistic view and is interdependent, involving friends and whānau, 
compared with the naturalistic, individualistic view of mainstream health models (M. Durie, 
1994).  
 
As lifestyles in developed countries, such as New Zealand, Australia and the US become 
more sedentary, and food supply more abundant, prevalence of obesity related illnesses such 
as diabetes have increased dramatically (Diamond, 2003). The different ethnicities living 
Chapter 3  Methodology 
 72
within these developed countries seem to be disproportionately affected by this growing 
epidemic however, with indigenous groups suffering most (Gohdes; Moore & Lunt, 2000; 
Young, Reading, Elias, & O'Neil, 2000; Zimmet, 1979). Similar to indigenous people of other 
westernized nations, Māori in New Zealand are effected more by diabetes than New Zealand 
Europeans (K. McAuley et al., 2001; David Simmons & Thompson, 2004).  
 
Although intervention programs have been developed to slow the increasing prevalence of 
diabetes, recruitment of Māori into many research and health intervention programs has not 
been successful (Murphy et al., 2003; Ni Mhurchua et al., 2007; Tuttle, 2002). While 
programs aimed at improving Māori well-being have good intentions, success of these 
programs is not dependant on the quality of the information presented but whether its 
approach is culturally appropriate. Programs that are aimed at Māori are no doubt beneficial 
to Māori health but cannot be expected to succeed if recruitment and delivery of services 
continues to be approached in a Western manner. Māori will continue to respond poorly if 
researchers and those who develop interventions continue to overlook Māori views; and 
according to current trends, if Māori continue to respond poorly and today’s interventions are 
unsuccessful, an unsustainable burden on the health system (PricewaterhouseCoopers, 2001) 
and the endangerment of the Māori people is inevitable.    
 
3.2 The Scientific Approach 
Western science tends to be reductionist, and therefore seeks to understand behaviour and 
actions by breaking specific scientific units or concepts into smaller parts. According to a 
reductionist view, diseases and health conditions can be explained by understanding the 
cellular pathogenesis associated with the condition (H. Andersen, 2001). When related to 
diabetes for instance a biomedical view would seek to understand the condition’s relationship 
Chapter 3  Methodology 
 73
to cellular mechanisms and observe molecular imbalances within the cell. Though the 
physiological mechanisms of the condition are better understood due to this reductionist 
approach, the pharmaceutical interventions developed from this understanding does not seem 
to prove more effective than the prescription of whole body exercise (Stannard & Johnson, 
2006). It has also been argued that the reduction of many diseases to cellular and molecular 
mechanisms is so complex that no conclusive explanation of disease pathology can be 
obtained through reduction anyway (H. Andersen, 2001). This complexity could explain why 
the mechanism/s that leads to insulin resistance still elude researchers as research suggests 
that insulin resistance is a result of a disorder in more than one part of cellular metabolism or 
physiology (B. B. Kahn & Flier, 2000; Stannard & Johnson, 2004). 
 
A scientific understanding of physiological mechanisms is important however as it allows us 
to better understand the reason behind trends observed by epidemiological studies. 
Physiological findings have also been a driving force in motivating legislative change related 
to health problems (Booth, Gordon, Carlson, & Hamilton, 2000).  
 
Another characteristic of Western science is that it is naturalistic in that it disregards any 
supernatural phenomenon as an explanation (Harris & Mercier, 2006). A disregard for that 
which many indigenous people may consider as spiritual knowledge, whether intended or not, 
can be a barrier to Māori acceptance of scientific research as a valuable tool in their 
development. 
 
Although science exists as a vehicle to understand the unexplained, it has drawn a clear 
distinction between itself and indigenous understanding. Because indigenous knowledge 
includes a spiritual aspect, approaches research differently and is based on a different set of 
Chapter 3  Methodology 
 74
values, it is pushed aside and branded irrelevant to scientific knowledge. Science in its purest 
and most basic sense however is not just the standard chemistry, physics and biology 
associated with modern sciences, but is defined as knowledge, whether obtained through the 
usual scientific methodology of observation, experimentation and theory or not (Harris & 
Mercier, 2006).  
 
3.3 The Holistic Approach – The Māori Approach  
The WHO and other health organisations are taking a more holistic view of health than typical 
medical sciences have taken in the past. A holistic approach maintains that all levels of social, 
psychological and biological health are linked to each other so that if one changes, the others 
are affected. According to this view, the characteristics of wellbeing, a collective outcome, 
cannot be understood by focusing on the characteristics of each component separately (H. 
Andersen, 2001). 
 
Where Māori views differ from that of the holistic view taken by the WHO however is that 
the Māori holistic view of health includes a spiritual aspect. This spiritual aspect does not 
only encompass the Māori link to God and the divine, but includes a strong sense of unity 
with their environment, which link is thought to be the most defining element of indigeneity 
(Mason Durie, 2004; Kame'eleihiwa, 1992). Indigenous people globally have holistic views 
similar to the Māori concept of health and wellbeing, known to Māori as hauora (Smylie, 
Anderson, Ratima, Crengle, & Anderson, 2006). This concept includes a balance between the 
mental (hinengaro), physical (tinana), social/extended family (whānau) and spiritual (wairua) 
dimensions of health (M. H. Durie, 1985; Smylie, Anderson, Ratima, Crengle, & Anderson, 
2006). Māori Health models such as Whare Tapa Wha, developed by Mason Durie, represent 
these four aspects of health and have been adopted by many as a contemporary Māori view of 
Chapter 3  Methodology 
 75
health. Māori are also more concerned about the health and wellbeing of whānau collectively 
rather than that of the individual (M. Durie, 1994).  
 
Though it may be seen that science has disregarded the legitimacy of indigenous knowledge, 
indigenous people often dismiss the scientific method due to its failure to recognise spiritual 
aspects of learning (Mason Durie, 2004). In order for Māori to develop in a colonized world, 
Māori health researchers must ask the question “How can Māori health issues be better 
understood through Western-science, while still maintaining cultural integrity?”  
 
3.4 Exercise Science – The Bridge 
Though research would suggest many indigenous groups to be physiologically susceptible to 
diabetes and its complications, a few studies show that prevalence of diabetes is low in 
indigenous people that continue to live a traditional lifestyle, but high in those who adopt a 
westernized way of living (Eaton & Konner, 1985; Zimmet, 1979). Though it is unlikely and 
impractical to think that Māori could (or would) return to a completely traditional way of 
living, the implication is that intense exercise (as prescribed in a gym culture) may not be 
necessary for the maintenance of health in Māori, but that the everyday activities similar to a 
traditional lifestyle, such as cultivation of land, hunting, physical labour or even kapahaka, 
would suffice. It is important then that the benefits of exercise (or increased physical activity) 
are understood by Māori. 
 
Exercise physiology is the study of how our bodies’ structures and functions are altered when 
we are exposed to acute and chronic bouts of exercise (Wilmore, Costill, & Kenney, 2005). 
Literature regarding the research approach of exercise physiology is scarce however, though 
one editorial in the Journal of Physiology concluded ‘that the main regulatory and adaptive 
Chapter 3  Methodology 
 76
responses to acute and chronic exercise defy simple reductionist explanations’ (Joyner & 
Saltin, 2008). This implies that the study of exercise physiology is best approached at a 
holistic level. 
  
As type-2 diabetes is a disorder within skeletal muscle, it would be ideal to study muscle in a 
contracting state in order to better understand the effects of exercise on the physiological 
mechanisms associated with diabetes. Nevertheless, because it is not (yet) possible to study 
muscle cells directly in vivo during exercise, the responses to exercise of other systems (e.g. 
respiratory, cardiovascular systems) are observed instead. These systems, which also adjust to 
the energy requirements of contracting muscle at the onset of exercise, are related to skeletal 
muscle metabolism and are therefore an indirect way of observing the metabolic changes that 
occur in vivo. Therefore in order to understand diabetes and its relationship to muscular 
metabolism, the body must ultimately be studied as a system. 
 
Exercise physiology does follow the scientific methodology of observation, experiment and 
theory, but because it tends to be more holistic in its approach to health issues than other 
physiological sciences, it could prove a valuable tool in bridging a gap between western 
science and indigenous knowledge. Durie expresses that research and methods need to make 
sense to Māori (M. Durie, 1994), and because exercise science retains this systems approach 
and requires an understanding of the whole body response to exercise, its application is easy 
and often appeals more to those that are inexperienced in scientific research.  
 
Understanding the physiology of exercise in Māori also allows us to better understand the 
quantity and the type of exercise that could best benefit Māori (Booth, Gordon, Carlson, & 
Hamilton, 2000). Regardless of molecular or cellular explanation, the simple truth is that 
Chapter 3  Methodology 
 77
physical activity and exercise is likely to reduce the risk of diabetes in Māori similarly to non-
Māori (Burstein, Epstein, Shapiro, Charuzi, & Karnieli, 1990; Cauza et al., 2005), along with 
many other chronic diseases. While the study of exercise physiology may appear to focus on 
just one aspect of the four part health model, it is interesting that literature suggests that the 
effects of exercise are not only limited to tinana but cross to other aspects of the hauora model 
by improving mental health, whānau health and overall well being (Fox, 2007; Penedo & 
Dahn, 2005).  
 
It is my concern as a Māori exercise physiologist however, that the emphasis on the holistic 
view of health among Māori and their movement toward such multi-dimensional health 
models as Whare Tapa Wha, may lead some Māori to under-emphasize the importance of 
physical health (tinana) in Māori development. Though tinana is only one part of a multi-
dimensional view of health, it must not be neglected as less important than the other parts. If 
traditional views incorporated into everyday life are the desire for Māori, then physical 
activity and healthy living, normal parts of the ‘traditional Māori lifestyle’, must be placed 
near the top of priorities. Whether consciously for the benefit of health or not, pre-European 
Māori were physically active. In saying this it is not intended to take away from the other very 
important parts of hauora represented in such successful models as Te Whare Tapa Wha, but 
to emphasize the importance of re-introducing, rather than ‘educating’, Māori to the 
physically active lifestyles their bodies are genetically designed for (Booth, Gordon, Carlson, 
& Hamilton, 2000; Stannard & Johnson, 2006). Exercise physiology will allow Māori to 
better understand exercise and how it relates to their wellbeing, and could therefore prove a 
valuable tool toward achieving hauora. The idea that exercise and eating healthy is a Western 
view, introduced by Pakeha, is false. However the problem may be in the way the ‘physically 
Chapter 3  Methodology 
 78
active lifestyle’ is presented in New Zealand, which may not appeal to Māori and cause the 
message to appear as ‘Western medicine’.  
 
3.5 What is happening in New Zealand? 
In their 2000 world health report, the WHO made the statement that one of three principal 
goals for health-care systems should be ‘responsiveness to the expectations of the population’ 
(Smylie, Anderson, Ratima, Crengle, & Anderson, 2006). In the past the New Zealand health 
system, dominated by European influence (HWAC, 2006), has not been ‘responsive to the 
expectations’ of Māori, as their views and cultural concepts toward health have not been 
included. Recently however, the New Zealand government has attempted to better meet 
expectations of its Māori population with strategies such as He Korowai Oranga and 
Whakatātaka  which were developed by the Ministry of Health  as part of a framework for the 
public health sector to follow in its support of Māori communities and whānau. The main goal 
of He Korowai Oranga is achieving ‘whānau ora’; the realisation of maximum health and 
well-being among Māori families and communities. These strategies acknowledge the 
importance of a close relationship between the government and Māori at all levels i.e. Iwi 
(tribe), hapu (sub-tribes), community and whānau (family) levels etc. An underlying theme 
that runs through all of these government strategies is that Māori need to be involved in all 
aspects of Māori health and that the incorporation of a Māori approach is necessary for 
successful implementation. 
 
3.6 Implications – The Interface 
Throughout the world much has been said regarding the incorporation of indigenous values 
into national health systems. During a national Māori health conference held in 1984, Hui 
Chapter 3  Methodology 
 79
Whakaoranga, Māori recommended that ‘health and educational institutions recognise culture 
as a positive resource’(Hui Whakaoranga, 1984). Harris & Mercier also comment that 
learning from more than one knowledge system can only increase knowledge and help people 
make better-informed decisions in a multi-cultural world (Harris & Mercier, 2006). Just as the 
scientific world can benefit from the addition of Māori knowledge and views, Māori can 
benefit from scientific knowledge and methods of research. Māori researchers agree that 
kaupapa Māori research is not against the use of scientific methods (Cram, Smith, & 
Johnstone, 2003; Mason Durie, 2004), but are concerned about cultural sensitivity and 
whether Māori benefit directly from the research they’re involved in (Bishop, 1998). In order 
for Māori to benefit from science however, Māori must first be able to trust in researchers and 
scientific research systems. This will be possible with more Western-trained Māori scientists 
and an understanding and acceptance of Māori values among researchers and research 
institutions (Harris & Mercier, 2006). 
 
Focusing research on the prevention of diabetes, rather than treatment, is the key to ridding 
Māori of the burden of diabetes and other obesity related chronic illnesses. When related to 
health, many physiologists take a reductionist approach, racing to understand the molecular 
mechanisms of diabetes in search of the ‘diabetes pill’, and therefore aiming at the 
secondary/tertiary treatment of the disorder. Exercise physiology on the other hand, is focused 
on primary prevention (preventing obesity, insulin resistance, chronic illness etc.) and the 
response of the whole body system to exercise. While pharmaceutical interventions have 
proven effective in treating insulin resistance, it is often at the detriment of other bodily 
systems (Doggrell, 2006; Stannard & Johnson, 2006). In contrast, exercise improves insulin 
sensitivity better than pharmaceutical intervention and affects other body systems 
(cardiovascular, respiratory, neural systems) in a positive way (Stannard & Johnson, 2006). 
Chapter 3  Methodology 
 80
Similarly, traditional Māori health practices focused on primary prevention. Although 
procedures existed for the treatment of disease, the concept of ‘tapu’ which applied to all 
aspects of Māori life, was preventative by nature, acting as a type of illness and accident 
prevention (M. Durie, 1994). Not only is it more humane to avoid the suffering of diabetes but 
it is also much more cost effective in terms of the economic burden on health care (Booth, 
Gordon, Carlson, & Hamilton, 2000). 
 
3.7 My Research 
The aim of my research is not only to better understand how insulin resistance leads to 
diabetes in Māori, but to develop a suitable methodology for research at the interface of 
science and Māori health and to help Māori develop trust in scientific research techniques. As 
discussed earlier, exercise physiology and Māori health are focused on primary prevention, 
which is the main purpose for studying a cohort of non-diabetic Māori men that are clinically 
and self-identified as healthy. 
 
Māori are under-represented as scientific researchers (Harris & Mercier, 2006) as well as 
participants in scientific research, but one who observes my insulin resistance study is able to 
witness a rather rare scene of Māori participants participating as part of a whānau-like group 
in exercise physiology testing procedures, conducted by, and under, the direction of a team of 
Māori researchers. Much like the approach taken in public health studies, this research 
involves larger cohorts than most exercise physiology studies with between 24 and 50 
subjects in each study. 
  
Because the achievement of hauora is the ultimate goal of Māori research, there also needs to 
be changes when measuring the outcome of interventions and research to observe overall 
Chapter 3  Methodology 
 81
wellbeing rather than just the absence of illness (Mason Durie, 2004). I hope to apply this to 
my research by including questionnaires and group-interviews based on Māori perspectives of 
health, as part of the research process. By doing this, I will be able to gain an understanding of 
how participants feel about exercise and the methods of research that were used in my 
research, and the effect of these methods on wairua, hinengaro, tinana and whānau; similar to 
principles of a framework developed by Māori health researchers to measure mental health 
outcomes (M. H. Durie & Kingi, 1998). 
 
Research also suggests that data collected on Māori and other indigenous groups is taken but 
often not compiled and fed back to the groups or communities from which they came (Smylie, 
Anderson, Ratima, Crengle, & Anderson, 2006). In my research however, I have compiled 
results of findings and reported back to the individuals and the group as a priority activity, 
explaining the implications of my research findings and how it may affect the individual 
participant as well as Māori collectively. 
 
3.8 Limitations of the methodology 
As this approach is an innovation it suffers the same limitations as all innovations: 
a) There are no examples to follow, b) there may be incompatible principles/methods which 
need to be reconciled, and c) It may or may not be relevant to participants of other ethnic 
groups/cultures 
 
Also, as Māori are usually represented at a lower socioeconomic status and health status, it 
can often be difficult to recruit Māori participants that show a true representation of the NZ 
Māori community. Many of the Māori who participated in this research came from friend, 
Chapter 3  Methodology 
 82
church, employment or university networks so are likely more healthy, better educated and 
better employed than average NZ Māori.  
 
What’s more, ethnicity is self-identified in the cohorts I studied. As discussed previously (see 
Section 2.7.1), ethnicity is more closely associated with socio-cultural identity than genetic 
make-up, making the interpretation of ethnicity data difficult when observing physiological 
characteristics. Thus, those who identified as Māori in my studies did not necessarily share 
genetic traits. Nevertheless, this problem is not unique to my studies, and discussion regarding 
the issue of ethnicity data in scientific research, is abundant in current literature.   
 
3.9 Discussion 
The paradox with Western science is that while it discovers treatment and technologies to 
better understand and conquer diabetes, the same advances in technology have led to the 
sedentary lifestyle that supports the condition. The purpose of this paper is not to dispute the 
value of scientific research which obviously speaks for itself, nor argue that a Māori approach 
to research is more effective, but to present the possibility of using both methods of research 
to benefit Māori and non-Māori alike. As indigenous people, it is our right to have Māori 
views used in the research process and in the application of findings into intervention, 
especially when we are hardest hit by the research topic. Although Māori are becoming more 
involved in Māori-focused research, the ultimate goal in this sense should be Māori-driven 
scientific research that is developed by Māori for Māori, using a combination of scientific and 
Māori methods while maintaining cultural integrity. The scientific idea that indigenous 
knowledge is only applicable to pre-colonization times is a barrier that will only hinder health 
research. The New Zealand health system, which is burdened by ‘unhealthy Māori’, would 
also benefit greatly by investing more into improved Māori wellbeing. 
Chapter 3  Methodology 
 83
Research in the past has been focused on the secondary and tertiary treatment of diabetes, 
whereas a shift in focus is needed toward primary prevention. As habits are formed early in 
life, these interventions should appeal to rangatahi (young people) and tamariki (children). 
The editors of The New England Journal of Medicine wrote “We need to do a better job of 
educating people about healthful diets…Encouraging lifelong, regular exercise in children 
may well have the greatest effect in terms of preventing obesity”(Kassirer & Angell, 1998).  
On the other hand, while interventions such as “healthy eating – healthy action” (also known 
as Oranga kai- Oranga Pumau or HEHA)(Ministry of Health, 2007) have focused on 
‘educating’ Māori about key health issues such as the need for exercise and a balanced diet, it 
is ignorant to believe that Māori are not aware of the dangers of a sedentary lifestyle and bad 
diet already. These well planned strategies with their high funded research and interventions 
may yield the greatest success among Māori when the approach is made culturally appropriate 
for Māori.  
 
The notion of applying Māori knowledge and views to current health research, alongside a 
physiological approach, will only benefit Māori and scientific knowledge equally. Durie puts 
it simply that exploring the interface of Māori views and science allows us to shift the focus 
from “proving the superiority of one system over another to identifying opportunities for 
combining both”(Mason Durie, 2004). While we are at the exploratory stages of the interface 
between exercise physiology and Māori health research, the combination of these two 
methods could become a powerful tool in the future, to better understand the problems facing 
Māori and allow the development of more effective interventions. 
Chapter 4  Hypotheses & Research Design 
 84
4 Hypotheses and Research Design 
4.1 Research Design 
The following five studies were chosen to investigate three separate physiological factors, 
each of which have been shown to be associated with obesity and insulin sensitivity: Aerobic 
capacity (fitness), body composition (fatness), and skeletal muscle fibre-type proportion 
(fibre-type). With regard to Māori, no published studies have investigated fibre type 
proportion and only one study has investigated aerobic capacity. As Māori are worse affected 
by obesity and its co-morbidities, these studies were chosen to better understand how the 
aforementioned physiological factors might contribute to the development of insulin 
resistance and diabetes in Māori.  
 
 
 
 
 
 
Study 1 and Study 2 specifically investigate the relationship between aerobic capacity, body 
composition and markers of insulin sensitivity, so as to better understand the interaction 
between fitness, adiposity and insulin sensitivity. In Study 1, aerobic capacity, adiposity and 
markers of insulin sensitivity, were re-tested in a Māori cohort from whom data was collected 
in a previous study. This was done to confirm the findings of that previous work by observing 
whether those relationships remained constant over time despite small changes in the 
independent variables. Study 2 included a cohort of non-Māori men, the purpose of which was 
to observe whether the aforementioned relationships differed between Māori and non-Māori.  
Study 1 
Aerobic fitness and adiposity as 
predictors of fasting insulin 
concentration in healthy Māori 
men. Is the relationship consistent 
over time? 
Study 2 
Aerobic capacity and adiposity as 
predictors of fasting insulin 
concentration: A comparison 
between glucose tolerant Māori 
and non-Māori men 
Chapter 4  Hypotheses & Research Design 
 85
 
 
 
 
 
 
Following on from the investigation of body composition and insulin sensitivity, Study 3 
sought to validate the use of BIA as a suitable method for body composition assessment in 
Māori. BMI is the most frequently used method of assessment in population studies and 
clinical assessment, but ‘cut-off points’ do not take into account ethnicity-related differences 
in body composition. On the other hand, the most accurate means of measuring body fatness, 
such as hydrodensitometry, requires expensive equipment, trained technicians and is often 
impractical for those being measured. Thus, if BIA, which is now relatively cheap and easy to 
administer, proves valid in Māori it could provide a simple means of measuring body fatness 
in the laboratory and field. 
 
 
 
 
 
 
In addition to the focus of fitness and fatness as physiological determinants of insulin 
sensitivity, Study 4 and Study 5 were aimed at understanding the relationship between insulin 
sensitivity and another physiological determinant: muscle fibre type. Because measurement of 
fibre type proportion requires muscle biopsies, Study 4 was chosen to investigate the 
Study 3 
Bioelectrical impedance: A valid 
measure of body composition in 
Māori when compared to 
hydrodensitometry 
 
Study 4 
The repeatability of percutaneous 
electrically stimulated twitch 
characteristics in healthy men 
 
Study 5 
Measuring muscle contractile 
properties to predict insulin 
sensitivity in healthy men 
 
Chapter 4  Hypotheses & Research Design 
 86
feasibility of employing an innovative, non-invasive technique, developed to measuring 
muscle contractile properties (and therefore muscle composition indirectly). Study 5 then 
focused on whether the contractile properties measured with this technique, could predict 
insulin sensitivity. 
Chapter 4  Hypotheses & Research Design 
 87
4.2 Hypotheses 
Study 1: The relationship between aerobic capacity, body fatness and blood markers of 
insulin sensitivity in a group of healthy Māori men will remain, even if variables such as body 
fatness and aerobic capacity change over time. 
 
Study 2: The relationship between aerobic capacity, body fatness and blood markers of 
insulin sensitivity are similar in a group of healthy Māori and non-Māori men. 
 
Study 3: Measures of body composition obtained with a bioelectrical impedance apparatus 
correlate well with those obtained by hydrodensitometry, in healthy Māori and non-Māori 
men. 
 
Study 4: Measuring time to peak force (and half peak force) of an electrically evoked twitch 
in the vastus lateralis, will produce repeatable results as well as variance between subjects. 
 
Study 5: Twitch characteristics obtained with the method previously described, will 
accurately predict insulin sensitivity in a cohort of healthy men. 
 
Chapter 5  Study 1 
 88
5 Study 1: Aerobic fitness and adiposity as 
predictors of fasting insulin concentration in healthy 
Māori men. Is the relationship consistent over time?   
5.1 Introduction 
Obesity, as a result of excessive energy intake and a sedentary lifestyle, is primarily 
responsible for accumulation of fatty acids and triglycerides within myocytes (muscle cells). 
This accumulation is associated with impaired insulin signalling within these cells; a disorder 
known as insulin resistance. Insulin-mediated glucose uptake from the blood is reduced due to 
this insulin resistance, and eventually contributes to the hyperglycaemia which defines 
diabetes (Goodpaster, Thaete, Simoneau, & Kelley, 1997; Phillips et al., 1996).   
 
A single bout of exercise on the other hand, improves insulin action for at least 48 hours post-
exercise (Mikines, Sonne, Farrell, Tronier, & Galbo, 1988). Although the mechanisms as to 
why this occurs are not completely understood, it is widely known that the muscle’s fuel 
requirement increases during contractile activity (exercise), increasing the rate of fat oxidation 
and thereby reducing intramyocellular triglycerides (IMTG) (N. A. Johnson et al., 2003), 
which are closely associated with the insulin signalling disorder (Goodpaster, Theriault, 
Watkins, & Kelley, 2000; Greco et al., 2002; Koopman et al., 2006; Luc J. C. van Loon, 
2004).  The maximal rate of fat oxidation within muscle cells is limited by the mitochondrial 
content of the cell (J. O. Holloszy, 1967; J O  Holloszy & Coyle, 1984).   
 
Chronic contractile activity (exercise training) not only decreases body fat, which 
independently improves insulin sensitivity (Golay, Felber, Dusmet, Gomez, & Curchod, 
1985), but increases the body’s ability to oxidise lipid such as IMTG, improving insulin 
Chapter 5  Study 1 
 89
signalling independently of weight loss (DeFronzo, Sherwin, & Kraemer, 1987; Dengel, 
Pratley, Hagberg, Rogus, & Goldberg, 1996; Duncan et al., 2003). This improvement in 
muscle oxidative capacity, is largely due to an increase in mitochondrial content in response 
to chronic muscular contraction (J O  Holloszy & Coyle, 1984).   
 
Surprisingly, endurance trained athletes also have an elevated concentration of IMTG, yet are 
very insulin sensitive (Goodpaster, He, Watkins, & Kelley, 2001; Morgan, Short, & Cobb, 
1969; Schrauwen-Hinderling et al., 2003). However, acute exercise bouts along with 
increased muscle oxidative capacity within endurance trained individuals, results in a rapid 
turnover of these fats and probably ensures insulin sensitivity (Goodpaster, He, Watkins, & 
Kelley, 2001). Thus, it appears that insulin sensitivity is more a function of the rate of fat 
oxidation within myocytes, rather than the concentration of fat per se.  
 
It has been previously shown that in a group of young, apparently healthy Māori men, 
variations in fasting insulin concentrations and insulin sensitivity (HOMA-IR) can be partially 
accounted for by variations in both whole body adiposity and aerobic fitness (Stannard, 
Holdaway, Sachinwalla, & Cunningham, 2007).  Whilst maximal aerobic fitness (VO2max) 
may be limited by oxygen delivery to the working muscles (cardiac output) (González-Alonso 
& Calbet, 2003), a high VO2max also requires high oxidative capacity of the working muscle 
to utilize that oxygen (Bassett Jr & Howley, 2000).   
 
Although a number of studies have investigated aerobic fitness and its relationship to insulin 
sensitivity, almost all of these have been done with Caucasian subjects, and have measured 
oxidative capacity using invasive biopsy procedures (C. R. Bruce et al., 2003; David E. 
Kelley, Goodpaster, Wing, & Simoneau, 1999). Only the previously mentioned study by 
Chapter 5  Study 1 
 90
Stannard et al. has investigated this relationship in Māori, and they used a non-invasive 
method to measure the aerobic fitness of the working muscle (VO2maxLBM). Because this 
study was the first of its kind within Māori, and its results differed somewhat from the results 
of others using similar testing methods in Caucasian men (Coon, Bleecker, Drinkwater, 
Meyers, & Goldberg, 1989), it is important that further studies be conducted to confirm these 
findings. One method of confirming these relationships is to retest the same study participants 
after a period of time to see if the same associations exist. This proves particularly powerful if 
there are changes in one related variable such as fitness or adiposity and the other related 
variable also changes accordingly.   
 
Thus, the purposes of this study was to investigate whether the relationships observed 
between %BF, aerobic capacity and blood markers of insulin sensitivity in a cohort of Māori 
men, would exist over time after changes in fitness and body composition had occurred. 
 
Chapter 5  Study 1 
 91
5.2 Methods 
5.2.1 Study Design  
Three years prior to this study, 24 ‘healthy’ Māori men participated in a study to investigate 
the relationship between aerobic capacity (as measured by VO2maxLBM), body composition 
and markers of insulin sensitivity (Stannard, Holdaway, Sachinwalla, & Cunningham, 2007). 
The original study was the first of its kind using a cohort of young Māori men and produced 
results that differed to previous studies using only persons of Caucasian heritage.  It was the 
intention of this study to follow-up with these same 24 Māori men and to observe whether the 
relationships established in the first study were consistent over time when variables such as 
body weight, body fat %, and aerobic capacity (VO2maxLBM) had changed. 
 
5.2.2 Participants 
24 Māori men, aged 31 ± 6 (mean ± SD), who were part of a study investigating these 
relationships 2-3 years earlier were contacted to be re-tested. Only 12 of the 24 participants 
from the first study responded and agreed to be tested in the repeated study. All participants 
were originally recruited from Māori service providers and a sports group from Whanganui, 
Manawatu and Wellington. All were of varying physical activities though none were ‘well-
trained’. 
 
Chapter 5  Study 1 
 92
All participants completed a health screening questionnaire1 and none reported diagnosed 
diabetes or cardiovascular disease, though the results of one participant suggested insulin 
resistance in the previous (Sydney) study.  
  
Table 5-1 – Participant physical characteristics 
Name Age (yrs) Height (m) Weight (kg) BMI (kg/m2) 
LB 22 1.74 70.50 23.4 
RW 23 1.82 88.20 26.6 
TD 22 1.89 99.50 27.9 
SW 25 1.81 94.50 28.8 
NB 32 1.78 75.00 23.7 
RC 32 1.82 90.00 27.2 
CH 37 1.80 102.70 31.7 
BC 36 1.83 91.40 27.3 
JH 35 1.77 79.10 25.2 
TK 38 1.74 94.70 31.3 
JG 32 1.79 85.50 26.7 
JR 27 1.82 152.70 46.1 
Mean 31 1.80 93.02 28.6 
Standard Deviation 6.0 0.04 20.9 6.0 
                                                 
1
 The questionnaire used was based upon the Physical Activity Readiness Questionnaire (PAR-Q) which originates from the 
British Columbia Dept of Health (Canada), as revised by Thomas et al. (S. Thomas, Reading, & Shephard, 1992) and 
Cardinal et al. (Cardinal, Esters, & Cardinal, 1996) and with additional requirements of the Massey University Human Ethics 
Committee. 
Chapter 5  Study 1 
 93
5.2.3 Protocol 
Each participant began a water-only fast in the evening after eating dinner.  In the morning 
(after at least 10 hours fasted) participants underwent an oral glucose tolerance test (OGTT) in 
order to establish fasting insulin/glucose levels and insulin/glucose levels 2 hours after a 
glucose load.  After a light breakfast following the OGTT, body composition was measured 
by hydrodensitometry.  Participants then performed a treadmill test to establish their VO2max.  
 
In following with a kaupapa Māori approach, I aimed to promote whanaungatanga among the 
research participants, a concept associated with building relationships and finding common 
ground, so testing sessions were done in groups of 3-4 men, rather than individually. 
Participants also shared meals, before and after testing, in order to build relationships between 
researchers and participants. In most instances, karakia (prayers) were also offered before 
testing sessions. At the conclusion of the study, all participants received a compiled report of 
their results as well as the mean results of the group and population norms. 
 
5.2.3.1  Blood biochemistry 
After an overnight fast (10-12 hours post-prandial), participants underwent a standard oral 
glucose tolerance test (OGTT) at an accredited blood pathology lab (Aotea Pathology Ltd, 
Wellington). Fasting blood samples were collected for analysis of fasting glucose (FastGluc) 
and fasting insulin (FastIns) before a 75ml glucose beverage was given. Two hours after the 
glucose load was taken, blood samples were again collected for analysis of glucose and 
insulin, 2 hours post-glucose load (2hrGluc & 2hrIns).  
 
Analysis of blood was also performed at these accredited laboratories where serum was 
extracted and frozen for analysis. Serum Glucose concentrations were measured using a 
Chapter 5  Study 1 
 94
Roche 917 autoanalyser with Gluco-quant® kits (Roche Diagnostics, Manheim, Germany). 
Serum insulin was measured with a Roche Elecsys® 2010 via an immunometrical method 
using electrochemical illuminescence. Blood biochemistry followed the same methods used 
by Stannard et al. (Stannard, Holdaway, Sachinwalla, & Cunningham, 2007). Extraction and 
analysis of blood was performed at the same lab in Wellington for both arms of the study. 
 
Insulin sensitivity was estimated using the homeostasis model assessment-insulin resistance 
(HOMA-IR) model (Matthews et al., 1985).   
 
5.2.3.2  Maximal oxygen uptake 
Respiratory exchange of participants was measured during an incremental exercise protocol 
on a WOODWAY® treadmill (the Desmo ‘05’, Wisconsin, USA). The Bruce protocol (R. A. 
Bruce, 1971) was chosen as it was used in the initial study and was practical for those that 
were unfamiliar with treadmill running or were relatively unfit. Participants were encouraged 
to continue exercising for as long as possible, as the speed and gradient of the treadmill 
increased every 3 minutes, in order to obtain intensities capable of eliciting maximal VO2.  
Volitional fatigue was defined as when the subject grasped the safety railing of the treadmill 
to finish the treadmill test.  
 
Expired respiratory gases were sampled throughout the entire exercise test by an online gas 
analysis system using Moxus (version 2.1.05) software (AEI technologies Ltd, Illinois, USA). 
Expired air was collected via a Hans Rudolf mouthpiece with a turbine attached to measure 
expired volume. Expired air entered a mixing chamber and was analyzed for expired fraction 
of CO2 using an infrared carbon dioxide sensor and expired oxygen using a zirconium cell-
based oxygen sensor (AEI, Illinois, USA).  
Chapter 5  Study 1 
 95
Both the analysers and software were calibrated each morning before testing was performed 
using a predetermined mixture of O2 and CO2 (4.9% CO2, 15.6% O2 in Nitrogen) while the 
turbine was calibrated with a 3000ml Hans Rudolf calibration syringe.  Data were averaged 
by the software and displayed in 10 second intervals. VO2max was defined as the highest 
rolling 1 minute average VO2 (six 10 second recordings) obtained during the test. VO2max 
was expressed relative to lean body mass (VO2maxLBM) rather than relative to whole-body 
mass as it better represents the aerobic capacity of working muscle, independent of fatness 
(Coon, Bleecker, Drinkwater, Meyers, & Goldberg, 1989; Stannard, Holdaway, Sachinwalla, 
& Cunningham, 2007). 
 
5.2.3.3  Body composition 
Hydrodensitometry was performed in the morning between 1-2 hours after a small meal (the 
first meal post-fast) for determination of body volume. Participants wore tight fitting shorts or 
a swimming costume while in the water tank. Body weight on land was measured using a 
Sartorius electric scale (EA Model, Goettingen, Germany) and height was measured using a 
wall mounted ‘Harpenden’ stadiometer (Holtain Ltd, Crosswell, UK).   
 
Participants were given up to 10 minutes to practice the submersion technique and familiarize 
themselves with the procedure. They were suspended in the water by a harness-like seat 
attached by ropes to a force transducer sited above the tank.  The transducer output, once 
calibrated would be amplified and converted to a digital signal and used to indicate the 
underwater weight of the subjects. Once familiar with what was required, subjects were asked 
to exhale as much air from their lungs as was comfortable while submerging completely 
underwater by bowing the head forward as far as was comfortable. Once the subject was fully 
submerged and relatively stable, the measurement of underwater weight began using Labview 
Chapter 5  Study 1 
 96
software (Eagle Scientific, USA) programmed by our group. Participants were required to 
stay submerged for 5 seconds as underwater weight was measured. The software provided a 
graph of underwater weight, during the 5 second period. An average of the 5 second reading 
was taken, in order to account for the sinusoidal fluctuations in transducer output that 
occurred from the natural motion of the water.  
 
After the 5 second underwater period, participants were given a signal from the operator to 
raise their head out of the water. After participants had raised their head out of the water, 
residual lung volume was measured via oxygen dilution using an oxygen re-breathing 
technique described by van der Ploeg et al. (van der Ploeg, Gunn, Withers, Modra, & 
Crockett, 2000). The re-breathing apparatus consisted of a 5L (rubber) anaesthetic bag 
(Vacumed, California, USA) containing 4.25L of pure oxygen attached to Hans Rudolf 3 way 
‘sliding type TM’ Manual directional control valve (2870 series, Missouri, USA). A disposable 
‘SureGard’ respiratory filter was used as a mouthpiece for participants, and the control valve 
allowed the operator to open or close the flow of oxygen from the bag to the mouthpiece.  
 
After raising their head out of the water after the weighing, the mouthpiece of the apparatus 
was guided into their mouth and participants were asked not to take their first breath until the 
mouthpiece was securely in the mouth, the lips sealed around it, and the valve opened that 
allowed the participant to breathe the closed circuit oxygen. After 5-6 breaths in the 
apparatus, the valve was closed and the gas analysed for O2 and CO2 concentration using the 
same gas analysers used in the exercise test above. Body fat percentage was then calculated 
using previously established methods (Brozek, Grande, Anderson, & Keys, 1963).  
 
Chapter 5  Study 1 
 97
Each participant performed the measurement protocol at least 3 times. Percentage body fat 
was taken as the average of the 3 measurements. If deviation outside 2% body fat occurred, a 
fourth measurement was also taken, and measurements that deviated outside 2% body fat 
were disregarded. 
 
Two point calibration of the load cell was performed (once with no weight and again with a 
5kg weight) each morning before testing took place.   
 
5.2.4 Statistical Analyses 
Paired t-tests were used to identify any significant changes in the variables measured, between 
the first test (Sydney) and the second test (Wellington). Simultaneous multiple regression was 
performed to ascertain whether the dependant variables relating to insulin sensitivity (HOMA-
IR, Fast-Ins, 2hr-Ins, Fast-Gluc, 2hr-Gluc), could be significantly predicted by the 
independent variables (%BF, VO2maxLBM). Testing date was also included as an 
independent variable, to observe the effect of time on these relationships. SPSS version 10 
was used for all statistical analyses, while graphs were made using Microsoft excel 2003. 
Chapter 5  Study 1 
 98
5.3 Results 
It was my initial intention to re-test all of the 24 Māori men that participated in the initial 
study (Sydney study).  However, only 12 of the participants from the Sydney study were 
available to take part in the repeated test (Wellington study). Of the original 24 participants, 
six had moved out of the country, five were unable to be contacted, one was not interested in 
participating and one could not participate due to injury.  
 
The results of the 12 participants in this study will only be compared to the results of these 
same participants in the Sydney study. Because of this, the results from the Sydney study of 
the 12 participants referred to in this text, may vary from those described in the original study 
(Stannard, Holdaway, Sachinwalla, & Cunningham, 2007), which included all 24 participants. 
 
Tables of individual results for this study are found in the appendix (Appendix A – Study 1). 
 
5.3.1 Body composition and aerobic capacity 
Mean body weight increased from 92.3kg ± 20.7 (mean ± SD) in the initial study to 93.7kg ± 
20.9 in the repeated study, and %BF decreased from 22.4% ± 9.2 to 20.9% ± 8.5, although 
neither of these changes were statistically significant (p>0.05)(Table 1.2). Of all the variables 
measured, only LBM, which increased from an average of 70.3kg ± 8.7 to 71.9kg ± 10.0, and 
VO2maxLBM, which decreased from 59.4 ± 6.4 ml/kg l.b.m/min to 50.3 ± 2.1 ml/kg 
l.b.m/min, changed significantly (p<0.05) between the two testing periods.  Only one 
participant had an increase in VO2maxLBM between testing sessions (from 50.08 to 60.84).  
 
Chapter 5  Study 1 
 99
Table 5-2 – Body composition and aerobic capacity 
 
Sydney 
Study 
 
Wellington 
Study 
 
Paired  
t-Test 
Mean SD Mean SD p 
Weight (kg) 92.3 20.7 93.7 20.9 0.535 
BMI (kg/m2) 28.5 6.1 28.8 6.0 0.595 
%BF 22.4 9.2 20.9 8.5 0.223 
LBM (kg) 70.3 8.7 71.9 10.0 0.019 
VO2max LBM  
(ml.kgLBM -1.min-1) 
59.4 6.4 50.3 7.4 0.002 
 
 
Figure 3 – Group mean of lean body mass (LBM) in Sydney and Wellington study  
Sydney Wellington
50.0
55.0
60.0
65.0
70.0
75.0
80.0
85.0
M
ea
n
 
LB
M
 
(kg
)
 
 
Chapter 5  Study 1 
 100
Figure 4 – Group mean for VO2maxLBM in Sydney and Wellington study  
Sydney Wellington
0.00
10.00
20.00
30.00
40.00
50.00
60.00
70.00
VO
2m
ax
LB
M
 
(m
l.k
gL
BM
 
-
1 .
m
in
-
1 )
 
5.3.2 Blood biochemistry 
HOMA-IR decreased along with fasting insulin from 75.2 ± 64.9 to 58.3 ± 46.6 and fasting 
glucose from 5.1 ± 0.5 to 4.9 ± 0.3. While 2hrIns increased from 117.3 ± 76.8 to 178.1 ± 
226.7 along with 2hrGluc from 4.1 ± 0.9 to 4.3 ± 1.1. None of the changes in blood 
biochemistry were statistically significant (p>0.05).  
  
Chapter 5  Study 1 
 101
 
Table 5-3 – Blood Biochemistry within group 
 
Sydney Study 
 
Wellington Study 
 
Paired t-Test 
Mean SD Mean SD p 
HOMA-IR 2.5 2.4 1.8 1.5 0.055 
FastIns (pmol.L-1 ) 75.2 18.7 58.3 46.6 0.068 
FastGluc (mmol.L-1 ) 5.1 0.5 4.9 0.3 0.111 
2hrIns (pmol.L-1 ) 117.3 76.8 178.1 226.7 0.395 
2hrGluc (mmol.L-1 ) 4.1 0.9 4.3 1.1 0.593 
 
 
Figure 5 – Group mean for HOMA-IR in Sydney and Wellington study 
Sydney Wellington
0.00
1.00
2.00
3.00
4.00
5.00
6.00
HO
M
A
 
Chapter 5  Study 1 
 102
5.3.3 Multiple regression analysis 
Simultaneous multiple regression analysis was performed with the results of the Wellington 
and Sydney study combined using the ‘testing date’, i.e. Wellington or Sydney, as an 
independent variable. 
 
The simultaneous multiple regression analysis showed that 84% of the variance in HOMA-IR 
could be explained by %BF (p<0.001), VO2maxLBM (p=0.013) and testing date (p=0.027) 
combined. Similarly, 86% of the variation in FastIns could be explained by %BF, 
VO2maxLBM (p<0.05) and testing date which was marginally significant (p=0.051). Sixty 
percent of FastGluc variance could be attributed to VO2maxLBM and testing date (p<0.05), 
but %BF was not significant (p>0.05). VO2maxLBM was the only significant predictor of 
2hrIns. 
 
When multiple regression analysis was performed on the two studies separately, %BF 
(p=0.004) and VO2maxLBM (p=0.041) were significant contributors to the variation in 
HOMA-IR and FastIns in the Wellington study, although VO2maxLBM was marginally 
significant (p=0.054). In the earlier Sydney study however (including only the 12 participants 
re-tested in Wellington), only %BF was a significant contributor to the variation in HOMA-IR 
and FastIns. No other significant relationship existed between the independent variables of 
%BF and VO2maxLBM, and the other dependant variables analyzed (FastGluc, 2hrIns, 
2hrGluc) (See Appendix Table 0-5)  
  
Chapter 5  Study 1 
 103
Table 5-4 – Multiple regression analysis for combined results 
Dependant variable Independent variable r r² t P 
HOMA-IR - 0.844 0.712 - <0.001 
 
% body fat - - 5.001 <0.001 
 
VO2maxLBM - - -2.712 0.013 
 
Testing date  - - -2.392 0.027 
Fasting serum insulin - 0.863 0.745 - <0.001 
 
% body fat - - 5.804 <0.001 
 
VO2maxLBM - - -2.496 0.021 
 
Testing date - - -2.077 0.051 
Fasting serum glucose - 0.602 0.362 - 0.027 
 
% body fat - - -1.732 0.099 
 
VO2maxLBM - - -2.787 0.011 
 
Testing date - - -3.095 0.006 
2hr serum insulin - 0.690 0.477 - 0.008 
 
% body fat - - 1.500 0.151 
 
VO2maxLBM - - -2.784 0.012 
 
Testing date - - -0.442 0.664 
2hr serum glucose - 0.527 0.277 - 0.970 
 
% body fat - - -1.159 0.261 
 
VO2maxLBM - - -2.601 0.018 
 
Testing date - - -0.951 0.354 
 
Chapter 5  Study 1 
 104
5.4 Discussion 
The purpose of this study was to observe the relationship between adiposity, aerobic fitness 
and markers of insulin sensitivity in a group of healthy Māori men, and investigate whether 
those relationships changed over time.  
 
Both the separated and combined results from Sydney and Wellington showed that %BF 
predicts fasting insulin levels and HOMA-IR values in young, healthy Māori men. This 
finding is consistent with studies in non-Māori, which have shown that body composition is 
associated with insulin sensitivity (Bogardus, Lillioja, Mott, Hollenbeck, & Reaven, 1985; 
Yki-Jarvinen & Koivisto, 1983).  Bogardus et al. (Bogardus, Lillioja, Mott, Hollenbeck, & 
Reaven, 1985) point out that early reports of this relationship may have been misleading due 
to the indirect methods which essentially infer body composition, such as BMI and skin-fold 
measurement, employed to assess body composition. In fact, although the association between 
body fat and insulin sensitivity is commonly assumed, the use of valid methods to measure 
body composition, when observing its relationship to insulin sensitivity, has rarely been used. 
Consequently, body composition in this study was measured using hydrodensitometry, which 
is termed the ‘gold standard’ method to measure body composition. Then again, this method 
is not practical for large scale studies due to the time and skill required to perform the test, nor 
is it appropriate for certain populations, such as the obese or elderly. For this reason, 
development of appropriate techniques to measure, rather than infer body composition, could 
prove valuable in early identification of Māori that are at risk of type-2 diabetes; before 
glucose intolerance occurs. On the other hand, because body fat distribution is more closely 
related to insulin sensitivity than body fat percentage, methods of body composition 
measurement should actually focus on quantifying abdominal/visceral fat independent of 
subcutaneous fat. In this regard, hydrodensitometry does not take into account body fat 
Chapter 5  Study 1 
 105
distribution and therefore %BF derived from this method is made up in large part by 
subcutaneous fat stores. Accordingly, a simple measure of abdominal adiposity, such as waist 
circumference, may provide a simple and more effective way of predicting diabetes risk. 
Accordingly, Bell et al. found that the combination of measuring waist circumference and 
blood triglycerides was a simple and accurate method of predicting insulin sensitivity in 
Māori (Bell, McAuley, Mann, Murphy, & Williams, 2004). For that reason, this study would 
have benefited from the inclusion of waist circumference measures. 
 
A second major finding of the present study is that the time between the first and second tests 
(a representation of ageing, albeit small) had a significant effect upon HOMA-IR, FastIns and 
FastGluc (p<0.05), revealing a relationship between ageing and insulin sensitivity in this 
cohort. Surprisingly however, the effect of age on insulin sensitivity was not a detrimental 
effect i.e. increased HOMA-IR, fasting insulin etc. In fact FastIns, FastGluc and HOMA-IR 
actually decreased, although the decrease was not significant for FastGluc (p=0.111), 
approached significance for FastIns (0.068) and was only marginally significant for HOMA-
IR (p=0.055). This is in contrast to most epidemiological and physiological studies, which 
indicate that aging  is associated with insulin resistance and diabetes (Fink, Kolterman, 
Griffin, & Olefsky, 1983; Ministry of Health, 2008; Rowe, Minaker, Pallotta, & Flier, 1983). 
On the other hand, although type-2 diabetes was originally thought to be a disorder associated 
with ageing, hence the name ‘mature/adult onset diabetes’, research now shows that there is 
no association between age and insulin sensitivity after accounting for differences in %BF and 
VO2max (Bogardus, Lillioja, Mott, Hollenbeck, & Reaven, 1985). Although my results could 
be confounded by technical error associated with collection and analysis of repeated blood 
samples, this is not likely a factor in this study as blood was collected and analyzed in the 
same location (Aotea Pathology, Wellington) for both arms of the study (See section 5.2.3.1). 
Chapter 5  Study 1 
 106
Therefore, in these participants, the improvement in insulin sensitivity with age is likely a 
result of the significant increase in lean body mass within this cohort. Because muscle is a 
major site of insulin-mediated glucose disposal in the body, an increase in muscle mass would 
improve whole-body insulin sensitivity, and thus the glyceamic response to a glucose load. 
The reduction in relative fat mass in this cohort, albeit small, might also improve insulin 
sensitivity.  
 
In these studies (study 1 and 2 of the thesis), fasting insulin/glucose, 2 hour insulin/glucose 
and HOMA-IR, have been used to represent ‘insulin sensitivity’. Because these measures do 
not give a clear understanding of the dynamic response of glucose to insulin secretion over 
time, more precise methods and models, such as the glucose/insulin clamp and frequently 
sampled glucose tolerance test, have been advocated in lab-based studies (Breda, Cavaghan, 
Toffolo, Polonsky, & Cobelli, 2001; Caumo, Bergman, & Cobelli, 2000; DeFronzo, 1979). 
While these methods are more precise measures of ‘insulin sensitivity’, the simple measures 
(fasting insulin, HOMA-IR) used in this study are currently the standard diagnostic tool for 
insulin resistance, glucose intolerance and diabetes in the clinical setting and are likely more 
acceptable for participants than methods involving regular measurement (See section 2.3.3).  
 
On another note, VO2maxLBM was also a significant predictor of HOMA-IR in the 
Wellington study but not in the Sydney study. However, when the results of all 24 participants 
from the original Sydney study are referred to (Stannard, Holdaway, Sachinwalla, & 
Cunningham, 2007), VO2maxLBM was a significant predictor of HOMA-IR in this group as 
well, signalling a lack of statistical power in the individual ‘arms’ of this repeated study. 
However, when the results of both ‘arms’ (Sydney and Wellington) were combined, and the 
year tested was analysed as an independent variable, VO2maxLBM, was a significant 
Chapter 5  Study 1 
 107
predictor of HOMA-IR, FastIns, FastGluc. This supports the findings of Bruce et al., although 
unlike my study where %BF (p<0.001) was a more significant contributor than VO2maxLBM 
(p=0.013), they showed that oxidative capacity and VO2max were even better predictors of 
insulin sensitivity than adiposity (C. R. Bruce et al., 2003). This could be due to the difference 
in methods used. Bruce et al. expressed VO2max relative to total body weight; whereas 
VO2max in my study was expressed relative to lean body mass only (VO2maxLBM).  
 
During exercise, muscle is the main tissue for energy utilization and oxygen consumption 
within the body, whereas adipose tissue does not contract and consumes relatively little 
oxygen. Therefore, expressing VO2max relative to total body weight, as is common place in 
exercise physiology, underestimates the relative ’fitness’ of the muscle. What’s more, 
considering the role of body fatness upon insulin sensitivity, including fat in the measurement 
of relative aerobic capacity could confound results. Alternatively, expressing VO2max relative 
to lean body mass alone represents, in part, the maximal oxidative capacity of the muscle. 
Thus, in this study, measurement of relative aerobic capacity was made independent of body 
fatness. The only other study that I know of which investigates these relationships, using 
VO2max relative to lean body mass (VO2maxLBM) as a measure of aerobic fitness, found 
that VO2maxLBM was not a significant predictor of insulin sensitivity at all (Coon, Bleecker, 
Drinkwater, Meyers, & Goldberg, 1989). However, Coon et al. used a much older, cohort (46-
73 years old), which likely had a very different ethnic background and much larger range of 
ages than my Māori participants. In contrast, on two separate occasions, VO2maxLBM has 
been shown to be a significant predictor of insulin sensitivity in a cohort of Māori men. These 
findings further support the possibility of using VO2max, relative to lean body mass, as a non-
invasive way of identifying Māori at risk of insulin resistance, who do not yet show impaired 
glycaemia.  
Chapter 5  Study 1 
 108
 
Another interesting finding from this study was the degree to which VO2max decreased 
between tests. Maximal oxygen consumption (VO2max) reportedly decreases with age, and 
the significant point of decline within sedentary individuals occurs when people reach their 
20s and 30s (Hawkins & Wiswell, 2003). Accordingly, in the present study aerobic capacity 
relative to lean body mass, decreased significantly between testing sessions (over 2-3 years) 
within this cohort of Māori men, all of which were in their 20s or 30s. Although one 
participant had a 21% increase in VO2maxLBM, the rest of the cohort (minus the one 
participant with the increase) had an average decline in VO2maxLBM of 18%. This result is 
quite dramatic considering that researchers propose the age-related decline in VO2max to be 
just 10% per decade (Buskirk & Hodgson, 1987; Hawkins & Wiswell, 2003). LBM within the 
group, which actually increased from the first study, was the only other variable which had a 
significant change between tests (p<0.05). Although this increase in LBM would lower 
VO2maxLBM calculation and partially explain the observed decrease in VO2maxLBM, even 
absolute VO2max decreased by 16% on average in this group (not including the participant 
who showed an increase). Taken at face value, these findings suggest that young Māori men 
may have a greater rate of decline in oxidative capacity with age when compared with other 
populations. Nevertheless, this cohort’s level of physical activity in the years between testing 
sessions was not recorded, so it is difficult to explain this marked decrease in aerobic 
capacity. Furthermore, whilst the method of gas composition analysis was the same between 
both the Sydney and Wellington arms of the study, the method of assessing gas volume was 
different. Douglas bags were used in the Sydney study and a turbine was used in the 
Wellington study. Both methods were calibrated, but it is possible that the observed reduction 
in VO2max could be explained by this technical difference.  
 
Chapter 5  Study 1 
 109
Since insulin sensitivity is positively correlated to aerobic fitness, it is interesting that 
HOMA-IR did not significantly increase as VO2maxLBM declined. However, as discussed 
previously, the increase in lean body mass and small decrease in relative fat mass likely 
counteracted any possible decrease in aerobic capacity, and ageing. There are a number of 
applications from these findings. Firstly, in contrast to common belief that cardiovascular 
exercise is the most appropriate mode of training for ‘metabolic health’, exercise such as 
resistance training which is aimed at increasing muscle mass (hypertrophy) without 
necessarily improving aerobic fitness, may be just as effective in improving and maintaining 
metabolic health in Māori. This is important considering that some may find it difficult or 
unacceptable to perform particular types of cardiovascular exercise. Secondly, in the short-
term at least, the supposedly adverse effects of ageing on insulin sensitivity can be prevented 
or even reversed with the maintenance or improvement of lean body mass.  
 
5.4.1 Chapter Summary 
This study strengthens the findings of the original study by Stannard et al. (Stannard, 
Holdaway, Sachinwalla, & Cunningham, 2007), and shows that even over time, a relationship 
exists between adiposity, aerobic fitness and insulin sensitivity in Māori men. My results have 
also shown that the detrimental effects of age and reduced aerobic fitness upon ‘metabolic 
health’ are reduced with increased muscle mass and reduced body fat. Clearly however, 
further studies of this relationship in both Māori and non-Māori would make the observations 
of this study stronger if a larger sample size can be employed. 
 
Chapter 6  Study 2 
 110
6 Study 2: Aerobic capacity and adiposity as 
predictors of fasting insulin concentration: A 
comparison between glucose tolerant Māori and 
non-Māori men.  
 
6.1 Introduction 
When compared with New Zealanders of European ancestry, Māori in New Zealand are 
disproportionately afflicted by diabetes and it’s precursor condition insulin resistance (David 
Simmons & Thompson, 2004). Physiological characteristics of an individual, such as skeletal 
muscle fibre type proportion (Lillioja et al., 1987a), aerobic capacity (C. R. Bruce et al., 
2003), and body composition (Bogardus, Lillioja, Mott, Hollenbeck, & Reaven, 1985), are all 
associated with risk of insulin resistance. It is thought that these physiological variables all 
impact on whether there is accumulation of lipid, and subsequent impairment of insulin-
mediated glucose disposal, within liver, and particularly, skeletal muscle (Seppälä-Lindroos et 
al., 2002; Stannard & Johnson, 2004). 
 
On the other hand, population-based studies have identified lower standards of employment, 
education, and living standards as contributors to an increase in diabetes prevalence (Ministry 
of Health, 2004b). When taken in isolation, the latter approach suggests that Māori, who are 
overrepresented in the lower deciles of socioeconomic status, have a lower degree of 
‘metabolic’ health because they are exposed to a lifestyle that is conducive to obesity and 
insulin resistance. Alternatively, it is possible that a physiological disposition for lipid 
Chapter 6  Study 2 
 111
accumulation and/or reduced lipid turnover within Māori predisposes this population to 
greater risk of obesity and metabolic disorder.  
 
Separating socioeconomic from physiological factors is difficult however, because any 
phenotype, including that which results in diabetes, develops from the interaction between 
inherent traits and environmental factors (Neil Pearce, Foliaki, Sporle, & Cunningham, 2004). 
Furthermore, comparing Māori and European participants in a physiological study can be 
difficult as ethnicity is a self identified trait and not necessarily a genetic or biological 
characteristic. On the other hand, because epidemiological data is also based on self-
identification of ethnic identity, it would be expected that a comparative study of 
physiological factors between two different self-identified ethnic groups would produce 
similar trends to epidemiological data. Nevertheless, an understanding of which physiological 
characteristics may be associated with increased risk of developing insulin resistance and 
diabetes may prove useful in preventing insulin resistance and prescribing treatment.   
 
Previous research has identified both aerobic capacity (VO2max) of the lean body mass, and 
body adiposity independently, as predictors of insulin resistance in young Māori men 
(Stannard, Holdaway, Sachinwalla, & Cunningham, 2007). Others have also shown the 
significance of adiposity and associated increases in blood lipids in predicting insulin 
sensitivity in Māori  (K. McAuley et al., 2001). Aerobic capacity is, in part, a function of the 
oxidative capacity of the muscle and thus also its ability to utilize (oxidise) lipids.  
Accordingly, oxidative capacity of skeletal muscle and resting whole body lipid oxidation 
rates have been shown to be significant predictors of whole body insulin sensitivity 
(Goodpaster, He, Watkins, & Kelley, 2001). 
 
Chapter 6  Study 2 
 112
Regardless of the epidemiological trends previously discussed, Māori actually appear to be 
more physically active than New Zealanders of European Origin (NZEO) (Utter, Scragg, 
Schaaf, & Fitzgerald, 2006), and sporting participation rates in Māori are high (Chadwick & 
Palmer, 2006). Furthermore, Māori have less adiposity when compared to NZEO of a similar 
BMI (Swinburn, Ley, Carmichael, & Palnk, 1999), and even at a similar %BF, Māori are still 
more insulin resistant than NZEO (K. McAuley, Williams, Mann, Goulding, & Murphy, 
2002).  These data beg the question as to whether the observed relationship between physical 
fitness, adiposity and insulin sensitivity exist in Māori and non-Māori.  Understanding these 
relationships better, may assist in shedding light upon the reasons why Māori are more 
predisposed to developing type-2 diabetes in New Zealand. To date, only one published study 
has shown that the relationship between aerobic capacity, body composition and insulin 
sensitivity exists in Māori (Stannard, Holdaway, Sachinwalla, & Cunningham, 2007). 
Additionally, I have shown that this relationship is consistent over time (see Chapter 5). 
Nevertheless, neither of these studies tried to compare these relationships between two ethnic 
groups.  
 
Thus, the purpose of this study was to investigate whether the relationships observed between 
%BF, aerobic capacity and blood markers of insulin sensitivity, differed between Māori and 
non-Māori men.   
 
Chapter 6  Study 2 
 113
6.2 Methods 
6.2.1 Study Design 
The methods used in this study, including the measurement of blood biochemistry, aerobic 
capacity and body composition, were identical to those in Study 1. However, this was a cross-
sectional study rather than a repeated measures study, including a non-Māori cohort for 
comparative purposes.  
 
6.2.2 Participants  
Forty eight healthy men (31 Māori, 17 non-Māori) aged 28 ± 5 (mean ± SD) years old. All 
participants answered a health questionnaire (see chapter 5) and had no cardiovascular, 
respiratory or metabolic disorders. All participants were randomly selected from employment, 
university or church networks. 
 
Participants were asked to self-identify their ethnicity. All those who were included as Māori 
identified as such, while in the non-Māori group all but one participant (Asian) identified as 
European. Pacific Islanders were excluded from all studies in this body of research. 
 
Chapter 6  Study 2 
 114
Table 6-1 – Participant physical characteristics 
 
Entire Group 
n = 48 
 
Māori 
n = 31 
 
Non-Māori 
n = 17 
 
Mean SD Mean SD Mean SD 
Age (yrs) 27.9 5.40 28.5 5.5 26.7 5.1 
Height (m) 1.80 0.07 1.80 0.07 1.79 0.07 
Weight (kg) 87.8 17.2 93.7 17.4 77.0 10.4 
BMI (kg/m2) 27.1 4.7 28.8 4.8 24.1 2.7 
 
6.2.3 Statistical Analyses 
Independent samples t-tests were used to identify any significant differences in the variables 
measured, between the Māori and non-Māori cohort. Simultaneous multiple regression was 
performed to ascertain whether the dependant variables relating to insulin sensitivity (HOMA-
IR, Fast-Ins, 2hr-Ins, Fast-Gluc, 2hr-Gluc), could be significantly predicted by the 
independent variables (%BF, VO2maxLBM). Ethnicity was also included as an independent 
variable, to determine the effect of ethnicity on these relationships. SPSS version 10 was used 
for all statistical analyses. 
 
6.3 Results 
Although 61 participants took part in this study (39 Māori, 22 non-Māori), I was unable to 
collect accurate VO2max measures for 13 of these subjects (8 Māori and 5 non-Māori) so data 
are reported only for 48 participants.  
 
Chapter 6  Study 2 
 115
6.3.1 Body Composition 
The Māori group had a significantly greater mean weight (93.7kg Māori, 77.0kg non-Māori), 
BMI, %BF and LBM than the non-Māori group (p<0.01). Neither age nor height were 
significantly different between groups (p>0.05).  
 
Table 6-2 – Body composition (percent body fat and total lean body mass) 
 
Entire Group Māori Non-Māori 
Mean SD Mean SD Mean SD 
%BF 19.7 7.0 21.5 7.4 16.5 5.0 
LBM (kg) 69.3 9.3 72.4 9.2 63.8 6.5 
 
6.3.2 Aerobic capacity 
The non-Māori group had a significantly greater VO2maxLBM than the Māori group (60.8 
L/min/kg l.b.m and 53.0 L/min/kg l.b.m: p<0.05). The non-Māori group also had a higher 
absolute VO2max than the Māori (3853 ml/min and 3805 ml/min), though this was not 
significant (p>0.05).   
 
Table 6-3 – Aerobic capacity (absolute and relative to lean body mass) 
 
Entire Group Māori Non-Māori 
Mean SD Mean SD Mean SD 
VO2max (ml) 3822 507 3805 505 3853 525 
VO2maxLBM 
(ml.kgLBM -1.min-1) 
 
55.8 8.6 53.0 6.9 60.8 9.3 
 
Chapter 6  Study 2 
 116
6.3.3 Blood Biochemistry 
Eight of the 48 participants did not receive 2hr insulin blood samples, and 5 did not receive 
2hr glucose samples. None of the blood markers measured were significantly different 
between the Māori and non-Māori group (p>0.05). 
 
Table 6-4 – Blood Biochemistry 
 
Entire Group Māori Non-Māori 
Mean SD n Mean SD n Mean SD n 
HOMA-IR 1.67 1.28 48 1.88 1.32 31 1.46 1.23 17 
Fast Ins (pmol.L-1 ) 55 38.7 48 60 39.3 31 47 37.2 17 
Fast Gluc (mmol.L-1 ) 4.8 0.4 48 4.8 0.5 31 4.8 0.2 17 
*2hr Ins (pmol.L-1 ) 143 141.3 40* 161 165.0 26* 110 75.5 14* 
*2hr Gluc (mmol.L-1 ) 4.3 1.1 43* 4.4 1.1 27* 4.0 1.0 16* 
* 8 participants did not have 2hr insulin samples, 5 did not have 2hr glucose 
 
6.3.4 Statistical Analysis 
The results of the simultaneous multiple regression analyses are found in Table 6-5. While 
ethnicity was used as an independent variable (along with %BF and VO2maxLBM) in the 
initial multiple regression analysis of the entire group, it was not a significant predictor of any 
of the dependant variables (p>0.05). The results of the entire group seen in Table 6-5 are 
from a simultaneous multiple regression analysis which used only %BF and VO2maxLBM as 
independent variables, ethnicity was excluded.  
 
 
  
117
Table 6-5 – Simultaneous multiple linear regression analysis 
 
Entire Group Māori non-Māori 
Dependant variable Independent 
variable 
r r² t-
value 
P-
value 
n r r² t-
value 
P-
value 
n r r² t-
value 
P-
value 
n 
HOMA-IR - 0.55
3 
0.30
6 
- <0.001 n=4
8 
0.66
6 
0.44
3 
- <0.001 n=3
1 
0.34
0 
0.11
6 
- 0.423 n=1
7 
% body fat - - 2.902 0.006 - - - 3.465 0.002 - - - 0.396 0.698 - 
VO2maxLBM - - -3.172 0.003 - - - -3.222 0.003 - - - -1.354 0.197 - 
Fasting serum 
insulin 
- 0.57
9 
0.33
6 
- <0.001 n=4
8 
0.69
4 
0.48
1 
- <0.001 n=3
1 
0.35
7 
0.12
7 
- 0.386 n=1
7 
% body fat - - 3.317 0.002 - - - 4.018 <0.001 - - - 0.435 0.670 - 
VO2maxLBM - - -3.194 0.003 - - - -3.151 0.004 - - - -1.428 0.175 - 
Fasting serum 
glucose 
- 0.27
1 
0.07
4 
- 0.179 n=4
8 
0.37
3 
0.13
9 
- 0.123 n=3
1 
0.20
9 
0.04
4 
- 0.732 n=1
7 
% body fat - - -1.189 0.241 - - - -1.174 0.250 - - - 0.548 0.592 - 
VO2maxLBM - - -1.547 0.129 - - - -1.771 0.087 - - - -0.721 0.483 - 
2hr serum insulin - 0.56
0 
0.31
4 
- 0.001 n=4
0 
0.67
3 
0.45
2 
- 0.001 n=2
6 
0.42
2 
0.17
8 
- 0.340 n=1
4 
% body fat - - 2.436 0.020 - - - 2.827 0.010 - - - -0.814 0.433 - 
VO2maxLBM - - -3.141 0.003 - - - -3.285 0.003 - - - -1.049 0.317 - 
2hr serum glucose - 0.26
8 
0.07
2 
- 0.225 n=4
3 
0.34
5 
0.11
9 
- 0.219 n=2
7 
0.27
6 
0.07
6 
- 0.597 n=1
6 
% body fat - - 0.946 0.350 - - - 1.177 0.251 - - - -1.023 0.325 - 
VO2maxLBM - - -1.398 0.170 - - - -1.293 0.208 - - - 0.166 0.871 - 
Chapter 6  Study 2 
 118
Multiple regression analysis showed that both %BF and VO2maxLBM were significant 
predictors of HOMA-IR, Fast-Ins and 2hr-Ins in the entire group as well as the Māori cohort. 
%BF and VO2maxLBM were not significant predictors of any of the dependant variables 
(HOMA-IR, Fast-Ins, Fast-Gluc, 2hr-Ins, 2hr-Gluc) in the non-Māori cohort however. 
 
6.4 Discussion 
Population studies have shown that Māori are more obese and insulin resistant than NZEO (K. 
A. McAuley et al., 2002; Ministry of Health, 2008). Since Māori are also overrepresented in 
lower socioeconomic levels, it is believed that the health disparity between Māori and NZEO 
is due to environmental factors rather than a physiological disposition. Accordingly, Pearce et 
al. suggest that historical, cultural and socioeconomic factors, which influence lifestyle and 
access to health care, are responsible for ethnic differences in health (Neil Pearce, Foliaki, 
Sporle, & Cunningham, 2004). On the other hand, it appears that Māori are more insulin 
resistant than NZEO even after adiposity is adjusted for (K. McAuley, Williams, Mann, 
Goulding, & Murphy, 2002), suggesting an inherent disposition toward insulin resistance. 
While differences in insulin sensitivity and obesity between Māori and NZEO are clear from 
population studies, almost no studies have investigated how ethnicity may alter the 
relationship between the physiological factors related to such metabolic disorders. Thus, the 
purpose of this study was to investigate whether the relationship between aerobic capacity, 
body composition and markers of insulin sensitivity differed between Māori and non-Māori. 
 
We were able to show that VO2maxLBM and %BF independent of each other are both 
significant predictors of blood markers associated with insulin sensitivity in a healthy group 
of young, adult New Zealand males. When this group was split into two self-identified ethnic 
groups, VO2maxLBM and %BF were only predictors of these blood markers in the Māori 
Chapter 6  Study 2 
 119
group, with no significant interactions seen in the non-Māori group. While the relationship 
observed in the Māori cohort is in accordance with the only published study of these 
relationships done in Māori (Stannard, Holdaway, Sachinwalla, & Cunningham, 2007), the 
absence of a relationship between insulin sensitivity and VO2maxLBM/%BF in the non-Māori 
group is a surprise. Interestingly, Bogardus et al. (Bogardus, Lillioja, Mott, Hollenbeck, & 
Reaven, 1985) showed that aerobic capacity and %BF were significantly correlated with 
insulin sensitivity in Caucasian and Native American men, but that these associations were 
only marginally statistically significant. What’s more, Bruce et al. (C. R. Bruce et al., 2003) 
reported that VO2max and oxidative capacity were even better predictors of insulin sensitivity 
than percentage body fat. However, Coon et al. note that most studies using aerobic capacity 
(VO2max) to predict insulin sensitivity have reported VO2max relative to total body weight 
rather than lean mass, so in these studies the influence of VO2max is not entirely independent 
of body fatness (See Section 5.4). In contrast, Coon et al. showed that when VO2max was 
expressed relative to lean body mass, it was not significantly correlated with markers of 
insulin sensitivity in a cohort of American men (Coon, Bleecker, Drinkwater, Meyers, & 
Goldberg, 1989).  
 
Although part of the result may be due to a lack of statistical power in the non-Māori cohort, 
Coon et al. found no significant interaction between VO2maxLBM and insulin sensitivity in a 
cohort over four times the size of the non-Māori group I tested  (Coon, Bleecker, Drinkwater, 
Meyers, & Goldberg, 1989). However, this does not explain why no significant relationship 
was observed between %BF and insulin sensitivity in my non-Māori group, as has been 
shown in these, and other studies (Boden, Chen, DeSantis, & Kendrick, 1993). On the other 
hand, our non-Māori cohort were fitter (had a higher VO2maxLBM) and leaner (lower %BF) 
than the Māori group and the lack of data spread in the non-Māori cohort was a weakness in 
Chapter 6  Study 2 
 120
this study. Thus, environmental factors such as physical activity levels or nutritional status, 
rather than ethnicity, may explain part of the result of this study. With this in mind, recording 
physical activity levels of participants prior to testing would have been ideal. Another 
possible confounder in regard to measurement of insulin sensitivity in this study was that the 
amount of glucose administered during the OGTT was the same (75g) for all participants 
regardless of body weight and LBM. Because of the role of skeletal muscle in the uptake of 
glucose from the blood, a greater total LBM should increase glucose disposal and therefore 
confound postprandial indices of insulin sensitivity (2hr insulin, 2hr glucose). On the other 
hand, an OGTT is the standard diagnostic test for insulin sensitivity in New Zealand and is 
used to identify insulin resistance/glucose tolerance in Māori and non-Māori. Furthermore, if 
the fasting insulin values, a measure unaffected by the glucose load, are analyzed against the 
independent variables of this study alone, the differences between the Māori and non-Māori 
cohort still remain. Indeed, fasting insulin has been shown to be as effective as HOMA-IR in 
predicting insulin sensitivity in Māori (Bell, McAuley, Mann, Murphy, & Williams, 2004) 
and is identified as an adequate means of measuring insulin sensitivity in glucose tolerant 
individuals such as the participants of this study (K. McAuley et al., 2001).   
 
Regardless of confounding factors, the results of the present study viewed in conjunction with 
the findings of the previously referred to studies by Coon et al. and Bogardus et al., suggest 
that the relationship between aerobic capacity and insulin sensitivity may be ethnically 
determined. If, as the results of this study suggest, the association between aerobic capacity 
and insulin sensitivity is strong in Māori but not NZEO, the focus for Māori-targeted 
intervention for reducing diabetes risk, would obviously benefit from inclusion of physical 
fitness rather than simply weight loss alone. Perhaps an emphasis on physical inactivity as a 
more important risk factor should be considered in Māori.  
Chapter 6  Study 2 
 121
 
Currently, the government vehicle of choice for increasing public physical activity levels is 
the ‘Push Play’ campaign delivered by SPARC (Sport and Recreation New Zealand) 
(SPARC., 2005). Through media and other types of community based promotion, ‘Push Play’ 
encourages people to engage in 30 minutes of physical activity per day (60 minutes for 
children), by incorporating physical activity in to everyday activities such as playing with the 
kids or walking to work. Similarly, the 10,000 Steps Program (and similar pedometer-based 
walking initiatives) promote increased physical activity by encouraging people to take at least 
10,000 steps per day, often incorporated into their occupational time. Although well 
intentioned, it is difficult to assess the success of such campaigns among Māori. As discussed 
previously (see section 3.5), in order for interventions to be effective for Māori, they must be 
culturally appropriate for Māori, and therefore involve Māori at all levels; from planning to 
implementation (Ministry of Health, 2002).  
 
Furthermore, such initiatives may motivate sedentary individuals to become more active, but 
the recommendations of these programs may not be sufficient to reduce adiposity or the 
effects of obesity and insulin resistance in Māori and others. On the other hand, the mode of 
exercises recommended to improve metabolic control (usually cardiovascular exercise such as 
walking, running, cycling, swimming etc) are difficult or unacceptable for some, particularly 
the already obese, to perform in public. Alternatively, resistance training has been shown to 
be an effective training method to improve insulin sensitivity in non-Māori (Evas & 
Plotnikoff, 2006; Ibanez et al., 2005; Poehlman, Dvorak, DeNino, Brochu, & Ades, 2000), 
and may be a more appropriate form of exercise for the obese and elderly. This is in 
accordance with the results of my previous study (see chapter 5), which suggests that 
increasing muscle mass (through resistance training), may be just as effective in improving 
Chapter 6  Study 2 
 122
insulin sensitivity in Māori, as cardiovascular exercise. Resistance training can be done 
indoors and perhaps could form the basis of a marae-based exercise initiative. 
  
There are limitations when comparing measures of health within separate ethnicities, as 
ethnicity is often more closely associated to cultural and social identity than it is to biological 
or genetic factors (Neil Pearce, Foliaki, Sporle, & Cunningham, 2004). Because of this, 
identifying what appears to be a physiological disposition to a disorder such as insulin 
resistance in Māori may actually be a result of a certain lifestyle adopted by those who 
identify as Māori. Identifying participants according to ‘heritage’ rather than self-identified 
‘ethnicity’ may be more appropriate to detect genetic dispositions in health.     
 
6.4.1 Chapter Summary 
We have shown that VO2maxLBM and %BF are predictors of insulin sensitivity, in glucose 
tolerant New Zealanders. When ethnicity was accounted for however, VO2maxLBM and 
%BF were only predictors of insulin sensitivity in the Māori group, but not the non-Māori 
group. Although literature is limited in Māori, the findings of this study are in accordance 
with other studies involving Māori. While reduced statistical power within the non-Māori 
cohort may have affected the result in that group, other studies in non-Māori, have found 
similar results. 
Chapter 7  Study 3 
 123
7 Study 3: Bioelectrical impedance: A valid 
method of measuring body composition in Māori, 
when compared to hydrodensitometry.  
 
7.1 Introduction 
Adiposity is inversely related to insulin sensitivity, and it is believed that excess lipid 
accumulation, or stagnancy, is responsible for disruption of the insulin signalling cascade in 
insulin sensitive tissues (Boden, 1997; Boden, Chen, Ruiz, White, & Rossetti, 1994; Roden et 
al., 1996). Likewise, population studies have identified obesity as a strong predictor of 
diabetes, other non-communicable diseases such as stroke and breast cancer, and all-cause 
mortality (Solomon & Manson, 1997). Armed with this knowledge, researchers, clinicians, 
and other health practitioners have sought precise and reliable methods to assess body fat 
percentage (%BF) in individuals, so as to predict the development of insulin resistance, 
diabetes and other obesity related illness.   
 
The body mass index (BMI) was originally developed as a simple means of identifying those 
at risk of obesity related illnesses, with distinct “cut off” points developed to categorize the 
level of risk. These cut off points have come under scrutiny in both lay and scientific media, 
as healthy people are often mistakenly categorized as overweight or obese, while other groups 
are at greater risk of obesity, and its co-morbidities, at a relatively low BMI (Kosaka, Kuzuya, 
Yoshinaga, & Hagura, 1996; K. McAuley, Williams, Mann, Goulding, & Murphy, 2002).   
 
Body mass index is derived from a person’s weight in relation to their height (BMI = weight 
(kg)/height (m2)), measures that are easily and quickly obtained in the clinical healthcare 
Chapter 7  Study 3 
 124
setting.  Although a direct measure of body fat percentage is a more valid predictor of health 
status, obtaining an accurate measure of an individual’s body fat percentage can involve 
difficult and costly procedures. 
 
Epidemiological evidence indicates that particular BMI ranges, defined by these cut-offs, are 
associated with differing levels of mortality and risk of other ‘life-style’ diseases (Solomon & 
Manson, 1997).  Whether an individual’s measured BMI lies within a particular range is then 
often extrapolated to indicate the presence of obesity-related disease or mortality risk.  
However, the use of BMI for such purposes makes the broad assumption that increasing BMI 
is a function of increasing adiposity.  Whilst at a population level this is likely true, for an 
individual this assumption may not hold.  Thus, for diagnosis at an individual level, a more 
direct method for estimating body fatness must be found.   
 
A second, but important issue is that the sample from which population-based BMI risk 
assessment is based may not necessarily represent other populations (WHO, 1995).  A case in 
point is the use of BMI cut offs derived from populations of European decent, to predict 
health status in Māori.  What is considered a “healthy” BMI range for non-Māori may be 
different from Māori as physical characteristics, and the association with disease risk factors, 
probably differ significantly between these ethnic groups.  For example, Asians and Indians 
have been shown to have more body fat than Europeans at a given BMI, while Polynesians 
and Māori appear to have less adiposity than Caucasians at the same BMI (Swinburn, Ley, 
Carmichael, & Palnk, 1999; WHO, 2004).  
 
Hydrodensitometry, also known as hydrostatic weighing or underwater weighing (UWW), has 
become the ‘gold standard’ procedure for obtaining an accurate measure of percent body fat 
Chapter 7  Study 3 
 125
(%BF). This procedure is based on the Archimedes principle of water displacement, and relies 
on assumptions regarding density of body tissues in order to calculate %BF (Heyward & 
Wagner, 2004). Accurate measurement also requires calculation of residual lung volume 
whilst submerged; this is then accounted for when calculating true underwater weight. Thus, 
hydrodensitometry is a difficult procedure to perform, especially for those who are lacking in 
mobility, such as the clinically obese. It is therefore impractical for use in population studies 
or clinical assessment in heterogeneous groups.   
 
Bioelectrical impedance machines (BIA) on the other hand, have become more popular in 
recent years because they are now relatively cheap and easy to use. These machines relay a 
small electrical current through the body, and based on the current’s impedance through 
tissues, an estimation of total body water (TBW) can be obtained. Water is a good conductor 
of electricity, so the current flows more easily through fat-free mass (FFM) which has a high 
water content (~73% water), while a body with more fat ‘impedes’ the electrical current 
because of fat’s relatively small water content (Heyward & Wagner, 2004). From estimates of 
TBW, FFM (and therefore %BF) can then be predicted (See section 2.4.2.3).  The whole 
procedure takes less than a minute and requires little expertise from the operator or demand of 
the individual being tested.  If proven reliable, BIA could provide a practical solution for 
obtaining measures of body composition in population studies and clinical assessment.  
 
Thus the purpose of this study was to investigate the reliability and applicability of BIA for 
measuring body composition in a Māori and non-Māori cohort, when compared with 
hydrodensitometry.  
 
Chapter 7  Study 3 
 126
7.2 Methods 
7.2.1 Participants 
Forty eight men (28 Māori & 20 non-Māori) aged 27 ± 5 (mean ± SD) years old participated 
in this study. All participants completed a written health questionnaire and were considered 
healthy with no cardiovascular, respiratory or metabolic disorders. 
 
Table 7-1 – Participant physical characteristics 
 
Māori (n=28) Non-Māori (n=20) Entire Group (n=48) 
Mean SD Mean SD Mean SD 
Age (years) 26.6 5.1 27.0 4.4 26.7 4.8 
Height (M) 1.80 0.07 1.79 0.06 1.80 0.06 
Weight (kg) 95.7 18.2 76.8 9.0 87.8 17.6 
BMI (kg/m2) 29.4 5.1 23.9 2.4 27.1 5.0 
 
7.2.2 Protocol 
7.2.2.1 Hydrodensitometry  
Methods have been described in a previous chapter (Chapter 1). 
 
7.2.2.2 Bioelectrical impedance 
BIA measures were taken 1-2 hours after a small (standard) meal (the first meal after a 10 
hour fast). As BIA measurements were taken just before underwater weighing, subjects wore 
only the same swimsuits that were worn in the underwater weighing tank during the BIA 
procedure. The BIA apparatus used was the Biospace 230 INBODY by GBC Biomed NZ, 
Technipro PTY. This model used ‘direct segmental multifrequency’ impedance measures, and 
Chapter 7  Study 3 
 127
had an ‘8 point tactile electrode system’ to make impedance measurements i.e. two points of 
electrode contact per hand and two points of contact per foot. Electrode contact was made 
with the palms of the hands and at each thumb, as the participant grasped the electrode 
handles, as well as with the soles and balls of the feet, as the participant stood barefoot upon 
the electrode plate. The operator would then enter the participant’s name, age, gender, height 
and weight, and the analyser would take approximately 30 seconds to make the impedance 
measurement and produce a printout of %BF, total body water (TBW) and segmental 
impedance measures from the right leg, left leg, right arm, left arm and trunk. 
 
Participants avoided diuretics, exercise, and alcohol prior to testing; all factors which may 
reduce the reliability of the measurement (Heyward & Wagner, 2004).   
 
7.2.3 Statistical Analyses 
Independent samples t-tests were used to identify any significant differences in the variables 
measured, between the Māori and non-Māori cohort. Bivariate correlation analysis was 
performed, to determine the relationship between measures of percentage body fat from 
underwater weighing and BIA. Simultaneous multiple regression analysis was also employed 
to determine the effect of ethnicity on this relationship. All statistical analyses were 
performed using SPSS version 10, while scatter plots were made using Microsoft excel 2003. 
Chapter 7  Study 3 
 128
7.3 Results 
Forty eight people participated in this study (28 Māori and 20 non-Māori).  
7.3.1 Body composition 
The average %BF from BIA and underwater weighing, as well as the BMI of the entire group, 
Māori only and non-Māori only, are shown in the table below. Māori had significantly higher 
%BF, LBM and BMI values, compared with the non-Māori group (p<0.001). 
 
Table 7-2 – BMI and percentage body fat from underwater weighing and BIA 
 
Māori Non-Māori Whole Group 
Mean SD Mean SD Mean SD 
BMI (kg/m2) 29.4 5.1 23.9 2.4 27.1 5.0 
%BF (UWW) 22.9 7.0 15.9 5.0 20.0 7.1 
%BF (BIA) 20.6 7.9 13.6 3.7 17.7 7.3 
 
7.3.2 Statistical Analyses 
Bivariate correlation revealed that %BF obtained from the BIA apparatus was a significant 
predictor of %BF obtained using hydrodensitometry in the entire cohort (P<0.001), the Māori 
cohort (P<0.001) and the non-Māori cohort (P=0.017). BIA values explained 77% of variation 
in the entire group, 76% in the Māori group and 53% in the non-Māori group. 
 
Multiple regression analysis was also performed using ethnicity and %BFBIA as independent 
variables to ascertain the effect of ethnicity on the relationship between %BFBIA and 
%BFUWW. Ethnicity however, had no significant effect upon this relationship (P>0.05). 
 
Chapter 7  Study 3 
 129
Figure 6 – Correlation between underwater weighing and BIA (entire cohort) 
y = 0.793x + 1.8405
0.0
5.0
10.0
15.0
20.0
25.0
30.0
35.0
40.0
45.0
50.0
0.0 5.0 10.0 15.0 20.0 25.0 30.0 35.0 40.0 45.0
Underwater weighing (%BF)
BI
A 
(%
BF
)
 
Figure 7 – Correlation between underwater weighing and BIA (Māori cohort only) 
y = 0.8504x + 1.1261
0.0
5.0
10.0
15.0
20.0
25.0
30.0
35.0
40.0
45.0
50.0
0.0 5.0 10.0 15.0 20.0 25.0 30.0 35.0 40.0 45.0
Underwater weighing (%BF)
BI
A 
(%
BF
)
 
Chapter 7  Study 3 
 130
   
Figure 8 – Correlation between underwater weighing and BIA (non-Māori cohort only) 
y = 0.389x + 7.4204
0.0
5.0
10.0
15.0
20.0
25.0
30.0
35.0
40.0
45.0
50.0
0.0 5.0 10.0 15.0 20.0 25.0 30.0 35.0 40.0 45.0
Underwaterweighing (%BF)
BI
A 
(%
BF
)
 
 
7.4 Discussion 
Because of the association between adiposity and ‘lifestyle illnesses’ such as diabetes and 
cardiovascular disease, accurate measurement of body composition is important in identifying 
those most at risk of these disorders. Although useful to identify obesity trends in population 
studies, the body mass index (BMI) is not a direct measure of adiposity and can therefore 
under/overestimate obesity and associated health risks in certain populations. However, 
proven methods of body composition assessment, such as hydrodensitometry and DEXA, are 
not practical for population studies. On the other hand, BIA provides an easy and direct way 
to measure fatness, and therefore risk of insulin resistance, but its accuracy in certain 
populations has been questioned. Thus the purpose of this study was to investigate the 
Chapter 7  Study 3 
 131
accuracy of employing BIA to measure %BF in a group of Māori and non-Māori males, when 
compared to hydrodensitometry.  
 
We have shown that %BF values obtained with the BIA apparatus correlated well with %BF 
measures from underwater weighing. Using an earlier version of the apparatus I’ve used, 
Malavolti et al. showed that whole-body fat-free mass and appendicular lean body mass 
obtained from BIA, correlated well with measurements from DEXA (Malavolti et al., 2003). 
In conjunction with these findings, my results suggest that BIA could provide a valid and easy 
way to measure %BF in both Māori and non-Māori. Although the relationship between BIA 
and UWW seemed weaker in the non-Māori group (r=0.76 Māori vs. r=0.53 non-Māori), this 
is most likely due to an insufficient spread of data in this group.  
 
Although my results indicate that it may be appropriate for obtaining population data, the use 
of BIA for individual assessment must be approached with some caution. While it correlated 
well with UWW, BIA did underestimate %BF by an average of 2.3% in all three groups 
tested, when compared with UWW. This inaccuracy is relatively modest compared with the 
findings of other studies comparing BIA to hydrodensitometry. Heath et al. revealed that BIA 
under predicted %BF by 5.7% in men and 9.1% in women (Heath, Adams, Daines, & Hunt, 
1998). Studies have also shown that, when compared with hydrostatic weighing, BIA 
overestimates fatness in lean individuals and underestimates it in overweight individuals 
(Heath, Adams, Daines, & Hunt, 1998; Segal, Gutin, Presta, Wang, & Van Itallie, 1985). 
Accordingly, my study showed that BIA underestimated %BF for 21 of the 25 participants 
that were over 20% BF, and over estimated %BF in 4 out of the 5 ‘lean’ participants 
(%BF<10%), supporting the idea that BIA underestimates %BF in overweight individuals and 
Chapter 7  Study 3 
 132
overestimated it in lean individuals (Heath, Adams, Daines, & Hunt, 1998; Segal, Gutin, 
Presta, Wang, & Van Itallie, 1985). 
 
Figure 9 - Correlation between underwater weighing and BIA (entire cohort) 95% confidence 
intervals added 
 
 
What’s more, the degree to which BIA under/overestimates %BF, can be relatively large. In 
this study BIA under/overestimated %BF by at least 5% for 14 of the 48 participants, 3 of 
which were more than 10% under/overestimated. Other studies have recorded similarly large 
under/overestimations of %BF from BIA when compared with hydrostatic weighing (Heath, 
Adams, Daines, & Hunt, 1998). If technical error is responsible for these deviations, the re-
breathing protocol during underwater weighing is likely responsible, rather than the BIA 
apparatus.  
Chapter 7  Study 3 
 133
Figure 10 – The relationship between body fatness and the degree to which BIA 
under/overestimates percent body fat  
y = -0.207x - 2.3004
-15.0
-10.0
-5.0
0.0
5.0
10.0
15.0
20.0
-20.0 -15.0 -10.0 -5.0 0.0 5.0 10.0 15.0 20.0 25.0
variation from 20% BF
u
n
de
r/o
v
e
re
s
tim
a
tio
n
 
(%
BF
)
 
 
In order to obtain an accurate measurement of %BF from underwater weighing, residual lung 
volume, the volume of air remaining in the lungs after maximal expiration, should be 
accounted for (Buskirk, 1959). Buoyancy of an individual is increased as RV increases, so 
that a greater RV can make a person appear lighter under water than they actually are and 
therefore overestimate body fatness. This suggests that if RV is measured and accounted for, 
the amount of air in the lungs during the measurement of underwater weight should make no 
difference. On the other hand, Welch and Crisp (Welch & Crisp, 1958) revealed that body 
density was significantly different between lung volume measured at half expiration and full 
expiration. If so, measuring RV at full expiration is important, but because maximal 
expiration can be uncomfortable underwater and is measured subjectively, measuring RV is 
not without its limitations. After having been underwater for up to 10 seconds with air fully 
Chapter 7  Study 3 
 134
expired, many of the participants of this study found it difficult to stop themselves from 
inhaling, as was required of them before connection to the re-breathing apparatus. Although 
measures were taken to ensure that participants did not inhale before the re-breathing 
apparatus was in their mouths and the circuit open, it is possible that slight inhalation may 
have occurred which could alter the dilution and RV measurement.   
 
Another limitation of BIA is its need for specific equations to accurately calculate %BF. In 
order to make BIA valid for certain ethnic groups, ethnic specific equations are required. 
Rising et al. showed that BIA %BF measures calculated by the software provided with the 
analyser they used, underestimated lean body mass in Pima Indians when compared to 
underwater weighing, by 5.3 ± 8.6 kg (Rising, Swinburn, Larson, & Ravussin, 1991). 
Similarly, Jakicic et al. found that BIA underestimated LBM in Caucasians, while 
overestimating LBM in African-Americans (Jakicic, Wing, & Lang, 1998). In my study 
however, ethnicity did not have a significant effect upon this relationship (p>0.05), even 
though the same equation was used for both Māori and non-Māori. Furthermore, although 
BIA underestimated %BF when compared to UWW, the underestimation was the same in 
both ethnic groups (2.3%).  
 
This study also revealed that BMI correlated well with the %BF measures obtained in the 
group tested. Indeed the correlation between BMI and hydrodensitometry was slightly 
stronger than that for BIA and hydrodensitometry in the entire group (r=0.791 vs. r=0.773) 
and the Māori group (r=0.775 vs. r=0.756), but very similar in the non-Māori group (r=516 
vs. r=0.526). This suggests that BMI in this group of healthy Māori men is an acceptable way 
to identify the degree of insulin sensitivity. Although this implies that BMI may be a valid 
index of risk in the group which I tested, caution must still be used in applying BMI to 
Chapter 7  Study 3 
 135
individual assessment as it is not a measure of adiposity, which is the actual risk factor 
associated with metabolic disorder.  
 
In this study, just one type of BIA analyser was used. Studies have shown that impedance 
measures obtained from different BIA analysers can produce %BF measurements which differ 
significantly (Heyward & Wagner, 2004). While this does not affect the results of my study, it 
is important to understand this if different analysers are used interchangeably during 
population research and clinical practice. For example, if a BIA analyser is used to obtain 
repeated measures of body composition during a longitudinal study, the same analyser should 
be used. I was unable to obtain information regarding the equations this apparatus used to 
determine %BF, so it is questionable whether the results obtained with the Biospace 230 
INBODY BIA are comparable with other machines. 
 
The source of error, as a result of technician skill, is quite minor due to the nature of the 
apparatus. However, the position of electrode placement has been shown to be a major source 
of error in other tetra polar BIA analysers. While placement of electrodes in other methods 
requires technicians to place electrodes at certain points on the wrist and ankle while the 
participant lies supine, the specific apparatus used in this study is less intricate. Impedance 
with this apparatus is measured while the subject stands on the lower body electrode plates, 
meaning that contact with electrodes are with the soles of the feet, while upper body electrode 
contact comes from the palms of the hands while grasping the electrode handles. Because of 
this, electrode contact with the feet and hands is not completely standardized, creating another 
possible source of error. 
 
Chapter 7  Study 3 
 136
Hydration and food consumption is also a source for error when using BIA. Participants in 
this study were not in the fasted state, as is suggested by Heyward and Wagner (Heyward & 
Wagner, 2004), but food consumption was within 1-2hr postprandial. This consumption time 
should have little effect on the BIA measurements in this study however, considering most 
studies have shown food consumption and hydration to have little effect on impedance in the 
first 2hrs post-consumption (P. Deurenberg, Weststrate, Paymans, & Van Der Kooy, 1988; M. 
R. Gallagher, Walker, & O'Dea, 1998). Kushner et al. conclude that eating and drinking have 
little effect on the impedance measures until 2-4hr post consumption (Kushner, Gudivaka, & 
Schoeller, 1996), although others have found consumption to effect measures at just 1hr post 
meal (Rising, Swinburn, Larson, & Ravussin, 1991). 
 
Percentage body fat (%BF) obtained from BIA is based on assumptions of the impedance of 
certain tissues (See section 2.4.2.3). Whenever assumptions are made from one population to 
another, there is a possibility of error when applying the assumption to a different population. 
Because underwater weighing is the standard to which BIA is usually compared, and this 
method is based on assumptions regarding the density of certain tissues obtained from 
reference data in just a few cadavers (Heyward & Wagner, 2004), it is important that the 
principles and assumptions behind underwater weighing are understood when applied to the 
context of different ethnic groups. Bone density has been shown to differ between ethnicities, 
with Māori and Polynesian having denser bone than Caucasians (Reid, Mackie, & Ibbertson, 
1986; Swinburn, Ley, Carmichael, & Palnk, 1999). Because bone makes up a significant part 
of LBM in the 2 component model, applying assumptions of LBM density to ethnic groups 
with greater bone density must be done with caution.  
 
Chapter 7  Study 3 
 137
Interestingly, considering the body of literature dedicated to developing more direct measures 
of body composition to predict health status, no accepted guidelines for a healthy percentage 
body fat range exist. In 2000, Gallagher et al. published a paper which provided the ground 
work for establishing international healthy body fat ranges (D. Gallagher et al., 2000), but 
almost a decade later and, to my knowledge, standard body fat ranges have still not been 
established. Until such ranges are established, development of easier to use methods of body 
composition measurement, are of little value in public health assessment. On the other hand, 
the realization that body fat distribution is a more significant predictor of insulin sensitivity 
and health risk than total body fatness, may have halted the development of appropriate body 
fat ranges as the focus shifts toward body fat distribution and the measurement of ectopic fat 
stores.  
 
7.4.1 Chapter Summary 
We have shown that the Biospace 230 INBODY BIA analyser, provided %BF measures that 
correlated well with the ‘gold standard’ method of underwater weighing in a group of healthy 
males but that with the manufacturers equation, BF% was underestimated by an average of 
2.3% in the entire group (see Table 7-2) and this was similar in Māori and non-Māori. Also, 
BIA may not be appropriate for individual assessment, as it caused significant under/over 
estimation of %BF. While the results of this study do not demonstrate a significant 
contribution from ethnicity like other studies, this could be due to the small sample size 
employed. 
Chapter 8  Study 4 
 138
8 Study 4: The repeatability of percutaneous 
electrically stimulated twitch characteristics in 
healthy men 
 
8.1 Introduction 
Within the body, skeletal muscle is the main site of insulin-mediated glucose disposal (Baron, 
Brechtel, Wallace, & Edelman, 1988) and lipid oxidation (cited in Kim, Hickner, Cortright, 
Dohm, & Houmard, 2000). Thus, skeletal muscle insulin sensitivity largely determines whole 
body insulin sensitivity.  
 
A whole skeletal muscle is innervated by many motor neurons; each motor neuron innervates 
a set of muscle fibres (multi-nucleate cells called myocytes).  The motor neuron and the 
muscle fibres it innervates are called the motor unit.  The fibres in each motor unit area are all 
innervated together and possess similar functional characteristics and physiological 
properties. However, between motor units these fibre characteristics and properties are 
different and can be compared and classified by either histochemical, biochemical, 
morphological or physiological characteristics (Scott, Stevens, & Binder-Macleod, 2001). 
Although current methods of this “fibre typing” rely mostly on biochemical and histochemical 
analysis of muscle samples obtained through biopsy, muscle fibres are also characterized by 
certain contractile characteristics. In fact, the earliest fibre-typing research was based on 
observations of different contractile behaviour between different muscle groups in animals 
and humans (Needham, 1926). Fast twitch fibres were shown to have a shorter twitch time, a 
higher twitch tension and a greater fatigability than slow twitch fibres (Sadoyama, Masuda, 
Miyata, & Katsuta, 1988; Thorstensson & Karlsson, 1976). Twitch characteristics and 
Chapter 8  Study 4 
 139
muscular fatigability measured with methods other than biopsy, have also proven to correlate 
well with fibre type composition (E. Suter, Herzog, Sokolosky, Wiley, & Macintosh, 1993).  
 
The proportion of these varying fibre types, within a whole muscle, has also been shown to 
relate closely with body composition and insulin sensitivity. In this context, if a person 
possesses a greater proportion of fast twitch (type IIX glycolytic) muscle fibre, they are more 
likely to have increased adiposity and reduced insulin sensitivity (Hickey, Carey et al., 1995; 
Tanner et al., 2002; Wade, Marbut, & Round, 1990). Conversely, those characterised by more 
slow twitch (type I oxidative) fibres, are less likely to be obese or insulin resistant (Lillioja et 
al., 1987a). This association between muscle fibre-type and insulin sensitivity is likely due to 
the variation in oxidative capacity between muscle fibres, which alter the rate of lipid 
metabolism and therefore glucose transport (Daugaard et al., 2000) (See section 2.6.3 and 
2.6.4).  
 
It is thought that fibre type is genetically determined (Komi et al., 1977) and a number of 
studies have shown an ethnic difference in fibre type proportion in well-trained and obese 
individuals (Ama et al., 1986; Kohn, Essen-Gustavsson, & Myburgh, 2007; Tanner et al., 
2002). However, more recent studies have revealed that fibre type proportion can change in 
response to training or inactivity (Adams, Hather, Baldwin, & Dudley, 1993; Schantz, 
Billeter, Henriksson, & Jansson, 1982; Staron et al., 1991). Denervation, as occurs in the leg 
muscles of paraplegics, reduces the relative number of type 1 fibres (Burnham et al., 1997; 
Martin, Stein, Hoeppner, & Reid, 1992) and fatigue resistance (Rochester et al., 1995). These 
adaptations can be reversed by repeated electrical stimulation (Martin, Stein, Hoeppner, & 
Reid, 1992; Rochester et al., 1995), demonstrating that contractile activity alone, can induce a 
shift in the muscle fibre from fast, fatiguing fibres (IIX), to slower, less-fatiguing fibres  (IIA 
Chapter 8  Study 4 
 140
and I). However, in able-bodied people, changes in fibre type with training are generally 
small, restricted to conversion of type IIB to type IIA (P. Andersen & Henriksson, 1977; 
Houmard et al., 1993), or non-existant (P D  Gollnick et al., 1973). Simoneau and Bouchard 
(J. A. Simoneau & Bouchard, 1995) suggest that fibre type proportion is partly a function of 
both genetics and environment.   
 
Given the inherent nature of an individual’s fibre type and the apparently protective effects of 
slow fibre types against the development of insulin resistance, understanding an individual’s 
muscle composition could potentially be a means to identify early, those most at risk of 
metabolic disorders such as diabetes. Also, because the effects of different training intensities 
are fibre type specific (Cortez, Torgan, Brozinick, & Ivy, 1991), identifying fibre-type 
proportion could provide more effective exercise prescription.    
 
In order to obtain an estimate of muscle fibre proportion, muscle biopsies are usually 
required. Once a sufficient sample of muscle is obtained through biopsy, the individual fibres 
are identified by biochemical or histochemical analysis, or myosin heavy chain identification. 
The biopsy procedure is very invasive, contains some risk, and because of inter-muscular, as 
well as intra-muscular variation of fibre type, the reliability of obtaining a true representation 
of muscle composition has been challenged (Lexell, Henriksson-Larsen, & Sjostrom, 1983; 
Lexell, Taylor, & Sjostrom, 1985). Furthermore, even if biopsy samples are reliable, the 
classifications obtained through the different analysis techniques mentioned above do not 
always agree (Staron, 1997). 
 
Percutaneous (via the skin) electrical stimulation has been utilized for decades in many 
clinical settings to maintain muscle mass and strength during times of immobilization (Delitto 
Chapter 8  Study 4 
 141
& Snyder-Mackler, 1990; Lake, 1992). Electrical stimulation allows the muscle to produce 
force while bypassing the nervous processes involved with a voluntary contraction. In a 
research context, electrical stimulation has provided a way to simulate voluntary muscle 
contraction, and has been used in many physiological studies as a substitute for voluntary 
exercise. Only one study that I know of has applied it to the measurement of contractile 
properties however (Blimkie, Sale, & Bar-Or, 1990). 
 
Therefore, it was the intention of this study to a) observe whether twitch characteristics 
obtained by measuring the time to peak tension of an electrically stimulated twitch were 
repeatable, and b) whether variation could be observed between subjects, in order to ascertain 
the feasibility of investigating the electrically evoked twitch method and it’s relation to 
muscle fibre type. 
 
Chapter 8  Study 4 
 142
8.2 Methods 
8.2.1 Participants 
Nine men aged 27 ± 4.3 years (mean ± SD) volunteered to participate in this study. All 
participants completed a health questionnaire; none reported any cardiovascular problems, 
diabetes, used a pacemaker, or chronic injuries that would prevent them from performing the 
movements required. All of the participants participated in social sport but none were well-
trained. Their physical characteristics are described in the table below. 
 
Table 8-1 – Participant physical characteristics 
Participant Age (yrs) Height (m) Weight (kg) BMI (kg/m2) %BF 
IW 25 1.95 110.2 29.0 28.0 
NP 32 1.59 61.0 24.1 14.0 
HM 18 1.74 69.4 22.9 8.0 
WR 25 1.94 75.8 20.1 13.5 
GE 32 1.67 62.4 22.4 8.0 
CM 29 1.81 84.5 25.8 14.0 
TM 28 1.67 62.8 22.5 15.0 
DN 26 1.82 71.5 21.6 21.3 
JM 28 1.78 68.6 21.7 16.0 
Mean 27 1.77 74.0 23.3 15.3 
Standard Deviation 4.3 0.12 15.4 2.7 6.2 
 
Chapter 8  Study 4 
 143
8.2.2 Protocol 
Participants attended the Massey University Human Performance Laboratory at the Turitea 
campus, in the morning after breakfast, but before any form of exercise for that day.  All were 
asked not to perform any unnecessary leg exercise in the 3 days leading up to the test.  
 
Usually participants came in groups of 2-4, though this was not always the case. Each 
participant was asked to sit in a seated position on a specially designed rig and was secured to 
the seat by tightening a safety belt around the waste. A load cell (Sensatronics – Class; C3 
DIV 3000) was then attached to the participant’s dominant leg (identified as the leg they 
would naturally kick a ball with) just above the ankle (maleolus) with a Velcro® strap which 
could be adjusted to limit the slack between relaxation and contraction. This was connected to 
a custom made D.C amplifier and then to the ‘Power Lab 4/25’ (ADInstruments, Australia). 
This circuit of apparatus measured the amount of force that the lower leg exerted during an 
electrically stimulated leg extension. Percutaneous electrode pads were attached to the lateral 
part of the thigh on the vastus lateralis. One electrode was placed 2/5th of the distance from 
hip to knee and the second at 4/5th the distance. Chart v5.4.2 software by ADinstruments, 
Australia was used to collect data as well as control the stimulation, by way of a Digitimer 
Stimulator (model DS7A, Digitimer Ltd, England). This software provided numerical and 
graphical data of absolute force produced per unit time.  Two point calibration of the load cell 
was performed each morning before testing. 
 
After being attached to the measuring apparatus each participant went through a 
familiarisation routine involving a series of evoked twitches controlled by the Digitimer 
stimulator.  These were all at 10V and increased in amperage from 100-900 mA. Appropriate 
testing amperage was obtained when the maximum force produced by the electrical 
Chapter 8  Study 4 
 144
stimulation reached a plateau and no longer increased with an increase in amperage, or when 
the participant no longer felt comfortable with an increase in amperage.  
 
Five twitches were then measured at the ‘plateau amperage’ with a 30 second break between 
each administration. Each twitch was then analysed for time to peak-force (TPF), and time to 
50% peak-force (T50). 
 
Each participant returned to the laboratory two more times to conduct this repeat this testing 
protocol. The three testing sessions were 3-5 days apart for each participant. 
 
8.2.3 Analysis of twitch     
Figure 11 – Computer analysis of electrically evoked twitch 
 
Force produced by 
electrically induced 
leg extension 
Impulse 
administered 
Initial force (at 
time of impulse) 
(IF) 
Peak Force (PF) 
50% peak force 
50%PF = (PF – IF)/2 
Time to 50% 
peak force (T50) 
Time to peak force 
(TPF)  
 
Time to Force 
production (TFP)  
Chapter 8  Study 4 
 145
Each of the 5 twitches obtained at each testing session was analysed for time to peak force 
(TPF) and time to 50% peak force (T50) using the Chart v5.4.2 software (ADInstruments, 
Australia) (table 4.2).  
 
8.2.4 Statistical Analyses 
Repeat-measures analysis of variance was performed to investigate differences in the 
dependent variables (TPF, T50): a) within a testing session (individual twitch number), b) 
between testing days (session), c) between participants, and d) interactions between a), b), and 
c). Significance required a 95% level of confidence to detect a real effect (P ≤ 0.05).   
 
 
Chapter 8  Study 4 
 146
8.3 Results 
All participants completed all three trials with minimal discomfort and no reported side 
effects.   
 
8.3.1 Reliability of twitch characteristics 
Analysis of variance revealed that there were no significant interactions between individual 
twitches and subject, nor was there an interaction between individual twitch and session 
(p>0.05). However, significant interaction was revealed between subject and day, indicating 
that some subjects did differ from one day to another in TPF and T50 (P<0.001 for both), 
even though overall the difference between trial days was not significant within the group 
(P>0.05).  
 
There was no significant difference between individual twitches taken within each trial day 
either (P>0.05), indicating the repeatability of the test within sessions and between sessions. 
The analysis of variance also showed that there was significant variation between subjects in 
both TPF (P<0.001) and T50 (P<0.01). Table Figure 12 shows the repeatability of the T50 
and Figure 13, the repeatability of TPF (n=9) 
Chapter 8  Study 4 
 147
Figure 12 – Repeatability of the T50 measure for each participant, and the variation between 
participants (each colour represents a single participant) 
0.050
0.055
0.060
0.065
0.070
0.075
0.080
Day
1
Day
2
Day
3
Trial Day
Ti
m
e 
to
 
50
%
 
pe
ak
 
fo
rc
e
 
Figure 13 – Repeatability of the TPF measure for each participant, and the variation between 
participants (each colour represents a single participant) 
 
0.080
0.090
0.100
0.110
0.120
0.130
0.140
0.150
Day
1
Day
2
Day
3
Trial Day
Ti
m
e 
to
 
pe
ak
 
fo
rc
e
 
Chapter 8  Study 4 
 148
8.4 Discussion 
Because a higher proportion of fast-twitch muscle fibre is associated with a greater risk of 
obesity, insulin resistance and type-2 diabetes, measuring an individual’s fibre-type 
proportion may, in theory, aid in identifying those with a greater risk of developing type-2 
diabetes. The metabolic effects of different forms of exercise also appear to be fibre-type 
specific (Cortez, Torgan, Brozinick, & Ivy, 1991), so that identifying an individual’s fibre 
type may provide more specific exercise recommendations to maintain and improve insulin 
sensitivity. Due to the invasive nature of the biopsy procedure required for histochemical and 
biochemical identification of muscle fibre type, recruitment into fibre type-based studies and 
the scope of research in this area has been limited. A few studies suggests that contractile 
properties measured externally, are related to fibre-type proportion obtained through biopsy 
(Ivy, Withers, Brose, Maxwell, & Costill, 1981; Sadoyama, Masuda, Miyata, & Katsuta, 
1988). Thus, it was the purpose of this study to investigate the utility and reliability of twitch 
characteristics (measured as time to peak tension), during a percutaneous electrically 
stimulated twitch. It was not the intention of this study to investigate whether the electrically 
evoked twitch characteristics reported correlated well with fibre type proportions obtained 
through biopsy and histochemical analysis, but to investigate whether the measures were a) 
individual, and b) repeatable, and therefore warranted further investigation.  
 
A number of studies have made the comparison between contractile characteristics obtained 
with external stimuli and histo/bio-chemically analysed fibre type proportion from muscle 
samples, but using different methods to characterise contractile property (Ivy, Withers, Brose, 
Maxwell, & Costill, 1981; Sadoyama, Masuda, Miyata, & Katsuta, 1988; E. Suter, Herzog, 
Sokolosky, Wiley, & Macintosh, 1993). These studies have shown that conduction velocity 
(Sadoyama, Masuda, Miyata, & Katsuta, 1988), peak torque, peak power and the rate of 
Chapter 8  Study 4 
 149
power production (Ivy, Withers, Brose, Maxwell, & Costill, 1981) are related to 
histochemically determined fibre type composition. Others have also shown a relationship 
between muscle fatigue and histochemically determined fibre type (E. Suter, Herzog, 
Sokolosky, Wiley, & Macintosh, 1993; Thorstensson & Karlsson, 1976).  
 
In the present study I have been able to show that evoked twitch characteristics in the 
quadriceps muscles of an individual, in terms of time to peak tension at least, are highly 
repeatable within a testing session. While some individuals produced significantly different 
results on different days, overall there was no significant difference in both TPF and T50 
between testing sessions, suggesting that this measure is largely repeatable. It is important to 
note that there was a significant variation between the twitch characteristics of subjects tested, 
suggesting that a person’s muscle twitch characteristics are inherent and not subject to 
temporal or environmental influences. To my knowledge, only one previous study has 
employed percutaneous electrical-stimulation to determine twitch characteristics, although 
this study measured peak dynamic torque with a dynamometer (Blimkie, Sale, & Bar-Or, 
1990). One other study also used a force transducer, along with EMG, to measure fibre 
conduction velocity, but the measurement was taken during a voluntary contraction, rather 
than an electrically evoked contraction (Sadoyama, Masuda, Miyata, & Katsuta, 1988). 
Because a maximal voluntary contraction can be influenced by psychological factors such as 
motivation, an electrically stimulated twitch may provide a way to overcome the error 
associated with such psychological factors. To date there are no published studies which 
report measuring the time to peak force of an electrically stimulated twitch. 
 
Because I was unable to ascertain whether the measures I took (TPF, T50) were associated 
with other indicators of fibre type characteristics, or whether these measures truly represented 
Chapter 8  Study 4 
 150
the twitch characteristics of the muscle, the next obvious step in this line of research would be 
to compare these twitch characteristics to biopsy samples, in order to determine whether these 
twitch measures are correlated with fibre type proportion. As fibre type is also associated with 
insulin sensitivity and %BF, it would also be of worth to investigate the correlation of these 
twitch characteristics with measures of body composition and insulin sensitivity, and whether 
measurement of contractile properties can be employed to predict insulin sensitivity. 
  
Because of the innovative nature of the methods used in this study, presently, no reviews exist 
which discuss the percutaneous electrical stimulation method and its limitations in 
determining twitch characteristics as such. Nevertheless, care was taken in placing the 
electrode pads in the same spot on the thigh for each session. Besides placement of electrode 
pads, possibility for technical error from impulse administration was small, because the 
timing and magnitude of the impulse was controlled electronically. The variation that some 
participants showed in TPF and T50, between testing sessions, could be explained by either 
muscular fatigue during sessions. Also, attachment of the lower leg to the force transducer has 
been identified as a possible source of operational error. 
 
For field studies, or in laboratories where a dynamometer is not available, measurement of 
peak tension with a simple force transducer may be more appropriate. On the other hand, 
because T50 and TPF have not been compared with fibre type determined from muscle 
samples, as fatigue, peak torque, peak power and rate of power production have been (Ivy, 
Withers, Brose, Maxwell, & Costill, 1981), it is possible that these two measurements may 
not relate to fibre type, or even muscle contractile properties. Nevertheless, if further 
investigation demonstrates that this techniques correlates well with fibre type proportion and 
measures of insulin sensitivity, it could provide a non-invasive way to predict insulin 
Chapter 8  Study 4 
 151
sensitivity in population studies, and aid in the development of fibre-type specific exercise 
recommendations.  
 
8.4.1 Chapter Summary 
The time to peak force (TPF) and 50% peak force (T50) of an electrically evoked twitch in the 
vastus-lateralis is repeatable within a session of 5 twitches and between 3 separate testing 
sessions. Variability of twitch characteristics between subjects is significant. Further study is 
recommended, to assess the validity of this method, as a true representation of muscle 
contractile properties, and whether the contractile properties measured can predict insulin 
sensitivity. 
Chapter 9  Study 5 
 152
9 Study 5: Measuring muscle contractile 
properties to predict insulin sensitivity in healthy 
men  
 
9.1 Introduction 
Skeletal muscle fibres vary in functional and metabolic properties which meet the varying 
requirements placed upon them. Slow twitch fibres contain a slower myosin ATPase, a greater 
mitochondrial volume/density, oxidative enzyme activity, and greater capillarization than fast 
twitch fibres, making them more efficient and less fatigable during prolonged exercise. Fast 
twitch fibres on the other hand, are better suited to high-power, short bouts of muscular 
activity, and evidently contain a greater proportion (volume/volume) of glycolytic enzymes 
than slow twitch fibres (Wilmore, Costill, & Kenney, 2005; Zierath & Hawley, 2004). Whole-
body aerobic fitness, measured as maximal oxygen uptake, is largely determined by blood 
supply to the working muscle (González-Alonso & Calbet, 2003), but also relies upon 
sufficient mitochondrial content and oxidative enzyme activity to utilize this delivered 
oxygen; as indicated above, all of the latter are associated with slow twitch muscle fibre 
(Essén, Jansson, Henriksson, Taylor, & Saltin, 1975; He, Watkins, & Kelley, 2001; Zierath & 
Hawley, 2004).  
 
These noted variations between muscle fibres have implications for health status in humans. 
Chronic endurance exercise (endurance training) increases an individual’s oxidative capacity, 
mainly due to an increase of mitochondrial volume density and capillarization of the trained 
muscle (J. O. Holloszy & Booth, 1976). 
 
Chapter 9  Study 5 
 153
Table 9-1 – Twitch speed, aerobic capacity and fatigability of varying muscle fibre types 
  
This increase in aerobic/oxidative capacity is associated with improved whole body insulin 
sensitivity (C. R. Bruce et al., 2003) and an improved ability to oxidize lipid (Goodpaster, 
Katsiaras, & Kelley, 2003). Chronic muscle contraction, which occurs with endurance 
training, also appears to induce a conversion, albeit slow, from fast twitch to slower twitching 
fibres (Schantz, Billeter, Henriksson, & Jansson, 1982). Interestingly, possession of a greater 
proportion of slow twitch muscle fibres is associated with greater whole body insulin 
sensitivity (Lillioja et al., 1987a).  
 
One functional characteristic of fibre type is its time to reach peak force, or ‘twitch speed’. As 
the name suggests, fast twitch (type II) muscle fibres contract, but also relax at a greater 
velocity than slow twitch fibres (Harridge et al., 1996; Reiser, Moss, Giulian, & Greaser, 
1985). Consequently, the fast twitch fibres are unable to maintain this force production, and 
fatigue more rapidly than slow twitch fibres when forced to contract repeatedly (Thorstensson 
& Karlsson, 1976). Work by Suter et al. (E. Suter, Herzog, Sokolosky, Wiley, & Macintosh, 
1993) showed that the rate at which relative torque decreased after 60 maximal leg extensions, 
was related to the proportion of type II muscle fibres in the vastus lateralis identified by 
histochemical analysis. 
 
Fibre Type Twitch speed Aerobic capacity Fatigability 
Type 1  
(slow twitch/oxidative) 
Slow High Low 
Type 2A  
(fast twitch/oxidative) 
Fast Medium Medium 
Type 2B  
(fast twitch/glycolytic) 
Fast Low High 
Chapter 9  Study 5 
 154
The literature suggests that improved oxidative capacity, as a result of a high proportion of 
well-trained, slow twitch (oxidative) muscle, is a major factor in determining insulin 
sensitivity. Consequently, identifying those with a higher proportion of fast twitch muscle 
fibre may provide a glimpse at those most at risk of insulin resistance and diabetes. Although 
research is still inconclusive as to the roles of genetics and exercise in determining muscle 
composition, some studies suggest that ethnicity influences fibre-type proportion (Ama et al., 
1986; Kohn, Essen-Gustavsson, & Myburgh, 2007; Tanner et al., 2002). Thus, it is possible 
that high-risk populations, such as Māori, may be more prone to insulin resistance due to a 
genetically determined greater proportion of fast twitch muscle fibres. 
 
Although it is well known that fibre type is related to obesity and insulin resistance, no studies 
to my knowledge have attempted to observe a relationship between muscle contractile 
properties, measured externally using methods other than biopsy, and markers of insulin 
sensitivity. Certainly, no study has investigated these relationships in Māori. While I have 
shown in Study 4 that time to peak and 50% peak tension (TPF and T50) are repeatable and 
that there is variation between participants, these two measurements are unique to this study 
and have not been compared with fibre type in the way that fatigue, peak torque, peak power 
and rate of power production have been (Ivy, Withers, Brose, Maxwell, & Costill, 1981). 
Thus, T50 and TPF may not be a representation of fibre type proportion, or even contractile 
properties of muscle, at all. Nevertheless, it was the aim of this study to observe whether T50 
and TPF induced by an electrically stimulated isometric twitch, as well as the fatigability of 
muscle, were related to markers of insulin sensitivity.  
Chapter 9  Study 5 
 155
9.2 Methods 
9.2.1 Participants  
Twenty one men (12 Māori and 9 non-Māori) aged 27 ± 5 (mean ± SD) years old participated 
in this study. All participants completed a health questionnaire; none reported any 
cardiovascular problems, diabetes, used a pacemaker or had chronic injuries that would 
prevent them from performing the movements required. All of the participants participated in 
social sport but none were well-trained. Their physical characteristics are described in the 
table below. 
 
Table 9-2 – Participant Physical characteristics 
 
Mean SD 
Age (yrs) 27 5.0 
Height (m) 1.79 0.09 
Weight (kg) 83.6 17.2 
BMI (kg/m2) 26 4.6 
 
9.2.2 Protocol 
The methods used to obtain measures of blood biochemistry, body composition and aerobic 
capacity in this study, were identical to those previously described (see section 5.2). 
 
9.2.2.1  Electrically evoked twitch test  
Participants followed a protocol previously described in the methods section of Study 4, 
except they only participated in one session of 5 electrically stimulated twitches instead of the 
three separate sessions required of the participants in the feasibility study. 
Chapter 9  Study 5 
 156
 
9.2.2.2  Fatigue Test 
This test was performed on the Biodex isokinetic dynamometer (Biodex Medical systems, 
NY, USA), and was designed similar to a protocol used by Suter et al. (E. Suter, Herzog, 
Sokolosky, Wiley, & Macintosh, 1993).   
 
Participants were asked to sit in the seat of the Biodex apparatus and their dominant leg (see 
section 8.2) was attached to the dynamometer just above the ankle, with a specially designed 
padded strap. The arm of the dynamometer ran parallel to the lower leg and its pivot point 
was adjusted to be in line with, and 10-15mm from, the lateral condyle of the femur.  
 
After a warm up including stretching and 3 sets of 4 leg extensions at various velocities on the 
apparatus, participants became familiar with the force they would be exerting during the test. 
For the testing protocol, participants were asked to perform 60 knee extensions at an angular 
velocity of 90 degrees per second. Participants were encouraged throughout the test to put 
forth a maximal effort for each contraction. Peak torque was recorded for each contraction 
throughout the test. For a detailed description of the dynamometer apparatus and its use, see 
Thorstensen et al. (Thorstensson, Grimby, & Karlsson, 1976).   
 
The Biodex software (Biodex System 3) provided graphs (see Error! Reference source not 
ound.) and tables of the peak torque of all 60 contractions throughout the test. These tables 
were used to calculate the fatigability of an individual. To calculate fatigability i.e. the drop 
off of torque after 60 contractions, the mean of the peak torque for the last five contractions 
(mean of 56-60), was subtracted from the peak torque recorded in the test (Tmax), to obtain 
Chapter 9  Study 5 
 157
the absolute difference (Tdiff). The relative fatigue, the drop off as a percentage of Tmax 
(%diff), was also recorded. 
 
 
  
 
9.2.3 Statistical analyses 
VO2max was expressed relative to lean body mass (lbm – derived from two compartment 
hydrodensitometry) rather than relative to whole body mass because it: a) better represents the 
aerobic performance of the working muscle (Coon, Bleecker, Drinkwater, Meyers, & 
Goldberg, 1989); and b) provides for a truly independent variable in the multiple regression 
analysis.   
 
Simultaneous multiple regression was performed on the dependent variables which were 
markers of insulin sensitivity or glucose tolerance (HOMA-IR, fasting glucose, fasting 
insulin, two-hr glucose two-hr insulin). The independent variables included in the analysis 
were percentage body fat, relative VO2max/lbm (ml.kg-1 lbm.min-1), and T50 or TPF.  
Bivariate correlation was used to describe the relationship between T50 and the dependent 
variables. Bivariate correlation was also used to describe the relationship between measures 
Fmax (highest torque for test) T60 (mean of contraction 56-60) 
Tdiff 
Figure 14 – Computer Analysis of Tdiff by subtracting T60 from Fmax 
Chapter 9  Study 5 
 158
from the fatigue study (PerDecl, Absdiff and Fmax), and the previously noted dependent 
variables. Significance was ascribed at a confidence level of 95% (two tailed). All statistical 
analyses were performed on specialist software (SPSS Version 10).   
9.3 Results 
We were unable to obtain a *2hr glucose sample for one participant and *2 hr insulin samples 
for four participants.  
 
9.3.1 Blood Biochemistry 
Table 9-3 – Blood biochemistry 
 
Mean SD n 
HOMA-IR 1.76 1.5 21 
Fast Ins (pmol.L-1 ) 56 42 21 
Fast Gluc (mmol. L-1) 4.7 0.5 21 
*2hr Ins (pmol.L-1 ) 145 111 17 
*2hr Gluc (mmol. L-1) 4.3 0.9 20 
 
9.3.2 Body Composition and Aerobic Capacity 
Table 9-4 – Percent body fat, lean body mass and VO2maxLBM 
 
Mean SD 
%BF 18.0 7.8 
LBM (kg) 68.0 9.4 
VO2max LBM (ml.kgLBM -1.min-1) 55.43 7.8 
 
Chapter 9  Study 5 
 159
9.3.3 Twitch Test 
9.3.3.1 Multiple regression analysis   
Simultaneous multiple regression analysis revealed that 46% of the variance in HOMA-IR, 45 
% of the variance in fasting insulin, and 56% of the variance in 2hr insulin, could be predicted 
by the T50 value (p<0.05). TPF was not a significant predictor of any of the independent 
variables analyzed (p>0.05). 
 
9.3.3.2  Bivariate Correlation 
Bivariate correlation showed that T50 and TPF were not significantly correlated with 
VO2max/lbm (r<0.2, p>0.05). My results did show however that T50 was directly correlated 
with HOMA, fasting insulin, 2hr insulin and 2hr glucose (p<0.05). T50 was also significantly 
related to %BF as well as absolute body weight (p<0.01). TPF however, was not significantly 
related to any of the variables tested. There was no significant relationship between T50 and 
ethnicity or T50 and age (p>0.05). 
 
9.3.4 Fatigue Test 
9.3.4.1 Multiple regression analysis   
Simultaneous multiple regression revealed no significant contribution to markers of insulin 
sensitivity or glucose tolerance from Fmax, PerDecl or AbsDiff, when included with the other 
independent variables used (VO2max/lbm and %BF) (p>0.05) 
 
9.3.4.2 Bivariate Correlation 
Bivariate correlations showed that Fmax (maximal force produced during the fatigue test) and 
Absdiff (absolute difference between Fmax and the mean of the last 5 twitches) were 
significantly correlated with T50 and %BF (p<0.01). Fmax and Absdiff were also related to 
Chapter 9  Study 5 
 160
HOMA-IR measurements (p<0.05). PercDecl (The mean of the last 5 twitches as a percentage 
of Fmax), which represented the relative decline of force, was not related to T50, TPF, 
VO2max/lbm, %BF or any of the blood markers (p>0.05).  
Table 9-5 – Simultaneous multiple regression analysis 
Dependant variable Independent variable R r² t-value P-value n 
HOMA-IR  0.675 0.455  0.014 21 
T50   2.963 0.009  
VO2max/lbm   -1.563 0.137  
%BF   -1.484 0.156  
Fasting serum insulin  0.670 0.448  0.016 21 
T50   2.803 0.012  
VO2max/lbm   -1.650 0.117  
%BF   -1.241 0.232  
Fasting serum glucose  0.524 0.275  0.132 21 
T50   2.318 0.033  
VO2max/lbm   -0.109 0.915  
%BF   2.318 0.033  
2hr serum insulin  0.746 0.557  0.012 17 
T50   3.353 0.005  
VO2max/lbm   -1.282 0.222  
%BF   -1.702 0.113  
2hr serum glucose  0.619 0.383  0.047 20 
T50   1.650 0.119  
VO2max/lbm   -1.243 0.232  
%BF   0.350 0.731  
 
 
Chapter 9  Study 5 
 161
9.4 Discussion 
Muscle fibre type proportion is related to athletic performance as well as metabolic health. 
Measuring fibre type proportion in an individual can be difficult and invasive however, as it 
requires muscle samples obtained through biopsy. Fibre type proportion is also associated 
with a person’s muscle contractile characteristics and fatigability measured externally. Thus, 
the purpose of this study was to investigate the relationship between electrically evoked 
twitch characteristics, muscular fatigue and markers of insulin sensitivity.  
 
Although studies have shown that insulin sensitivity is associated with an individual’s 
histochemically determined fibre type proportion, to the author’s knowledge, none have 
investigated whether contractile characteristics measured externally could predict insulin 
sensitivity. In this study, I was able to show that electrically evoked twitch characteristics 
were significantly related to estimates of insulin sensitivity (HOMA-IR). There was no 
significant interaction between twitch characteristics and age, or twitch characteristics and 
ethnicity in this cohort. While the TPF measure was not a significant predictor of any of the 
variables studied, T50 was a significant predictor of fasting insulin, 2hr insulin (post glucose 
load) and 2 hr glucose (post glucose load). T50 was also significantly related to BF% in this 
cohort, but not VO2max/lbm LBM. 
 
According to findings of twitch characteristics and its relationship to fibre type, if a 
correlation was observed between T50 and insulin sensitivity, it was expected to be an inverse 
relationship, as fast twitch fibres are associated with obesity and insulin resistance i.e. the 
shorter the T50, the higher the %BF and HOMA values. Surprisingly however, these were not 
inverse relationships as expected. That is, those with the highest HOMA-IR score (more 
Chapter 9  Study 5 
 162
insulin resistant) were characterised by slower twitch speeds. This observation can probably 
be explained according to the methodology I’ve employed.  
 
Electrical conductivity in the thigh may explain the relationship between slower twitch time 
and higher HOMA-IR, because a higher HOMA-IR is strongly associated with increased 
%BF. While muscle is a relatively good electrical conductor, adipose tissue is not. Indeed it is 
possible that conduction of the electrical impulse may be different, but repeatable between 
individuals because of some other inherent or physical factor such as subcutaenous fatness (of 
the thigh). Therefore, the electrical impulse administered percutaneously would be impeded in 
those with greater adiposity surrounding the vastus lateralis, increasing T50. In order to 
account for the impedance, I also observed whether there was a relationship between 
adiposity and ‘time to force initiation’ (identified as the time from impulse administration to 
the moment contractile force went beyond 3 standard deviations from initial force). No 
significant relationship existed between ‘time to force initiation’ and obesity however 
(P>0.05), neither was there a significant change in the results when T50 and TPF were 
adjusted for ‘time to force initiation’. These findings suggest that factors associated with 
adiposity, other than impulse impedance, are responsible for the ‘adiposity/slow-twitch’ 
relationship.   
 
Furthermore, I expected to see a relationship between a subject’s ability to resist fatigue in the 
fatigability test, and their VO2max/lbm Suter et al. (E. Suter, Herzog, Sokolosky, Wiley, & 
Macintosh, 1993) showed that the decline in force during a dynamometer test similar to mine, 
was associated with fibre type. They found that those with a greater proportion of fast twitch 
fibres had a greater relative decline than those with a high proportion of slow twitch fibre. 
This study revealed that the relative decline was not significantly related to markers of insulin 
Chapter 9  Study 5 
 163
sensitivity or aerobic fitness (expressed as VO2max/lbm). Relative decline was not 
significantly related to T50 or TPF either. 
 
It appears that the methods I used to obtain twitch characteristics, namely TPF and T50 of an 
evoked twitch, are probably not valid as measures of fibre type and thus require further 
development. Indeed, T50 and TPF are innovative measures which have not been compared 
with muscle samples, so these two measures may not actually be associated with twitch 
characteristics in the muscle at all. However, the fatigue protocol, which has been shown to 
correlate with fibre type, did not produce any significant relationships between fatigability 
and markers of insulin sensitivity either. This could be due to the use of healthy subjects in 
my studies, as variations in oxidative capacity and insulin sensitivity do not only exist 
between different fibre types, but between the same fibre types of different subjects i.e. It is 
possible to have insulin sensitive type 2 fibres. Thus, observations of the protective qualities 
of type 1 fibre would only be distinguished in unhealthy subjects. Nonetheless, there are a 
number of useful applications for measuring fibre type proportion, warranting further 
development of non-invasive methods of fibre-typing.   
 
Because fibre type is associated with obesity and the ability to oxidize lipids, measuring an 
individual’s fibre type proportion, may allow early identification of those who are most 
susceptible to obesity and metabolic disorder, long before the individual has reached an 
unhealthy state.  
 
The type of exercise i.e. endurance-type or high intensity, and its effectiveness in improving 
metabolic profile, is also influenced by fibre-type, so that measurement of a person’s fibre 
type proportion may allow prescription of exercise that is more specific to an individual’s 
Chapter 9  Study 5 
 164
metabolic needs. ‘Fibre-type dependant’ exercise prescription would apply to primary 
prevention of obesity and insulin resistance, as well as the treatment of those who are obese 
and/or glucose intolerant. In regards to reducing body fat within the obese, Tanner et al. 
(Tanner et al., 2002) found that obese individuals with a higher proportion of slow-twitch 
fibres were better able to reduce body mass with intervention, when compared with those who 
had a higher proportion of fast-twitch fibre.  
 
Studies have shown that muscle fibre type proportion differs between ethnic groups. 
Interestingly, the same ethnic groups that are identified in the literature as having a greater 
proportion of fast-twitch muscle fibre (Ama et al., 1986; Tanner et al., 2002), are also 
identified in epidemiological studies as those with greater risk of obesity and diabetes (Cowie 
et al., 2006). The data highlighting this relationship is limited to only a few ethnic groups 
however, and certainly, there is no muscle fibre-type data from Māori. Because of the 
invasive methods required to accurately measure the proportion of skeletal muscle fibre types, 
it has not been possible to obtain population data comparing the fibre type proportions of 
different ethnic groups. Nevertheless, based on studies in other ethnic groups, it remains 
possible that Māori have a greater predisposition for obesity, insulin resistance and diabetes 
than NZEO, because of a genetically determined, higher proportion of fast-twitch muscle 
fibre.  
 
9.4.1 Chapter Summary 
The results of this study did reveal a relationship between electrically-evoked twitch 
characteristics and blood markers for insulin resistance. There was also a relationship between 
adiposity and twitch characteristics, but neither of these relationships were inverse as 
expected i.e. twitch time in this study increased as adiposity increased, suggesting flaws in the 
Chapter 9  Study 5 
 165
methods used. Fatigability, which has been shown to correlate with fibre type proportion 
measured in-vivo, was not a significant predictor of insulin sensitivity, adiposity or aerobic 
fitness either. The association between fibre type and metabolic health warrants further 
development of non-invasive techniques to enable further investigation in this area, and to 
explore whether fibre type can partly explain ethnic health disparities. 
 
Chapter 10  Conclusions 
 166
10 Conclusions 
 
The line of research presented in this thesis was undertaken in an effort to help explain why 
Māori appear more predisposed to insulin resistance and type-2 diabetes than New Zealanders 
of European Origin. I acknowledge the importance of socio-cultural and socio-economic 
factors, and the role they play in the prevalence of this condition, but this thesis has focussed 
on the physiological factors which may contribute. I also recognise the limitations of 
including ethnicity data in this physiological research. What’s more, I acknowledge that the 
study participants were by and large homogenous, particularly the Māori cohort; thus the 
results may only be applicable to young Māori men. Nevertheless, it is the age represented by 
these participants which is likely critical in the development of diabetes and cardiovascular 
disease; Māori men have arguably the lowest life expectancy of any group in NZ (Statistics 
New Zealand, 2008), as well as a lower onset age for diabetes (Joshy & Simmons, 2006).   
 
10.1  Summary of findings 
The results of the studies here-in have confirmed that the aerobic capacity of lean body mass 
(VO2max LBM) and percentage body fat (%BF) both help to predict insulin sensitivity 
(HOMA-IR) in a healthy group of young, Māori males. Furthermore, as body composition 
and fitness change with time, VO2max LBM and %BF still predict insulin sensitivity in this 
self-identified ethnic group. This relationship may vary between ethnicities however, though 
the reason for this remains unclear.  
 
As a consequence of the relationship between %BF and insulin sensitivity, and thus risk of 
diabetes, it is important that appropriate methods become available to measure body fatness in 
Chapter 10  Conclusions 
 167
Māori as, to date, there are no scientifically determined Māori-specific BMI cut-off values. 
Accordingly, I have shown that BIA is a valid and acceptable means of assessing body fatness 
in young Māori men, when compared to a two compartment, established method of measuring 
body composition (hydrodensitometry).  
 
The scientific literature indicates that muscle fibre type is also independently related to 
obesity and insulin resistance.  As part of my studies, I have attempted to develop a non-
invasive method of apportioning fibre type distribution by measuring the functional 
characteristics of muscle. Whilst this method proved reliable and acceptable to the 
participants, its validity in representing twitch characteristics representative of fibre type 
proportion, is questionable.  Furthermore, and likely as a consequence of its invalidity, twitch 
characteristics obtained with this method were not able to predict markers of insulin 
sensitivity.   
 
10.2  Applicability 
Measuring aerobic fitness, in combination with body composition, is a valid way of 
identifying risk of insulin resistance, as represented by elevated fasting insulin, in Māori men 
who maintain normoglycaemia.  It is important to note the importance of measuring both, as 
both body fatness and aerobic fitness of the lean body mass are independent predictors, but 
the latter requires knowledge of lean body mass.  In a clinical and public health setting, 
however, both may be difficult to obtain because of the facilities and equipment required to 
accurately make these measures.  However, there are valid and reliable means of predicting 
both of these. Indeed, in this thesis, I have described the validity of BIA to predict body 
fatness in young Māori men.  Maximal aerobic fitness, on the other hand, is more difficult to 
measure. There are some tests available, such as a ‘beep’ test which can infer maximal 
Chapter 10  Conclusions 
 168
aerobic capacity from performance, though skill may play some role (Léger & Lambert, 
1982).  Sub maximal testing may be more useful in the clinical, or public health setting, as 
such tests have shown to be valid in large population studies (Noonan & Dean, 2000).  
 
In connection with the findings of others, aerobic capacity (of lean body mass) may not be 
associated with insulin sensitivity in non-Māori however. Since aerobic capacity contributed 
significantly to the variance in insulin sensitivity in the Māori but not the non-Māori cohort, 
ethnicity may determine the extent to which aerobic fitness effects insulin sensitivity. This 
knowledge may help explain why Māori have reduced insulin sensitivity when compared to 
Europeans of the same BMI. It also highlights the possibility of a genetic influence on the 
development of insulin resistance. In context of the theories of genetic adaptation discussed 
earlier (on page 64), Māori and other ethnic groups which had to expend relatively more 
energy to obtain food, would have benefited from a physiological mechanism which 
controlled glucose uptake relative to energy requirements. Europeans on the other hand, may 
have adapted mechanisms which control glucose uptake irrespective of aerobic fitness, in 
response to increased food availability that did not require great energy expenditure.  
 
Fibre-type proportion has also been identified as a possible factor in ethnic health disparities, 
probably because of the varying oxidative properties between different muscle fibres, and the 
necessity of adequate oxidative capacity of muscle in maintaining sensitivity to insulin-
mediated glucose uptake. Thus, although I can only speculate, it is possible that a larger 
proportion of fast or slow-twitch muscle may influence the relationship between aerobic 
fitness and insulin sensitivity. Findings from my study of twitch characteristics, and their 
relationship with obesity and insulin sensitivity are unable to resolve this issue because of 
methodological issues.  Specifically, the electrical stimulation method (which measured the 
Chapter 10  Conclusions 
 169
time to peak force) showed that the speed of the electrically evoked twitch (measured by a 
force transducer) is probably not a function of fibre type, but of the absolute size of the 
participant, and thus probably the cross-sectional area of the thigh muscle which was 
stimulated. Nevertheless, these results do not rule out the future possibility of employing 
contractile characteristics to predict health risk, or in obtaining a representation of fibre type 
proportion if an appropriate tool can be developed and validated.  
 
Another aim of the research described within the thesis was to observe any changes, over a 
time period of two years, in health-related markers in a small group of young Maori men.  
Interestingly, the reduction in aerobic capacity seen within this group was not accompanied 
by a reduction in insulin sensitivity. However, the relationship between aerobic fitness per 
unit lean body mass and insulin sensitivity remained.  These results are likely due to the 
increased lean mass seen within this group after two years. It may be that since the initial 
tests, participants grew a little or undertook some form of resistance training and hypertrophy 
resulted. However, it is also likely that the reduction in aerobic capacity observed in Study 1 
may reflect a small but consistent difference in maximal ventilation measurement, because 
two slightly different methods were used at the different time points (Macfarlane, 2001). 
Douglas bags were used during the initial (Sydney) study, but in the second (Wellington) 
study, a turbine was used to measure ventilation; the use of which, has been shown by some to 
result in error when measuring ventilation (Yeh, Adams, Gardner, & Yanowitz, 1987). 
 
10.3  Implications 
The implication of my findings can be applied to public health studies, health interventions, 
and future physiological studies involving Māori. Firstly, public health studies often rely upon 
BMI, an indirect measure of fatness, to indicate risk of obesity-related illnesses such as 
Chapter 10  Conclusions 
 170
diabetes and CVD. The results of my study imply that BIA, which actually measures body 
fatness, though indirectly by way of ‘total body water’, can be utilized within a Māori 
population. On the other hand, BIA can significantly under/overestimate %BF in some 
people, so employing BIA for individual clinical assessment should be approached with some 
caution.  
 
Perhaps the most important contribution from this body of research is that improved physical 
fitness and a healthy body composition are associated with better metabolic health and a 
reduced risk of diabetes among young Māori men. The now proven association between 
aerobic fitness, in relation to lean body mass, and fasting insulin concentration in Māori, 
advocates the development of simple sub-maximal exercise tests to provide health promoters 
and physicians with an additional way to predict insulin sensitivity within Māori. However a 
VO2max test, as performed in these studies, may not be appropriate for the obese or elderly, 
so development of an appropriate alternative exercise test, which correlates well with aerobic 
capacity, may be more suitable. The association between aerobic capacity and insulin 
sensitivity also highlights the importance of increasing physical fitness among Māori who are 
most at risk of metabolic disorders. Thus, for Māori at least, monitoring ones fitness may be 
more important than just monitoring their weight.  
 
On the other hand, increasing lean body mass, through resistance training, allows maintenance 
of insulin sensitivity independent of aerobic fitness. Thus, resistance training may offer an 
alternative means of exercise for Māori, to maintain and improve glucose tolerance.  Yet, this 
means of physical activity has not been researched nor promoted in Māori to date, even 
though, anecdotally, it is well accepted in young Māori men.    
 
Chapter 10  Conclusions 
 171
The implications of fibre-type proportion upon health, also warrants further research in this 
area which has been limited to relatively small studies because of the invasive muscle 
biopsies required. Although the method I used to measure contractile properties was not a 
valid way to predict insulin sensitivity, further development of techniques may provide a 
method that does. The effectiveness of exercise in improving metabolic control is dependent 
upon the recruitment of fibres, so being able to identify fibre-type proportion within 
individuals, will allow exercise prescription which maximizes the effects of exercise specific 
to each individual. 
 
Importantly, the participation of Māori within this research was an innovation in itself. Māori 
health researchers have discussed the interface between ‘Western’ and ‘kaupapa Māori’ 
research methods, and have described ways to achieve this. Yet, physiological research 
involving Māori participants remains limited. The fact alone, that this research was conducted 
by a Māori exercise physiologist, is also a positive step toward desires of Māori to have 
leadership in issues relating to Māori; tying in with Māori health strategies and the principle 
of rangatiratanga. Additionally, I have sought to conduct my research at this interface 
according to principles discussed earlier (see chapter 3). As with any innovation however, 
there is no literature to compare this approach to, or measure the effectiveness of the 
methodology against. Nevertheless, the approach employed in this body of research may 
provide a methodological basis to be used and/or built upon for further scientific studies 
involving Māori. Thus, follow up with participants would be beneficial in assessing this 
approach and understanding the participants experience and view of this research. 
 
Chapter 10  Conclusions 
 172
10.4  Future Directions 
Physical activity, obesity and diabetes have a cyclic relationship (Hohepa, Schofield, & Kolt, 
2004). In other words, obesity is partly caused by physical inactivity but can also lead to 
further inactivity, as obese individuals may find it difficult to perform the physical tasks 
required to improve their body composition. Because of this, policy and interventions should 
focus on increasing awareness of physical activity levels among those who are still physically 
capable of making lifestyle changes. This falls inside the realm of prevention rather than 
treatment. Accordingly, SPARC have stated that efforts to increase physical activity should be 
focused on young people as well as two other groups; Māori and Pacific people (SPARC., 
2004).  
 
If Māori are the focus of policy efforts however, it is important that Māori are included in the 
development and implementation of policy. According to He Korowai Oranga (The Māori 
Health Strategy) this should be done at whānau (family), community and Iwi (tribal) levels for 
maximum Māori participation (Ministry of Health, 2002). District Health Boards (DHBs) and 
Public Health Organisations (PHOs) are also involved in the promotion of health care among 
Māori and the general population and should therefore play a part in policy making. In a study 
of diabetes among Canada’s Aboriginal people, Young et al. make the observation that non-
Aboriginal professionals, in this case non-Māori, involved in health care and policy making, 
need to understand how Māori interpret the experience of illness and respond to treatments 
prescribed, and to respect the logic of what may be to some a foreign system of thought 
(Young, Reading, Elias, & O'Neil, 2000). Training in cultural sensitivity and Māori systems 
of health is essential to develop policies that will achieve the outcomes of reducing the 
obvious health inequalities among Māori. Strategies such as He Korowai Oranga and 
Whakatātaka were developed as part of a framework for the public health sector to follow in 
Chapter 10  Conclusions 
 173
its support of Māori communities and whānau, the main goal of these strategies being the 
achievement of ‘whānau ora’; the realisation of maximum health and well-being among 
Māori families and communities. 
 
Future policy should also involve the reduction of environments that promote sedentary, 
energy saving lifestyles and the excessive consumption of high energy food. Obesity during 
childhood also adds to the risk of obesity continuing or worsening in adulthood (Guo, Roche, 
Chumlea, Gardner, & Siervogel, 1994), suggesting that policy should be targeted toward the 
prevention of obesity in children and adolescents. Indeed, healthy eating plans are currently 
being piloted in some NZ schools, as the MoH implement the Healthy Eating – Healthy 
Action plan (HEHA) (Ministry of Health, 2003). However, because children with obese 
parents are more likely to become obese adults themselves (Whitaker, Wright, Pepe, Seidel, & 
Dietz, 1997), interventions may be more effective when parents and family are involved. 
Because of the part whānau (family) plays in the achievement of hauora, this approach may 
also be more culturally appropriate for Māori. 
 
Along with policy changes, there are areas of research that need to be addressed. Firstly, 
because screening and assessment of health is often dependent upon measures such as BMI 
and waist circumference, Māori specific norms for these measurements should be established. 
Secondly, due to the importance of exercise in reducing obesity and risk of diabetes, 
identifying the most effective forms of exercise for Māori should be a priority. Although this 
should include investigation of the most effective forms of exercise in improving 
physiological markers associated with diabetes, it may be even more important to understand 
cultural, social and individual preferences to exercise. Without identifying preferred modes of 
Chapter 10  Conclusions 
 174
exercise, participation of ‘high-risk’ groups in exercise and physical activity will not likely 
improve. 
 
 
 References 
 175
References 
 
Abate, N., Garg, A., Peshock, R. M., Stray-Gundersen, J., & Grundy, S. M. (1995). 
Relationships of generalized and regional adiposity to insulin sensitivity in men. 
Journal of Clinical Investigation, 96(1), 88. 
Adams, G. R., Hather, B. M., Baldwin, K. M., & Dudley, G. A. (1993). Skeletal muscle 
myosin heavy chain composition and resistance training. Journal of Applied 
Physiology, 74(2), 911-915. 
Ahlborg, G., Felig, P., Hagenfeldt, L., Hendler, R., & Wahren, J. (1974). Substrate turnover 
during prolonged exercise in man: splanchnic and leg metabolism of glucose, free 
fatty acids, and amino acids. Journal of Clinical Investigation, 53(4), 1080. 
Alberti, K. G. M. M., Zimmet, P. Z., & World Health Organisation. (1998). Definition, 
diagnosis and classification of diabetes mellitus and its complications. Part 1: 
diagnosis and classification of diabetes mellitus. Provisional report of a WHO 
Consultation. Diabetic Medicine, 15(7), 539-553. 
Ama, P. F., Simoneau, J. A., Boulay, M. R., Serresse, O., Theriault, G., & Bouchard, C. 
(1986). Skeletal muscle characteristics in sedentary black and Caucasian males. 
Journal of Applied Physiology, 61(5), 1758-1761. 
Andersen, H. (2001). The history of reductionism versus holistic approaches to scientific 
research. Endeavour, 25(4), 153-156. 
Andersen, P., & Henriksson, J. (1977). Training induced changes in the subgroups of human 
type II skeletal muscle fibres. Acta Physiologica Scandinavica, 99, 123-125. 
Astrand, P. O., Rodahl, K., Dahl, H. A., & Stromme, S. B. (2003). Textbook of Work 
Physiology: Physiological Bases of Exercise (4th edition ed.). Champaign, Illinois: 
Human Kinetics. 
Astrup, A. (2001). Healthy lifestyles in Europe: prevention of obesity and type II diabetes by 
diet and physical activity. Public Health Nutrition, 4(2b), 499-515. 
Atkinson, M. A., & Maclaren, N. K. (1994). The pathogenesis of insulin-dependent diabetes 
mellitus. The New England Journal of Medicine, 331(21), 1428-1436. 
Balkau, B., Deanfield, J. E., Despres, J. P., Bassand, J. P., Fox, K. A. A., Smith Jr, S. C., et al. 
(2007). International day for the evaluation of abdominal obesity (IDEA): a study of 
waist circumference, cardiovascular disease, and diabetes mellitus in 168 000 primary 
care patients in 63 countries. Circulation, 116(17), 1942. 
 References 
 176
Barany, M. (1967). ATPase activity of myosin correlated with speed of muscle shortening. 
The Journal of General Physiology, 50(6), 197-218. 
Baron, A. D., Brechtel, G., Wallace, P., & Edelman, S. V. (1988). Rates and tissue sites of 
non-insulin and insulin-mediated glucose uptake in humans. American Journal of 
Physiology Endocrinology and Metabolism, 255(6), E769. 
Bassett Jr, D. R., & Howley, E. T. (2000). Limiting factors for maximum oxygen uptake and 
determinants of endurance performance. Medicine & Science in Sports & Exercise, 
32(1), 70. 
Bauman, A. (1997). Potential Health Benefits of Physical Activity in New Zealand. 
Wellington: Hillary Commission. 
Bauman, A., McLean, G., Hurdle, D., Walker, S., Boyd, J., van Aalst, I., et al. (2003). 
Evaluation of the national 'Push Play' campaign in New Zealand-creating population 
awareness of physical activity. The New Zealand Medical Journal, 116(1179), U535. 
Bell, D., McAuley, K. A., Mann, J., Murphy, E., & Williams, S. (2004). The 
hypertriglyceridaemic waist in New Zealand Māori. Asia Pacific Journal of Clinical 
Nutrition, 13(1), 74-77. 
Bergman, R. N., Kim, S. P., Catalano, K. J., Hsu, I. R., Chiu, J. D., Kabir, M., et al. (2006). 
Why visceral fat is bad: mechanisms of the metabolic syndrome. Obesity, 14, 16S-
19S. 
Bergman, R. N., & Mittelman, S. D. (1998). Central role of the adipocyte in insulin 
resistance. Journal of Basic and Clinical Physiology and Pharmacology, 9(2-4), 205. 
Bergström, J. (1975). Percutaneous needle biopsy of skeletal muscle in physiological and 
clinical research. Scandinavian Journal of Clinical & Laboratory Investigation, 35(7), 
609-616. 
Bergström, J., & Hultman, E. (1966). Muscle glycogen synthesis after exercise: an enhancing 
factor localized to the muscle cells in man. Nature, 210, 309-310. 
Billeter, R., Heizmann, C. W., Howald, H., & Jenny, E. (1981). Analysis of myosin light and 
heavy chain types in single human skeletal muscle fibers. European Journal of 
Biochemistry, 116(2), 389-395. 
Bishop, R. (1998). Freeing ourselves from neo-colonial domination in research: a Maori 
approach to creating knowledge. International Journal of Qualitative Studies in 
Education, 11(2), 199-219. 
Björntorp, P. (1991). Metabolic implications of body fat distribution. Diabetes care, 14(12), 
1132. 
 References 
 177
Blaak, E. E. (2007). Fatty acid metabolism in obesity and type 2 diabetes mellitus. 
Proceedings of the Nutrition Society, 62(03), 753-760. 
Blaak, E. E., Wagenmakers, A. J. M., Glatz, J. F. C., Wolffenbuttel, B. H. R., Kemerink, G. 
J., Langenberg, C. J. M., et al. (2000). Plasma FFA utilization and fatty acid-binding 
protein content are diminished in type 2 diabetic muscle. American Journal of 
Physiology- Endocrinology And Metabolism, 279(1), 146-154. 
Blair, D., Habicht, J., Sims, E., Sylwester, D., & Abraham, S. (1984). Evidence for an 
increased risk for hypertension with centrally located body fat and the effect of race 
and sex on this risk. The American Journal of Epidemiology, 119(4), 526-540. 
Blimkie, C., Sale, D., & Bar-Or, O. (1990). Voluntary strength, evoked twitch contractile 
properties and motor unit activation of knee extensors in obese and non-obese 
adolescent males. European Journal of Applied Physiology, 61, 313-318. 
Blomstrand, E., & Ekblom, B. (1982). The needle biopsy technique for fibre type 
determination in human skeletal muscle-a methodological study. Acta Physiologica 
Scandinavica, 116(4), 437-442. 
Boden, G. (1997). Role of fatty acids in the pathogenesis of insulin resistance and NIDDM. 
Diabetes, 46(1), 3-10. 
Boden, G., Chen, X., DeSantis, R. A., & Kendrick, Z. (1993). Effects of age and body fat on 
insulin resistance in healthy men. Diabetes Care, 16(5), 728. 
Boden, G., Chen, X., Rosner, J., & Barton, M. (1995). Effects of a 48-h fat infusion on insulin 
secretion and glucose utilization. Diabetes, 44(10), 1239-1242. 
Boden, G., Chen, X., Ruiz, J., White, J. V., & Rossetti, L. (1994). Mechanisms of fatty acid-
induced inhibition of glucose uptake. Journal of Clinical Investigation, 93(6), 2438. 
Boden, G., & Shulman, G. I. (2002). Free fatty acids in obesity and type2 diabetes: defining 
their role in the development of insulin resistance and beta-cell dysfunction. European 
Journal of Clinical Investigation, 32(s3), 14-23. 
Bogardus, C., Lillioja, S., Mott, D. M., Hollenbeck, C., & Reaven, G. (1985). Relationship 
between degree of obesity and in vivo insulin action in man. American Journal of 
Physiology Endocrinology and Metabolism, 248(3), E286-291. 
Booth, F. W., Gordon, S. E., Carlson, C. J., & Hamilton, M. T. (2000). Waging war on 
modern chronic diseases: primary prevention through exercise biology. Journal of 
Applied Physiology, 88(2), 774-787. 
 References 
 178
Bottinelli, R., Betto, R., Schiaffino, S., & Reggiani, C. (1994). Unloaded shortening velocity 
and myosin heavy chain and alkali light chain isoform composition in rat skeletal 
muscle fibres. Journal of Physiology, 478(2), 341-350. 
Bramley, D., Hebert, P., Tuzzio, L., & Chassin, M. (2005). Disparities in indigenous health: a 
cross-country comparison between New Zealand and the United States. American 
Journal of Public Health, 95(5), 844-850. 
Breda, E., Cavaghan, M. K., Toffolo, G., Polonsky, K. S., & Cobelli, C. (2001). Oral glucose 
tolerance test minimal model indexes of -cell function and insulin sensitivity. 
Diabetes, 50(1), 150. 
Brooke, M., & Kaiser, K. (1970). Three “myosin adenosine triphosphatase” systems: The 
nature of their pH lability and sulphydryl dependence. Journal of Histochemistry and 
Cytochemistry, 8, 670-672. 
Brooks, G. A., & Mercier, J. (1994). Balance of carbohydrate and lipid utilization during 
exercise: the "crossover" concept. Journal of Applied Physiology, 76(6), 2253-2261. 
Brozek, J., Grande, F., Anderson, J. T., & Keys, A. (1963). Densitometric analysis of body 
composition revision of some quantitative assumptions*. Annals of the New York 
Academy of Sciences, 110(1), 113-140. 
Bruce, C. R., Anderson, M. J., Carey, A. L., Newman, D. G., Bonen, A., Kriketos, A. D., et 
al. (2003). Muscle oxidative capacity Is a better predictor of insulin sensitivity than 
lipid status. Journal of Clinical Endocrinology and Metabolism, 88(11), 5444-5451. 
Bruce, R. A. (1971). Exercise testing of patients with coronary heart disease. Principles and 
normal standards for evaluation. Annals of clinical research, 3(6), 323. 
Bryant, N. J., Govers, R., & James, D. E. (2002). Regulated transport of the glucose 
transporter GLUT4. Nature Reviews Molecular Cell Biology, 3(4), 267-277. 
Buller, A., Eccles, J., & Eccles, R. (1960). Interactions between motor neurons and muscles in 
respect of the characteristic speeds of their responses. The Journal of Physiology, 
150(41), 7-439. 
Burke, R. E., Levine, D. N., Zajac Iii, F. E., Tsairis, P., & Engel, W. K. (1971). Mammalian 
motor units: physiological-histochemical correlation in three types in cat 
gastrocnemius. Science, 174(4010), 709. 
Burnham, R., Martin, T., Stein, R., Bell, G., MacLean, I., & Steadward, R. (1997). Skeletal 
muscle fibre type transformation following spinal cord injury. Spinal cord: the official 
journal of the International Medical Society of Paraplegia, 35(2), 86. 
 References 
 179
Burstein, R., Epstein, Y., Shapiro, Y., Charuzi, I., & Karnieli, E. (1990). Effect of an acute 
bout of exercise on glucose disposal in human obesity. Journal of Applied Physiology, 
69(1), 299-304. 
Buskirk, E. R. (1959). Underwater weighing and body density: a review of procedures. In J. 
Brozek & A. Henschel (Eds.), Techniques for Measuring Body Composition (pp. 90). 
Washington DC: National Academy of Sciences-National Research Council. 
Buskirk, E. R., & Hodgson, J. L. (1987). Age and aerobic power: the rate of change in men 
and women. Paper presented at the Federation Proceedings. 
Caldwell, S. H., Ikura, Y., Iezzoni, J. C., & Liu, Z. (2007). Has natural selection in human 
populations produced two types of metabolic syndrome (with and without fatty liver)? 
Journal of gastroenterology and hepatology, 22(s1 International Symposium on 
Energy Metabolism and Oxidative Stress in Liver Pathophysiology), 11-19. 
Calle, E. E., Thun, M. J., Petrelli, J. M., Rodriguez, C., & Heath, C. W. (1999). Body-mass 
index and mortality in a prospective cohort of US adults. The New England journal of 
medicine, 341(15), 1097. 
Campbell, P. J., Carlson, M. G., Hill, J. O., & Nurjhan, N. (2006). Regulation of free fatty 
acid metabolism by insulin in humans: role of lipolysis and reesterification. American 
Journal of Physiology- Endocrinology And Metabolism, 263(6), E1063. 
Cardinal, B. J., Esters, J., & Cardinal, M. K. (1996). Evaluation of the Revised Physical 
Activity Readiness Questionnaire in older adults. Medicine & Science in Sports & 
Exercise, 28(4), 468. 
Caton, J. R., MolÉ, P. A., Adams, W. C., & Heustis, D. S. (1988). Body composition analysis 
by bioelectrical impedance: effect of skin temperature. Medicine & Science in Sports 
& Exercise, 20(5), 489. 
Caumo, A., Bergman, R. N., & Cobelli, C. (2000). Insulin sensitivity from meal tolerance 
tests in normal subjects: a minimal model index. Journal of Clinical Endocrinology & 
Metabolism, 85(11), 4396. 
Cauza, E., Hanusch-Enserer, U., Strasser, B., Ludvik, B., Metz-Schimmerl, S., Pacini, G., et 
al. (2005). The relative benefits of endurance and strength training on the metabolic 
factors and muscle function of people with type 2 diabetes mellitus. Archives of 
Physical Medicine and Rehabilitation, 86(8), 1527-1533. 
Chadwick, P., & Palmer, F. (2006). Tama tu, tama ora. tama noho, tama mate–central. A 
report on the health and physical activity status of a sample of Ngati Raukawa, 
Muaupoko and Rangitaane iwi. 
 References 
 180
Clarys, J. P., Martin, A. D., & Drinkwater, D. T. (1984). Gross tissue weights in the human 
body by cadaver dissection. Human biology; an international record of research, 
56(3), 459. 
Coggan, A. R., & Coyle, E. F. (1987). Reversal of fatigue during prolonged exercise by 
carbohydrate infusion or ingestion. Journal of Applied Physiology, 63(6), 2388. 
Cohn, S. (1987). New concepts of body composition. In Vivo Body Composition Studies The 
Institute of Physical Sciences in medicine: London, 1-14. 
Conley, K. E., Jubrias, S. A., & Esselman, P. C. (2000). Oxidative capacity and ageing in 
human muscle. The Journal of Physiology, 526(1), 203-210. 
Coon, P. J., Bleecker, E. R., Drinkwater, D. T., Meyers, D. A., & Goldberg, A. P. (1989). 
Effects of body composition and exercise capacity on glucose tolerance, insulin, and 
lipoprotein lipids in healthy older men: a cross-sectional and longitudinal intervention 
study. Metabolism, 38(12), 1201-1209. 
Cornish, B. H., Thomas, B. J., & Ward, L. C. (1998). Effect of temperature and sweating on 
bioimpedance measurements. Applied Radiation and Isotopes, 49(5-6), 475-476. 
Cortez, M. Y., Torgan, C. E., Brozinick, J. T., Jr., & Ivy, J. L. (1991). Insulin resistance of 
obese Zucker rats exercise trained at two different intensities. The American Journal 
of Physiology Endocrinology and Metabolism, 261(5), E613-619. 
Costill, D. L., Daniels, J., Evans, W., Fink, W., Krahenbuhl, G., & Saltin, B. (1976). Skeletal 
muscle enzymes and fiber composition in male and female track athletes. Journal of 
Applied Physiology, 40(2), 149. 
Cowie, C. C., Rust, K. F., Byrd-Holt, D. D., Eberhardt, M. S., Flegal, K. M., Engelgau, M. 
M., et al. (2006). Prevalence of diabetes and impaired fasting glucose in adults in the 
US population. Diabetes Care, 29(6), 1263. 
Coyle, E. F., Hagberg, J. M., Hurley, B. F., Martin, W. H., Ehsani, A. A., & Holloszy, J. O. 
(1983). Carbohydrate feeding during prolonged strenuous exercise can delay fatigue. 
Journal of Applied Physiology, 55(1), 230. 
Crabtree, B., & Garlick, P. J. (1993). Fuels of tissues. In J. S. Garrow, A. Ralph & W. P. T. 
James (Eds.), Human Nutrition and Dietetics (9th ed.). New York: Churchill 
Livingston. 
Cram, F., Smith, L., & Johnstone, W. (2003). Mapping the themes of Maori talk about health. 
Journal of the New Zealand Medical Association, 116(1170). 
Crespin, S. R., Greenough Iii, W. B., & Steinberg, D. (1969). Stimulation of insulin secretion 
by infusion of free fatty acids. Journal of Clinical Investigation, 48(10), 1934. 
 References 
 181
Dannenbaum, D., Kuzmina, E., Lejeune, P., Torrie, J., & Gangbe, M. (2008). Prevalence of 
diabetes and diabetes-related complications in First Nations communities in Northern 
Quebec (Eeyou Istchee), Canada. Can J Diabetes, 32(1), 46-52. 
Daugaard, J. R., Nielsen, J. N., Kristiansen, S., Andersen, J. L., Hargreaves, M., & Richter, E. 
A. (2000). Fiber type-specific expression of GLUT4 in human skeletal muscle: 
influence of exercise training. Diabetes, 49(7), 1092-1095. 
Daugaard, J. R., & Richter, E. A. (2001). Relationship between muscle fibre composition, 
glucose transporter protein 4 and exercise training: possible consequences in non-
insulin-dependent diabetes mellitus. Acta Physiologica Scandinavica, 171(3), 267-
276. 
DeFronzo, R. A. (1979). Glucose clamp technique: a method for quantifying insulin secretion 
and resistance. American Journal of Physiology Endocrinology and Metabolism, 
237(3), 214-223. 
DeFronzo, R. A., Sherwin, R. S., & Kraemer, N. (1987). Effect of physical training on insulin 
action in obesity. Diabetes, 36(12), 1379-1385. 
Delitto, A., & Snyder-Mackler, L. (1990). Two theories of muscle strength augmentation 
using percutaneous electrical stimulation. Physical Therapy, 70(3), 158-164. 
Dengel, D. R., Pratley, R. E., Hagberg, J. M., Rogus, E. M., & Goldberg, A. P. (1996). 
Distinct effects of aerobic exercise training and weight loss on glucose homeostasis in 
obese sedentary men. Journal of Applied Physiology, 81(1), 318-325. 
Deurenberg, P., Deurenberg-Yap, M., & Schouten, F. J. M. (2002). Validity of total and 
segmental impedance measurements for prediction of body composition across ethnic 
population groups. European Journal of Clinical Nutrition, 56, 214-220. 
Deurenberg, P., Weststrate, A., J., & Seidell, J. C. (1991). Body mass index as a measure of 
body fatness: Age- and sex- specific prediction formulas. British Journal of Nutrition, 
65, 105-114. 
Deurenberg, P., Weststrate, J. A., Paymans, I., & Van Der Kooy, K. (1988). Factors affecting 
bioelectrical impedance measurements in humans. European Journal of Clinical 
Nutrition, 42(12), 1017. 
Deurenberg, P., Yap, M., & van Staveren, W. (1998). Body mass index and percent body fat: 
a meta analysis among different ethnic groups. International Journal of Obesity, 
22(12), 1164-1171  
Diamond, J. (2003). The double puzzle of diabetes. Nature, 423, 599-602. 
 References 
 182
Dietz, W. H., & Bellizzi, M. C. (1999). Introduction: the use of body mass index to assess 
obesity in children. The American Journal of Clinical Nutrition, 70(1), 123S-125. 
Doggrell, S. A. (2006). Muraglitazar: beneficial or detrimental in the treatment of type 2 
diabetes? Expert Opinion on Pharmacotherapy, 7(9), 1229-1233. 
Dresner, A., Laurent, D., Marcucci, M., Griffin, M. E., Dufour, S., Cline, G. W., et al. (1999). 
Effects of free fatty acids on glucose transport and IRS-1 - associated 
phosphatidylinositol 3-kinase activity. The Journal of Clinical Investigation, 103(2), 
253-259. 
Duncan, G. E., Perri, M. G., Theriaque, D. W., Hutson, A. D., Eckel, R. H., & Stacpoole, P. 
W. (2003). Exercise Training, Without Weight Loss, Increases Insulin Sensitivity and 
Postheparin Plasma Lipase Activity in Previously Sedentary Adults. Diabetes Care, 
26(3), 557-562. 
Durie, M. (1994). Whaiora: Maori Health Development. Auckland: Oxford University Press. 
Durie, M. (2004). Understanding health and illness: research at the interface between science 
and indigenous knowledge. International Journal of Epidemiology, 33(5), 1138-1143. 
Durie, M. H. (1985). A Maori perspective of health. Social Science and Medicine, 20(5), 483-
486. 
Durie, M. H., & Kingi, T. K. R. (1998). A framework for measuring Maori mental health 
outcomes. Department of Maori Studies, Te Pumanawa Hauora, Massey University, 
Palmerston North. 
Eaton, S. B., & Konner, M. (1985). Paleolithic nutrition. A consideration of its nature and 
current implications. New England Journal of Medicine, 312(5), 283-289. 
Edström, L., & Kugelberg, E. (1968). Histochemical composition, distribution of fibres and 
fatiguability of single motor units. Anterior tibial muscle of the rat. Journal of 
Neurology, Neurosurgery, and Psychiatry, 31(5), 424. 
Ellis, B. A., Poynten, A., Lowy, A. J., Furler, S. M., Chisholm, D. J., Kraegen, E. W., et al. 
(2000). Long-chain acyl-CoA esters as indicators of lipid metabolism and insulin 
sensitivity in rat and human muscle. The American Journal of Physiology 
Endocrinology and Metabolism, 279(3), E554-560. 
Enoka, R. M., & Stuart, D. G. (1984). Henneman's size principle': current issues. Trends in 
Neurosciences, 7(7), 226-228. 
Eriksson, J., Taimela, S., & Koivisto, V. A. (1997). Exercise and the metabolic syndrome. 
Diabetologia, 40(2), 125-135. 
 References 
 183
Eriksson, K. F., & Lindgärde, F. (1991). Prevention of type 2 (non-insulin-dependent) 
diabetes mellitus by diet and physical exercise The 6-year Malmö feasibility study. 
Diabetologia, 34(12), 891-898. 
Esmaillzadeh, A., Kimiagar, M., Mehrabi, Y., Azadbakht, L., Hu, F. B., & Willett, W. C. 
(2007). Dietary patterns, insulin resistance, and prevalence of the metabolic syndrome 
in women. American Journal of Clinical Nutrition, 85(3), 910-918. 
Essén, B., Jansson, E., Henriksson, J., Taylor, A. W., & Saltin, B. (1975). Metabolic 
characteristics of fibre types in human skeletal muscle. Acta Physiologica 
Scandinavica, 95(2), 153-165. 
Estabrooks, P. A., Lee, R. E., & Gyurcsik, N. C. (2003). Resources for physical activity 
participation: does availability and accessibility differ by neighborhood 
socioeconomic status? Annals of Behavioral Medicine, 25(2), 100-104. 
Evas, N. D., & Plotnikoff, R. C. (2006). Resistance training and type 2 diabetes. Diabetes 
Care, 29, 1933-1941. 
Fabbrini, E., Magkos, F., Mohammed, B. S., Pietka, T., Abumrad, N. A., Patterson, B. W., et 
al. (2009). Intrahepatic fat, not visceral fat, is linked with metabolic complications of 
obesity. Proceedings of the National Academy of Sciences, 106(36), 15430-15435. 
Febbraio, M. A., Snow, R. J., Stathis, C. G., Hargreaves, M., & Carey, M. F. (1994). Effect of 
heat stress on muscle energy metabolism during exercise. Journal of Applied 
Physiology, 77(6), 2827. 
Fell, R. D., Terblanche, S. E., Ivy, J. L., Young, J. C., & Holloszy, J. O. (1982). Effect of 
muscle glycogen content on glucose uptake following exercise. Journal of Applied 
Physiology, 52(2), 434-437. 
Ferrannini, E., Barrett, E. J., Bevilacqua, S., & DeFronzo, R. A. (1983). Effect of fatty acids 
on glucose production and utilization in man. Journal of Clinical Investigation, 72(5), 
1737. 
Ferrannini, E., & Mari, A. (1998). How to measure insulin sensitivity. Journal of 
Hypertension, 16(7), 895. 
Fink, R. I., Kolterman, O. G., Griffin, J., & Olefsky, J. M. (1983). Mechanisms of insulin 
resistance in aging. Journal of Clinical Investigation, 71(6), 1523. 
Flatt, J. P. (1995). Use and storage of carbohydrate and fat. American Journal of Clinical 
Nutrition, 61(4), 952S. 
Forman, J. N., Miller, W. C., Szymanski, L. M., & Fernhall, B. (1998). Differences in resting 
metabolic rates of inactive obese African-American and Caucasian women. 
 References 
 184
International journal of obesity and related metabolic disorders: journal of the 
International Association for the Study of Obesity (USA). 
Fox, K. R. (2007). The influence of physical activity on mental well-being. Public Health 
Nutrition, 2(3a), 411-418. 
Fridlyand, L. E., & Philipson, L. H. (2006). Cold climate genes and the prevalence of type 2 
diabetes mellitus. Medical Hypotheses, 67(5), 1034-1041. 
Gale, E. A. M. (2001). The discovery of type 1 diabetes. Diabetes, 50(2), 217-226. 
Gallagher, D., Heymsfield, S. B., Heo, M., Jebb, S. A., Murgatroyd, P. R., & Sakamoto, Y. 
(2000). Healthy percentage body fat ranges: an approach for developing guidelines 
based on body mass index. American Journal of Clinical Nutrition, 72(3), 694-701. 
Gallagher, D., Visser, M., Sepulveda, D., Pierson, R. N., Harris, T., & Heymsfield, S. B. 
(1996). How useful is body mass index for comparison of body fatness across age, 
sex, and ethnic groups? The American Journal of Epidemiology, 143(3), 228-239. 
Gallagher, M. R., Walker, K. Z., & O'Dea, K. (1998). The influence of a breakfast meal on 
the assessment of body composition using bioelectrical impedance. European journal 
of clinical nutrition, 52(2), 94-97. 
Gardner, L. I., Stern, M. P., Haffner, S. M., Gaskill, S. P., Hazuda, H. P., Relethford, J. H., et 
al. (1984). Prevalence of diabetes in Mexican Americans. Relationship to percent of 
gene pool derived from native American sources. Diabetes, 33(1), 86-92. 
Garvey, W. T., Maianu, L., Zhu, J. H., Brechtel-Hook, G., Wallace, P., & Baron, A. D. 
(1998). Evidence for defects in the trafficking and translocation of GLUT4 glucose 
transporters in skeletal muscle as a cause of human insulin resistance. Journal of 
Clinical Investigation, 101(11), 2377. 
Gittelsohn, J., Wolever, T. M. S., Harris, S. B., Harris-Giraldo, R., Hanley, A. J. G., & 
Zinman, B. (1998). Specific patterns of food consumption and preparation are 
associated with diabetes and obesity in a native Canadian community. The Journal of 
Nutrition, 128(3), 541-547. 
Gohdes, D. Diabetes in North American Indians and Alaska Natives. Atlantic, 43, 87. 
Golay, A., Felber, J. P., Dusmet, M., Gomez, F., & Curchod, B. (1985). Effect of weight loss 
on glucose disposal in obese and obese diabetic patients. International Journal of 
Obesity, 9(3), 181. 
Gollnick, P. D., Armstrong, R. B., Saltin, B., Saubert, C. W. I., Sembrowich, W. L., & 
Shepherd, R. E. (1973). Effect of training on enzyme activity and fiber composition of 
human skeletal muscle. Journal of Applied Physiology, 34(1), 107-111. 
 References 
 185
Gollnick, P. D., Armstrong, R. B., Saubert, C. W. t., Piehl, K., & Saltin, B. (1972). Enzyme 
activity and fiber composition in skeletal muscle of untrained and trained men. 
Journal of Applied Physiology, 33(3), 312-319. 
González-Alonso, J., & Calbet, J. A. L. (2003). Reductions in systemic and skeletal muscle 
blood flow and oxygen delivery limit maximal aerobic capacity in humans. American 
Heart Association, 107(6), 824-830. 
Goodman, A. H. (2000). Why genes don't count (for racial differences in health). American 
Journal of Public Health, 90(11), 1699. 
Goodpaster, B. H., He, J., Watkins, S., & Kelley, D. E. (2001). Skeletal muscle lipid content 
and insulin resistance: evidence for a paradox in endurance-trained athletes. Journal of 
Clinical Endocrinology & Metabolism, 86(12), 5755-5761. 
Goodpaster, B. H., Katsiaras, A., & Kelley, D. E. (2003). Enhanced fat oxidation through 
physical activity is associated with improvements in insulin sensitivity in obesity. 
Diabetes, 52(9), 2191-2197. 
Goodpaster, B. H., Thaete, F. L., Simoneau, J.-A., & Kelley, D. E. (1997). Subcutaneous 
abdominal fat and thigh muscle composition predict insulin sensitivity independently 
of visceral fat. Diabetes, v46(n10), p1579(1577). 
Goodpaster, B. H., Theriault, R., Watkins, S. C., & Kelley, D. E. (2000). Intramuscular lipid 
content is increased in obesity and decreased by weight loss. Metabolism, 49(4), 467-
472. 
Goodpaster, B. H., Wolfe, R. R., & Kelley, D. E. (2002). Effects of obesity on substrate 
utilization during exercise. Obesity Research, 10(7), 575-584. 
Goodyear, P. L. J., & Kahn, M. D. B. B. (1998). Exercise, glucose transport and insulin 
sensitivity. Annual Review of Medicine, 49(1), 235-261. 
Graham, P., Jackson, R., Beaglehole, R., & De Boer, G. (1989). The validity of Māori 
mortality statistics. The New Zealand medical journal, 102(864), 124. 
Grandinetti, A., Kaholokula, J. K., Theriault, A. G., Mor, J. M., Chang, H. K., & Waslien, C. 
(2007). Prevalence of diabetes and glucose intolerance in an ethnically diverse rural 
community of Hawaii. Ethnicity & Disease, 17(2), 250-255. 
Greco, A. V., Mingrone, G., Giancaterini, A., Manco, M., Morroni, M., Cinti, S., et al. (2002). 
Insulin resistance in morbid obesity: reversal with intramyocellular fat depletion. 
Diabetes, 51(1), 144-151. 
 References 
 186
Guo, S. S., Roche, A. F., Chumlea, W. C., Gardner, J. D., & Siervogel, R. M. (1994). The 
predictive value of childhood body mass index values for overweight at age 35 y. 
American Journal of Clinical Nutrition, 59(4), 810-819. 
Hajduch, E., Balendran, A., Batty, I. H., Litherland, G. J., Blair, A. S., Downes, C. P., et al. 
(2001). Ceramide impairs the insulin-dependent membrane recruitment of protein 
kinase B leading to a loss in downstream signalling in L6 skeletal muscle cells. 
Diabetologia, 44(2), 173-183. 
Hamman, R. F., Wing, R. R., Edelstein, S. L., Lachin, J. M., Bray, G. A., Delahanty, L., et al. 
(2006). Effect of weight loss with lifestyle intervention on risk of diabetes. Diabetes 
Care, 29(9), 2102. 
Harridge, S. D. R., Bottinelli, R., Canepari, M., Pellegrino, M. A., Reggiani, C., Esbjörnsson, 
M., et al. (1996). Whole-muscle and single-fibre contractile properties and myosin 
heavy chain isoforms in humans. Pflügers Archiv European Journal of Physiology, 
432(5), 913-920. 
Harris, P., & Mercier, O. (2006). Te Ara Putaiao o nga tupuna, o nga mokopuna: science 
education and research. In M. Mulholland (Ed.), State of the Maori Nation (pp. 141-
155). Auckland: Reed Books. 
Hawkins, S. A., & Wiswell, R. A. (2003). Rate and mechanism of maximal oxygen 
consumption decline with aging: implications for exercise training. Sports Medicine, 
33(12), 877. 
Hawley, J. (2001). The fuels for exercise. Australian Journal of Nutrition and Dietetics, 
58(1), 19-22. 
He, J., Watkins, S., & Kelley, D. E. (2001). Skeletal muscle lipid content and oxidative 
enzyme activity in relation to muscle fiber type in type 2 diabetes and obesity. 
Diabetes, 50(4), 817-823. 
Heath, E. M., Adams, T. D., Daines, M. M., & Hunt, S. C. (1998). Bioelectric impedance and 
hydrostatic weighing with and without head submersion in persons who are morbidly 
obese. Journal of the American Dietetic Association, 98(8), 869-875. 
Hegarty, B. D., Furler, S. M., Ye, J., Cooney, G. J., & Kraegen, E. W. (2003). The role of 
intramuscular lipid in insulin resistance. Acta Physiologica Scandinavica, 178(4), 373-
383. 
Henriksen, E. (2002). Exercise effects of muscle insulin signaling and action: invited review: 
effects of acute exercise and exercise training on insulin resistance. Journal of Applied 
Physiology, 93(2), 788-796. 
 References 
 187
Henriksen, E. J., Bourey, R. E., Rodnick, K. J., Koranyi, L., Permutt, M. A., & Holloszy, J. O. 
(1990). Glucose transporter protein content and glucose transport capacity in rat 
skeletal muscles. American Journal of Physiology Endocrinology and Metabolism, 
259(4), 593-598. 
Herman, W. H., Hoerger, T. J., Brandle, M., Hicks, K., Sorensen, S., Zhang, P., et al. (2005). 
The cost-effectiveness of lifestyle modification or metformin in preventing type 2 
diabetes in adults with impaired glucose tolerance. Annals of Internal Medicine, 
142(5), 323-332. 
Heymsfield, S. B., Wang, J., Heshka, S., Kehayias, J. J., & Pierson, R. N. (1989). Dual-
photon absorptiometry: comparison of bone mineral and soft tissue mass 
measurements in vivo with established methods. American Journal of Clinical 
Nutrition, 49(6), 1283. 
Heyward, V. H., & Wagner, D. R. (2004). Applied body composition assessment (2 ed.). 
Champaign, Illinois: Human Kinetics. 
Hickey, M. S., Carey, J. O., Azevedo, J. L., Houmard, J. A., Pories, W. J., Israel, R. G., et al. 
(1995). Skeletal muscle fiber composition is related to adiposity and in vitro glucose 
transport rate in humans. The American Journal of Physiology Endocrinology and 
Metabolism, 268(3), E453-457. 
Hickey, M. S., Weidner, M. D., Gavigan, K. E., Zheng, D., Tyndall, G. L., & Houmard, J. A. 
(1995). The insulin action-fiber type relationship in humans is muscle group specific. 
The American Journal of Physiology Endocrinology and Metabolism, 269(1), E150-
154. 
Himes, J. H., & Dietz, W. H. (1994). Guidelines for overweight in adolescent preventive 
services: recommendations from an expert committee. The Expert Committee on 
Clinical Guidelines for Overweight in Adolescent Preventive Services. American 
Journal of Clinical Nutrition, 59(2), 307-316. 
Hohepa, M., Schofield, G., & Kolt, G. (2004). Adolescent obesity and physical inactivity. The 
New Zealand Medical Journal, 117(1207), U1210. 
Hollenbeck, C., & Reaven, G. M. (1987). Variations in insulin-stimulated glucose uptake in 
healthy individuals with normal glucose tolerance. Journal of Clinical Endocrinology 
& Metabolism, 64(6), 1169-1173. 
Holloszy, J. O. (1967). Biochemical adaptations in muscle. Effects of exercise on 
mitochondrial oxygen uptake and respiratory enzyme activity in skeletal muscle. 
Journal of Biological Chemistry, 242(9), 2278. 
 References 
 188
Holloszy, J. O., & Booth, F. W. (1976). Biochemical adaptations to endurance exercise in 
muscle. Annual Review of Physiology, 38(1), 273-291. 
Holloszy, J. O., & Coyle, E. F. (1984). Adaptations of skeletal muscle to endurance exercise 
and their metabolic consequences. Journal of Applied Physiology, 56(4), 831-838. 
Holten, M. K., Zacho, M., Gaster, M., Juel, C., Wojtaszewski, J. F. P., & Dela, F. (2004). 
Strength training increases insulin-mediated glucose uptake, GLUT4 content, and 
insulin signaling in skeletal muscle in patients with type 2 diabetes. Diabetes, 53(2), 
294-305. 
Hoppeler, H., Howald, H., Conley, K., Lindstedt, S. L., Claassen, H., Vock, P., et al. (1985). 
Endurance training in humans: aerobic capacity and structure of skeletal muscle. 
Journal of Applied Physiology, 59(2), 320-327. 
Host, H. H., Hansen, P. A., Nolte, L. A., Chen, M. M., & Holloszy, J. O. (1998). Rapid 
reversal of adaptive increases in muscle GLUT-4 and glucose transport capacity after 
training cessation. Journal of Applied Physiology, 84(3), 798-802. 
Hotu, S., Carter, B., Watson, P. D., Cutfield, W. S., & Cundy, T. (2004). Increasing 
prevalence of type 2 diabetes in adolescents. Journal of Paediatrics and Child Health, 
40(4), 201-204. 
Houmard, J. A., Shinebarger, M. H., Dolan, P. L., Leggett-Frazier, N., Bruner, R. K., 
McCammon, M. R., et al. (1993). Exercise training increases GLUT-4 protein 
concentration in previously sedentary middle-aged men. The American Journal of 
Physiology Endocrinology and Metabolism, 264(6), E896-901. 
Hu, F. B., Leitzmann, M. F., Stampfer, M. J., Colditz, G. A., Willett, W. C., & Rimm, E. B. 
(2001). Physical activity and television watching in relation to risk for type 2 diabetes 
mellitus in men. Archives of Internal Medicine, 161(12), 1542-1548. 
Hu, F. B., Li, T. Y., Colditz, G. A., Willett, W. C., & Manson, J. E. (2003). Television 
watching and other sedentary behaviors in relation to risk of obesity and type 2 
diabetes mellitus in women. Journal of the American Medical Association, 289(14), 
1785-1791. 
Hui Whakaoranga. (1984, 19-22 March 1984). Paper presented at the National Maori Health 
Conference, Hone Waititi Marae, Auckland. 
Hunter, D. W. (1998). Race and Athletic Performance: A Physiological Review. African 
Americans in sport: contemporary themes, 85. 
HWAC. (2006). Fit for purpose and for practice: a review of the medical workforce in New 
Zealand. Wellington, NZ: Health Workforce Advisory Committee. 
 References 
 189
Ibanez, J., Izquierdo, M., Arguelles, I., Forga, L., Larrion, J. L., Garcia-Unciti, M., et al. 
(2005). Twice-weekly progressive resistance training decreases abdominal fat and 
improves insulin sensitivity in older men with type 2 diabetes. Diabetes Care, 28(3), 
662-667. 
Ingjer, F. (1979). Capillary supply and mitochondrial content of different skeletal muscle fiber 
types in untrained and endurance-trained men. A histochemical and ultrastructural 
study. European Journal of Applied Physiology and Occupational Physiology, 40(3), 
197-209. 
Ivy, J. L. (1997). Role of exercise training in the prevention and treatment of insulin 
resistance and non-insulin-dependent diabetes mellitus. Sports Medicine, 24(5), 321-
336. 
Ivy, J. L., Withers, R. T., Brose, G., Maxwell, B. D., & Costill, D. L. (1981). Isokinetic 
contractile properties of the quadriceps with relation to fiber type. European Journal 
of Applied Physiology, 47(3), 247-255. 
Jacob, S., Machann, J., Rett, K., Brechtel, K., Volk, A., Renn, W., et al. (1999). Association 
of increased intramyocellular lipid content with insulin resistance in lean nondiabetic 
offspring of type 2 diabetic subjects. Diabetes, 48(5), 1113-1119. 
Jakicic, J. M., Wing, R. R., & Lang, W. (1998). Bioelectrical impedance analysis to assess 
body composition in obese adult women: the effect of ethnicity. International journal 
of obesity, 22(3), 243-249. 
Janssen, I., Katzmarzyk, P. T., & Ross, R. (2004). Waist circumference and not body mass 
index explains obesity-related health risk. American Journal of Clinical Nutrition, 
79(3), 379-384. 
Johnson, N. A., Stannard, S. R., Mehalski, K., Trenell, M. I., Sachinwalla, T., Thompson, C. 
H., et al. (2003). Intramyocellular triacylglycerol in prolonged cycling with high- and 
low-carbohydrate availability. Journal of Applied Physiology, 94(4), 1365-1372. 
Johnson, N. A., Stannard, S. R., Rowlands, D. S., Chapman, P. G., Thompson, C. H., 
O'Connor, H., et al. (2006). Effect of short-term starvation versus high-fat diet on 
intramyocellular triglyceride accumulation and insulin resistance in physically fit men. 
Experimental Physiology, 91(4), 693-703. 
Joshy, G., & Simmons, D. (2006). Epidemiology of diabetes in New Zealand: revisit to a 
changing landscape. The New Zealand Medical Journal, 119(1235), U1999. 
Joyner, M. J., & Saltin, B. (2008). Exercise physiology and human performance: systems 
biology before systems biology! The Journal of Physiology, 586(1), 9-9. 
 References 
 190
Kahn, B. B., & Flier, J. S. (2000). Obesity and insulin resistance. Journal of Clinical 
Investigation, 106(4), 473-481. 
Kahn, S. E., Prigeon, R. L., Schwartz, R. S., Fujimoto, W. Y., Knopp, R. H., Brunzell, J. D., 
et al. (2001). Obesity, body fat distribution, insulin sensitivity and islet beta-cell 
function as explanations for metabolic diversity. The Journal of Nutrition, 131(2), 
354S-360. 
Kame'eleihiwa, L. (1992). Native land and foreign desires: how shall we live in harmony? 
Honolulu, Hawaii: Bishop Museum Press. 
Kasa-Vubu, J. Z., Lee, C. C., Rosenthal, A., Singer, K., & Halter, J. B. (2005). Cardiovascular 
fitness and exercise as determinants of insulin resistance in postpubertal adolescent 
females. Journal of Clinical Endocrinology & Metabolism, 90(2), 849-854. 
Kassirer, J. P., & Angell, M. (1998). Losing weight-an ill-fated new year's resolution. New 
England Journal of Medicine, 338(1), 52. 
Katzmarzyk, P. T., Church, T. S., Janssen, I., Ross, R., & Blair, S. N. (2005). Metabolic 
syndrome, obesity, and mortality: impact of cardiorespiratory fitness. Diabetes Care, 
28(2), 391-397. 
Kelley, D. E., Goodpaster, B., Wing, R. R., & Simoneau, J.-A. (1999). Skeletal muscle fatty 
acid metabolism in association with insulin resistance, obesity, and weight loss. The 
American Journal of Physiology Endocrinology and Metabolism, 277(6), E1130-1141. 
Kelley, D. E., He, J., Menshikova, E. V., & Ritov, V. B. (2002). Dysfunction of mitochondria 
in human skeletal muscle in type 2 diabetes. Diabetes, 51(10), 2944. 
Kelly, G. S. (2000). Insulin resistance: lifestyle and nutritional interventions. Alternative 
Medicine Review: A Journal of Clinical Therapeutic, 5(2), 109. 
Khaled, M. A., McCutcheon, M. J., Reddy, S., Pearman, P. L., Hunter, G. R., & Weinsier, R. 
L. (1988). Electrical impedance in assessing human body composition: the BIA 
method. American Journal of Clinical Nutrition, 47(5), 789-792. 
Kim, J. Y., Hickner, R. C., Cortright, R. L., Dohm, G. L., & Houmard, J. A. (2000). Lipid 
oxidation is reduced in obese human skeletal muscle. American Journal of Physiology 
Endocrinology and Metabolism, 279(5), E1039. 
Klein, S., Fontana, L., Young, L., Coggan, A., Kilo, C., Patterson, B., et al. (2004). Absence 
of an effect of liposuction on insulin action and risk factors for coronary heart disease. 
New England Journal of Medicine, 350(25), 2549-2557. 
 References 
 191
Kohn, T. A., Essen-Gustavsson, B., & Myburgh, K. H. (2007). Do skeletal muscle phenotypic 
characteristics of Xhosa and Caucasian endurance runners differ when matched for 
training and racing distances? Journal of Applied Physiology, 103(3), 932-940. 
Komi, P. V., Viitasalo, J. H., Havu, M., Thorstensson, A., Sjödin, B., & Karlsson, J. (1977). 
Skeletal muscle fibres and muscle enzyme activities in monozygous and dizygous 
twins of both sexes. Acta Physiologica Scandinavica, 100(4), 385. 
Koopman, R., Manders, R. J. F., Jonkers, R. A. M., Hul, G. B. J., Kuipers, H., & van Loon, L. 
J. C. (2006). Intramyocellular lipid and glycogen content are reduced following 
resistance exercise in untrained healthy males. European Journal of Applied 
Physiology, 96(5), 525-534. 
Kosaka, K., Kuzuya, T., Yoshinaga, H., & Hagura, R. (1996). A prospective study of health 
check examinees for the development of non-insulin-dependent diabetes mellitus: 
relationship of the incidence of diabetes with the initial insulinogenic index and degree 
of obesity. Diabetic Medicine: A Journal of the British Diabetic Association, 13(9 
Suppl 6), S120. 
Kushner, R. F., Gudivaka, R., & Schoeller, D. A. (1996). Clinical characteristics influencing 
bioelectrical impedance analysis measurements. American Journal of Clinical 
Nutrition, 64(3), 423-427. 
Kyle, U. G., Bosaeus, I., De Lorenzo, A. D., Deurenberg, P., Elia, M., Gómez, J. M., et al. 
(2004). Bioelectrical impedance analysis-part I: review of principles and methods. 
Clinical Nutrition, 23(5), 1226-1243. 
Kyle, U. G., Bosaeus, I., De Lorenzo, A. D., Deurenberg, P., Elia, M., Manuel Gómez, J., et 
al. (2004). Bioelectrical impedance analysis—part II: utilization in clinical practice. 
Clinical Nutrition, 23(6), 1430-1453. 
Lake, D. A. (1992). Neuromuscular electrical stimulation. An overview and its application in 
the treatment of sports injuries. Sports Medicine 13(5), 320. 
Lean, M. E. J., Han, T. S., & Seidell, J. C. (1998). Impairment of health and quality of life in 
people with large waist circumference. The Lancet, 351(9106), 853-856. 
Lee, I. M., Manson, J. E., Hennekens, C. H., & Paffenbarger Jr, R. S. (1993). Body weight 
and mortality. A 27-year follow-up of middle-aged men. Journal of the American 
Medical Association, 270(23), 2823. 
Lee, J. M., Okumura, M. J., Davis, M. M., Herman, W. H., & Gurney, J. G. (2006). 
Prevalence and determinants of insulin resistance among US adolescents: a 
population-based study. Diabetes Care, 29(11), 2427. 
 References 
 192
Lee, S., Kuk, J. L., Davidson, L. E., Hudson, R., Kilpatrick, K., Graham, T. E., et al. (2005). 
Exercise without weight loss is an effective strategy for obesity reduction in obese 
individuals with and without type 2 diabetes. Journal of Applied Physiology, 99(3), 
1220-1225. 
Lee, S. J., Kuk, J. L., Katzmarzyk, P. T., Blair, S. N., Church, T. S., & Ross, R. (2005). 
Cardiorespiratory Fitness Attenuates Metabolic Risk Independent of Abdominal 
Subcutaneous and Visceral Fat in Men. Diabetes Care, 28(4), 895. 
Léger, L. A., & Lambert, J. (1982). A maximal multistage 20-m shuttle run test to predict O 2 
max. European Journal of Applied Physiology, 49(1), 1-12. 
Lewis, G. F., Carpentier, A., Adeli, K., & Giacca, A. (2002). Disordered fat storage and 
mobilization in the pathogenesis of insulin resistance and type 2 diabetes. Endocrine 
Reviews, 23(2), 201-229. 
Lexell, J., Henriksson-Larsen, K., & Sjostrom, M. (1983). Distribution of different fibre types 
in human skeletal muscles 2. A study of cross-sections of whole m. vastus lateralis. 
Acta Physiologica Scandinavica, 117(1), 115-122. 
Lexell, J., Taylor, C., & Sjostrom, M. (1985). Analysis of sampling errors in biopsy 
techniques using data from whole muscle cross sections. Journal of Applied 
Physiology, 59(4), 1228-1235. 
Lillioja, S., Young, A., Cutler, C., Ivy, J., Abbott, W., Zawadzki, J., et al. (1987a). Skeletal 
muscle capillary density and fiber type are possible determinants of in vivo insulin 
resistance in man. Journal of Clinical Investigation, 80(2), 415–424. 
Lillioja, S., Young, A., Cutler, C., Ivy, J., Abbott, W., Zawadzki, J., et al. (1987b). Skeletal 
muscle capillary density and fiber type are possible determinants of in vivo insulin 
resistance in man. J Clin Invest, 80(2), 415–424. 
Lindstrom, J., Eriksson, J. G., Valle, T. T., Aunola, S., Cepaitis, Z., Hakumaki, M., et al. 
(2003). Prevention of diabetes mellitus in subjects with impaired glucose tolerance in 
the Finnish Diabetes Prevention Study: results from a randomized clinical trial. 
Journal of the American Society of Nephrology, 14(90002), S108-113. 
Lohman, T. G. (1981a). Skinfolds and body density and their relation to body fatness: a 
review. Hum Biol, 53(2), 181-225. 
Lohman, T. G. (1981b). Skinfolds and body density and their relation to body fatness: a 
review. Human Biology, 53(2), 181-225. 
 References 
 193
Lohman, T. G., Pollock, M. L., Slaughter, M. H., Brandon, L. J., & Boileau, R. A. (1984). 
Methodological factors and the prediction of body fat in female athletes. Medicine & 
Science in Sports & Exercise, 16(1), 92-96. 
Lowell, B. B., & Spiegelman, B. M. (2000). Towards a molecular understanding of adaptive 
thermogenesis. Nature, 404(6778), 652-660. 
Lukaski, H. C. (1987). Methods for the assessment of human body composition: traditional 
and new. American Journal of Clinical Nutrition, 46(4), 537-556. 
Luke, A., Durazo-Arvizu, R., Rotimi, C., Prewitt, T. E., Forrester, T., Wilks, R., et al. (1997). 
Relation between body mass index and body fat in Black population samples from 
Nigeria, Jamaica, and the United States. The American Journal of Epidemiology, 
145(7), 620-628. 
Macfarlane, D. J. (2001). Automated metabolic gas analysis systems: a review. Sports 
Medicine, 31(12), 841-861. 
MacIntosh, B. R., Herzog, W., Suter, E., Wiley, J. P., & Sokolosky, J. (1993). Human skeletal 
muscle fibre types and force: velocity properties. European Journal of Applied 
Physiology, 67(6), 499-506. 
Malavolti, M., Mussi, C., Poli, M., Fantuzzi, A. L., Salvioli, G., Battistini, N., et al. (2003). 
Cross-calibration of eight-polar bioelectrical impedance analysis versus dual-energy 
X-ray absorptiometry for the assessment of total and appendicular body composition 
in healthy subjects aged 21-82 years. Annals of Human Biology, 30(4), 380-391. 
Marin, P., Andersson, B., Krotkiewski, M., & Bjorntorp, P. (1994). Muscle fiber composition 
and capillary density in women and men with NIDDM. Diabetes Care, 17(5), 382-
386. 
Martin, T. P., Stein, R. B., Hoeppner, P. H., & Reid, D. C. (1992). Influence of electrical 
stimulation on the morphological and metabolic properties of paralyzed muscle. 
Journal of Applied Physiology, 72(4), 1401-1406. 
Mathers, J., & Wolever, T. (2002). Digestion and metabolism of carbohydrates. In M. Gibney, 
H. Vorster & F. Kok (Eds.), Introduction to Human Nutrition. Oxford, UK: Blackwell 
Science Ltd. 
Matsuda, M., & DeFronzo, R. A. (1997). In vivo measurement of insulin sensitivity in 
humans. Clinical Research in Diabetes and Obesity: Methods, Assessment, and 
Metabolic Regulation, 23. 
Matthews, D. R., Hosker, J. P., Rudenski, A. S., Naylor, B. A., Treacher, D. F., & Turner, R. 
C. (1985). Homeostasis model assessment: insulin resistance and β-cell function from 
 References 
 194
fasting plasma glucose and insulin concentrations in man. Diabetologia, 28(7), 412-
419. 
Mazess, R. B., Barden, H. S., Bisek, J. P., & Hanson, J. (1990). Dual-energy x-ray 
absorptiometry for total-body and regional bone- mineral and soft-tissue composition. 
American Journal of Clinical Nutrition, 51(6), 1106-1112. 
McAuley, K., Williams, S., Mann, J., Goulding, A., & Murphy, E. (2002). Increased risk of 
type 2 diabetes despite same degree of adiposity in different racial groups. Diabetes 
Care, 25(12), 2360. 
McAuley, K., Williams, S., Mann, J., Walker, R., Lewis-Barned, N., Temple, L., et al. (2001). 
Diagnosing insulin resistance in the general population. Diabetes Care, 24(3), 460-
464. 
McAuley, K. A., Williams, S. M., Mann, J. I., Goulding, A., Chisholm, A., Wilson, N., et al. 
(2002). Intensive lifestyle changes are necessary to improve insulin sensitivity. 
Diabetes Care, 25(3), 445-452. 
Mei, Z., Grummer-Strawn, L. M., Pietrobelli, A., Goulding, A., Goran, M. I., & Dietz, W. H. 
(2002). Validity of body mass index compared with other body-composition screening 
indexes for the assessment of body fatness in children and adolescents. American 
Journal of Clinical Nutrition, 75(6), 978-985. 
Michael, M. D., Kulkarni, R. N., Postic, C., Previs, S. F., Shulman, G. I., Magnuson, M. A., et 
al. (2000). Loss of insulin signaling in hepatocytes leads to severe insulin resistance 
and progressive hepatic dysfunction. Molecular Cell, 6(1), 87-97. 
Mikines, K. J., Sonne, B., Farrell, P. A., Tronier, B., & Galbo, H. (1988). Effect of physical 
exercise on sensitivity and responsiveness to insulin in humans. The American Journal 
of Physiology Endocrinology and Metabolism, 254(3), E248-259. 
Ministry of Health, N. Z. (1999a). NZ Food, NZ People. Key results of the 1997 New Zealand 
Nutrition Survey. Wellington, NZ: Ministry of Health. 
Ministry of Health, N. Z. (1999b). Taking the Pulse: the New Zealand Health Survey 1996/7. 
Wellington, NZ: Ministry of Health. 
Ministry of Health, N. Z. (2002). He Korowai Oranga: Maori Health Strategy. Wellington, 
NZ: Ministry of Health. 
Ministry of Health, N. Z. (2003). Healthy Eating - Healthy Action, Oranga Kai - Oranga 
Pumau: A Background. Wellington: Ministry of Health. 
Ministry of Health, N. Z. (2004a). Mortality and demographic data 2000. Wellington, NZ: 
Ministry of Health. 
 References 
 195
Ministry of Health, N. Z. (2004b). A Portrait of Health: Key Results of the 2002/2003 New 
Zealand Health Survey. Wellington, NZ.: Ministry of Health. 
Ministry of Health, N. Z. (2007). Healthy Eating - Healthy Action, Oranga Kai - Oranga 
Pumau.   Retrieved 13/03/2008, 2008, from 
www.moh.govt.nz/healthyeatinghealthyaction 
Ministry of Health, N. Z. (2008). A portrait of Health: Key results of the 2006/07 New 
Zealand Health Survey. Wellington, NZ: Ministry of Health. 
Misra, A., Wasir, J. S., & Vikram, N. K. (2005). Waist circumference criteria for the 
diagnosis of abdominal obesity are not applicable uniformly to all populations and 
ethnic groups. Nutrition, 21(9), 969-976. 
Moore, M. P., & Lunt, H. (2000). Diabetes in New Zealand. Diabetes Research and Clinical 
Practice, 50(Supplement 2), S65-S71. 
Morgan, T. E., Short, F. A., & Cobb, L. A. (1969). Effect of long-term exercise on skeletal 
muscle lipid composition. American Journal of Physiology, 216(1), 82-86. 
Moro, C., Bajpeyi, S., & Smith, S. R. (2008). Determinants of intramyocellular triglyceride 
turnover: implications for insulin sensitivity. American Journal of Physiology 
Endocrinology and Metabolism, 294(2), E203. 
Murphy, E., McAuley, K. A., Bell, D., McLay, R. T., Chisholm, A., Hurley, R., et al. (2003). 
A new approach to design and implement a lifestyle intervention programme to 
prevent type 2 diabetes in New Zealand Maori. Asia Pacific Journal of Clinical 
Nutrition, 12(4), 419-422. 
National Institutes of Health Technology. (1996). Bioelectrical impedance analysis in body 
composition measurement: National Institutes of Health Technology Assessment 
Conference Statement. American Journal of Clinical Nutrition, 64(3), 524S-532. 
Needham, D. M. (1926). Red and white muscle. Physiological Reviews, 6(1), 1-27. 
Neel, J. V. (1962). Diabetes mellitus: a thrifty genotype rendered detrimental by progress. The 
American Journal of Human Genetics, 14(353), 62. 
Ni Mhurchua, C., Blakelya, T., Walla, J., Rodgersa, A., Jianga, Y., & Wiltona, J. (2007). 
Strategies to promote healthier food purchases: a pilot supermarket intervention study. 
Public Health Nutrition, 10(6), 608-615. 
Noonan, V., & Dean, E. (2000). Submaximal exercise testing: clinical application and 
interpretation. Physical Therapy, 80(8), 782-807. 
 References 
 196
Nord, R., & Payne, R. (1995). Dual-energy X-ray absorptiometry vs underwater weighing 
comparison of strengths and weaknesses Asia Pacific Journal of Clinical Nutrition, 4, 
173-175. 
Norton, K. (1996). Anthropometric estimation of body fat. In K. Norton & T. Olds (Eds.), 
Anthropometrica: a Textbook of Body Measurement for Sports and Health Courses 
(pp. 171). Sydney, Australia: University of New South Wales Press. 
O'Dea, K. (1984). Marked improvement in carbohydrate and lipid metabolism in diabetic 
Australian Aborigines after temporary reversion to traditional lifestyle. Diabetes, 
33(6), 596-603. 
O’Dea, K., White, N. G., & Sinclair, A. J. (1988). An investigation of nutrition-related risk 
factors in an isolated Aboriginal community in northern Australia: advantages of a 
traditionally-orientated lifestyle. The Medical Journal of Australia, 148(4), 177-180. 
Oakes, N. D., Kennedy, C. J., Jenkins, A. B., Laybutt, D. R., Chisholm, D. J., & Kraegen, E. 
W. (1994). A new antidiabetic agent, BRL 49653, reduces lipid availability and 
improves insulin action and glucoregulation in the rat. Diabetes, 43(10), 1203-1210. 
Okosun, I. S., Cooper, R. S., Rotimi, C. N., Osotimehin, B., & Forrester, T. (1998). 
Association of waist circumference with risk of hypertension and type 2 diabetes in 
Nigerians, Jamaicans, and African-Americans. Diabetes Care, 21(11), 1836-1842. 
Okosun, I. S., Liao, Y., Rotimi, C. N., Choi, S., & Cooper, R. S. (2000). Predictive values of 
waist circumference for dyslipidemia, type 2 diabetes and hypertension in overweight 
White, Black, and Hispanic American adults. Journal of Clinical Epidemiology, 53(4), 
401-408. 
Okosun, I. S., Tedders, S. H., Choi, S., & Dever, G. E. (2000). Abdominal adiposity values 
associated with established body mass indexes in White, Black and Hispanic 
Americans. A study from the Third National Health and Nutrition Examination 
Survey. International Journal of Obesity and Related Metabolic Disorders, 24(10), 
1279. 
Pan, D. A., Lillioja, S., Kriketos, A. D., Milner, M. R., Baur, L. A., Bogardus, C., et al. 
(1997). Skeletal muscle triglyceride levels are inversely related to insulin action. 
Diabetes, 46(6), 983-988. 
Pan, X. R., Li, G. W., Hu, Y. H., Wang, J. X., Yang, W. Y., An, Z. X., et al. (1997). Effects of 
diet and exercise in preventing NIDDM in people with impaired glucose tolerance. 
The Da Qing IGT and Diabetes Study. Diabetes Care, 20(4), 537-544. 
 References 
 197
Papa, S. (1996). Mitochondrial oxidative phosphorylation changes in the life span. Molecular 
aspects and physiopathological implications. Biochimica et Biophysica Acta (BBA) - 
Bioenergetics, 1276(2), 87-105. 
Parnell, W., Scragg, R., Wilson, N., Schaaf, D., & Fitzgerald, E. (2003). NZ food NZ 
children: key results of the 2002 national children’s nutrition survey. Wellington, 
Ministry of Health. 
Parra, E. J., Kittles, R. A., & Shriver, M. D. (2004). Implications of correlations between skin 
color and genetic ancestry for biomedical research. Nature genetics, S54-S60. 
Pearce, N., Foliaki, S., Sporle, A., & Cunningham, C. (2004). Genetics, race, ethnicity, and 
health. British Medical Journal, 328(7447), 1070-1072. 
Pearce, N., Pomare, E., Marshall, S., & Borman, B. (1993). Mortality and social class in 
Maori and non-Maori New Zealand men: changes between 1975-1977 and 1985-1987. 
The New Zealand Medical Journal, 106, 193-196. 
Penedo, F. J., & Dahn, J. R. (2005). Exercise and well-being: a review of mental and physical 
health benefits associated with physical activity. Current Opinion in Psychiatry, 18(2), 
189-193. 
Pereira, M. A., Kartashov, A. I., Ebbeling, C. B., Van Horn, L., Slattery, M. L., Jacobs, J. D. 
R., et al. (2005). Fast-food habits, weight gain, and insulin resistance (the CARDIA 
study): 15-year prospective analysis. The Lancet, 365(9453), 36-42. 
Pette, D. (2001). Plasticity in skeletal, cardiac, and smooth muscle: historical perspectives: 
plasticity of mammalian skeletal muscle. Journal of Applied Physiology, 90(3), 1119-
1124. 
Phillips, D. I., Caddy, S., Ilic, V., Fielding, B. A., Frayn, K. N., Borthwick, A. C., et al. 
(1996). Intramuscular triglyceride and muscle insulin sensitivity: evidence for a 
relationship in nondiabetic subjects. Metabolism: Clinical and Experimental, 45(8), 
947. 
Ploug, T., Galbo, H., & Richter, E. A. (1984). Increased muscle glucose uptake during 
contractions: no need for insulin. The American Journal of Physiology Endocrinology 
and Metabolism, 247(6), E726-731. 
Poehlman, E. T., Dvorak, R. V., DeNino, W. F., Brochu, M., & Ades, P. A. (2000). Effects of 
resistance training and endurance training on insulin sensitivity in nonobese, young 
women: a controlled randomized trial. Journal of Clinical Endocrinology & 
Metabolism, 85(7), 2463-2468. 
 References 
 198
Polonsky, K. S., Given, B. D., Hirsch, L., Shapiro, E. T., Tillil, H., Beebe, C., et al. (1988). 
Quantitative study of insulin secretion and clearance in normal and obese subjects. 
Journal of Clinical Investigation, 81(2), 435. 
Pratley, R. E. (2007). Gene–environment interactions in the pathogenesis of type 2 diabetes 
mellitus: lessons learned from the Pima Indians. Proceedings of the Nutrition Society, 
57(02), 175-181. 
Prentice, A. M., & Jebb, S. A. (2001). Beyond body mass index. Obesity Reviews, 2(3), 141-
147. 
PricewaterhouseCoopers. (2001). Type 2 Diabetes: Managing for Better Health outcomes. 
Wellington, NZ.: Diabetes New Zealand Inc. 
Raji, A., Seely, E. W., Arky, R. A., & Simonson, D. C. (2001). Body fat distribution and 
insulin resistance in healthy Asian Indians and Caucasians. Journal of Clinical 
Endocrinology & Metabolism, 86(11), 5366-5371. 
Randle, P. J., Garland, P. B., Hales, C. N., & Newsholme, E. A. (1963). The glucose fatty-
acid cycle. Its role in insulin sensitivity and the metabolic disturbances of diabetes 
mellitus. Lancet, 1(7285), 785. 
Reaven, G. M. (1988). Role of insulin resistance in human disease. Diabetes, 37(1), 1595-
1607. 
Rebbeck, T. R., & Sankar, P. (2005). Ethnicity, ancestry, and race in molecular epidemiologic 
research. Cancer Epidemiology Biomarkers & Prevention, 14(11), 2467. 
Reid, I. R., Mackie, M., & Ibbertson, H. K. (1986). Bone mineral content in Polynesian and 
white New Zealand women. British Medical Journal (Clinical Research Ed), 
292(6535), 1547-1548. 
Reiser, P. J., Moss, R. L., Giulian, G. G., & Greaser, M. L. (1985). Shortening velocity in 
single fibers from adult rabbit soleus muscles is correlated with myosin heavy chain 
composition. The Journal of Biological Chemistry, 260(16), 9077-9080. 
Richter, E. A., Garetto, L. P., Goodman, M. N., & Ruderman, N. B. (1982). Muscle glucose 
metabolism following exercise in the rat: increased sensitivity to insulin. Journal of 
Clinical Investigation, 69(4), 785-793. 
Richter, E. A., Mikines, K. J., Galbo, H., & Kiens, B. (1989). Effect of exercise on insulin 
action in human skeletal muscle. Journal of Applied Physiology, 66(2), 876-885. 
Rimbert, V., Boirie, Y., Bedu, M., Hocquette, J.-F., Ritz, P., & Morio, B. (2004). Muscle fat 
oxidative capacity is not impaired by age but by physical inactivity: association with 
 References 
 199
insulin sensitivity. Journal of the Federation of American Societies for Experimental 
Biology, 18(6), 737-739. 
Rising, R., Swinburn, B., Larson, K., & Ravussin, E. (1991). Body composition in Pima 
Indians: validation of bioelectrical resistance. American Journal of Clinical Nutrition, 
53(3), 594-598. 
Rocchini, A. P. (2002). Childhood obesity and a diabetes epidemic. New England Journal of 
Medicine, 346(11), 854-855. 
Rochester, L., Barron, M. J., Chandler, C. S., Sutton, R. A., Miller, S., & Johnson, M. A. 
(1995). Influence of electrical stimulation of the tibialis anterior muscle in paraplegic 
subjects. 2. Morphological and histochemical properties. Paraplegia, 33(9), 514. 
Roden, M., Price, T. B., Perseghin, G., Petersen, K. F., Rothman, D. L., Cline, G. W., et al. 
(1996). Mechanism of free fatty acid-induced insulin resistance in humans. Journal of 
Clinical Investigation, 97(12), 2859. 
Rosen, S. (1969). Three "myosin adenosine triphosphatase" systems: the nature of their pH 
lability and sulfhydryl dependence. Journal of Histochemistry and Cytochemistry, 
16(9), 670. 
Ross, R., Aru, J., Freeman, J., Hudson, R., & Janssen, I. (2002). Abdominal adiposity and 
insulin resistance in obese men. American Journal of Physiology- Endocrinology And 
Metabolism, 282(3), E657. 
Rowe, J. W., Minaker, K. L., Pallotta, J. A., & Flier, J. S. (1983). Characterization of the 
insulin resistance of aging. Journal of Clinical Investigation, 71(6), 1581. 
Rush, E., Plank, L., Chandu, V., Laulu, M., Simmons, D., Swinburn, B., et al. (2004). Body 
size, body composition, and fat distribution: a comparison of young New Zealand men 
of European, Pacific Island, and Asian Indian ethnicities. The New Zealand Medical 
Journal, 117(1207), U1203. 
Rush, E. C., Freitas, I., & Plank, L. D. (2009). Body size, body composition and fat 
distribution: comparative analysis of European, Maori, Pacific Island and Asian Indian 
adults. British Journal of Nutrition, 102(04), 632-641. 
Rush, E. C., Plank, L. D., & Robinson, S. M. (1997). Resting metabolic rate in young 
Polynesian and Caucasian women. International Journal of Obesity, 21(11), 1071-
1075. 
Rush, E. C., Puniani, K., Valencia, M. E., Davies, P. S., & Plank, L. D. (2003). Estimation of 
body fatness from body mass index and bioelectrical impedance: comparison of New 
 References 
 200
Zealand European, Maori and Pacific Island children. European Journal of Clinical 
Nutrition, 57(11), 1394-1401. 
Sachdev, P. (1998). The New Zealand Maori and the contemporary health system. In S. 
Okpaku (Ed.), Clinical Methods in Transcultural Psychiatry (pp. 111-136). 
Washington, D.C: American Psychiatric Press. 
Sadoyama, T., Masuda, T., Miyata, H., & Katsuta, S. (1988). Fibre conduction velocity and 
fibre composition in human vastus lateralis. European Journal of Applied Physiology, 
57(6), 767-771. 
Saltiel, A. R., & Kahn, C. R. (2001). Insulin signalling and the regulation of glucose and lipid 
metabolism. Nature, 414(6865), 799-806. 
Schantz, P., Billeter, R., Henriksson, J., & Jansson, E. (1982). Training-induced increase in 
myofibrillar ATPase intermediate fibers in human skeletal muscle. Muscle & Nerve, 
5(8), 628-636. 
Schrauwen-Hinderling, V. B., Schrauwen, P., Hesselink, M. K. C., van Engelshoven, J. M. 
A., Nicolay, K., Saris, W. H. M., et al. (2003). The increase in intramyocellular lipid 
content is a very early response to training. Journal of Clinical Endocrinology & 
Metabolism, 88(4), 1610-1616. 
Schulze, M. B., Manson, J. E., Ludwig, D. S., Colditz, G. A., Stampfer, M. J., Willett, W. C., 
et al. (2004). Sugar-sweetened beverages, weight gain, and incidence of type 2 
diabetes in young and middle-aged women. Journal of the American Medical 
Association, 292(8), 927-934. 
Scott, W., Stevens, J., & Binder-Macleod, S. A. (2001). Human skeletal muscle fiber type 
classifications. Physical Therapy, 81(11), 1810-1816. 
Scragg, R., Baker, J., Metcalf, P., & Dryson, E. (1991). Prevalence of diabetes mellitus and 
impaired glucose tolerance in a New Zealand multiracial workforce. The New Zealand 
Medical Journal, 104(920), 395-397. 
Segal, K. R., Gutin, B., Presta, E., Wang, J., & Van Itallie, T. B. (1985). Estimation of human 
body composition by electrical impedance methods: a comparative study. Journal of 
Applied Physiology, 58(5), 1565-1571. 
Seidell, J. C. (2000). Obesity, insulin resistance and diabetes - a worldwide epidemic. British 
Journal Of Nutrition, 83, 5-8. 
Seppälä-Lindroos, A., Vehkavaara, S., Häkkinen, A. M., Goto, T., Westerbacka, J., Sovijärvi, 
A., et al. (2002). Fat accumulation in the liver is associated with defects in insulin 
 References 
 201
suppression of glucose production and serum free fatty acids independent of obesity in 
normal men. Journal of Clinical Endocrinology & Metabolism, 87(7), 3023-3028. 
Serdula, M. K., Ivery, D., Coates, R. J., Freedman, D. S., Williamson, D. F., & Byers, T. 
(1993). Do obese children become obese adults? A review of the literature. Preventive 
medicine, 22(2), 167. 
Shield, A., & Zhou, S. (2004). Assessing Voluntary Muscle Activation with the Twitch 
Interpolation Technique. Sports Medicine, 34(4), 253-267. 
Shulman, G. I. (2000). Cellular mechanisms of insulin resistance. Journal of Clinical 
Investigation, 106(2), 171-176. 
Simmons, D., & Thompson, C. F. (2004). Prevalence of the metabolic syndrome among adult 
New Zealanders of Polynesian and European descent. Diabetes Care, 27(12), 3002-
3004. 
Simmons, D., Thompson, C. F., & Volklander, D. (2001). Polynesians: prone to obesity and 
type 2 diabetes mellitus but not hyperinsulinaemia. Diabetic Medicine, 18(3), 193-
198. 
Simoneau, J.-A., & Kelley, D. E. (1997). Altered glycolytic and oxidative capacities of 
skeletal muscle contribute to insulin resistance in NIDDM. Journal of Applied 
Physiology, 83(1), 166-171. 
Simoneau, J. A., & Bouchard, C. (1995). Genetic determinism of fiber type proportion in 
human skeletal muscle. Journal of the Federation of American Societies for 
Experimental Biology, 9(11), 1091-1095. 
Simoneau, J. A., Lortie, G., Boulay, M. R., Thibault, M. C., & Bouchard, C. (2008). 
Repeatability of fibre type and enzyme activity measurements in human skeletal 
muscle. Clinical Physiology and Functional Imaging, 6(4), 347-356. 
Siri, W. E. (1961). Body volume measurement by gas dilution. In J. Brozek (Ed.), Techniques 
for measuring body composition (pp. 108–117). Washington, DC: National Academy 
of Sciences, National Research Council. 
Smylie, J., Anderson, I., Ratima, M., Crengle, S., & Anderson, M. (2006). Indigenous health 
performance measurement systems in Canada, Australia, and New Zealand. The 
Lancet, 367(9527), 2029-2031. 
Sniderman, A. D., Bhopal, R., Prabhakaran, D., Sarrafzadegan, N., & Tchernof, A. (2007). 
Why might South Asians be so susceptible to central obesity and its atherogenic 
consequences? The adipose tissue overflow hypothesis. International Journal of 
Epidemiology, 36(1), 220-225. 
 References 
 202
Solomon, C. G., & Manson, J. E. (1997). Obesity and mortality: a review of the 
epidemiologic data. American Journal of Clinical Nutrition, 66(4), 1044S-1050. 
SPARC. (2004, May 2004). Sport, Recreation and Physical Activity in New Zealand. Paper 
presented at the Leisure and Heritage studies, Victoria University. 
SPARC. (2005). SPARC launches push play activator [Electronic Version]. Press Releases. 
Retrieved 16 May 2007 from http://www.sparc.org.nz/about-sparc/news/sparc-
launches-push-play-activator. 
Stannard, S. R., Holdaway, M. A., Sachinwalla, T., & Cunningham, C. W. (2007). Insulin 
sensitivity and intramyocellular lipid concentrations in young Maori men. Diabetic 
Medicine, 24(11), 1205-1212. 
Stannard, S. R., & Johnson, N. A. (2004). Insulin resistance and elevated triglyceride in 
muscle: more important for survival than ‘thrifty’genes? The Journal of Physiology, 
554(3), 595. 
Stannard, S. R., & Johnson, N. A. (2006). Energy well spent fighting the diabetes epidemic. 
Diabetes/Metabolism Research and Reviews, 22, 11-19. 
Stannard, S. R., Thompson, M. W., Fairbairn, K., Huard, B., Sachinwalla, T., & Thompson, 
C. H. (2002). Fasting for 72h increases intramyocellular lipid content in nondiabetic, 
physically fit men. The American Journal of Physiology Endocrinology and 
Metabolism, 283, 1185-1191. 
Staron, R. S. (1997). Human skeletal muscle fiber types: delineation, development, and 
distribution. Physiologie appliquée, nutrition et métabolisme, 22, 307-327. 
Staron, R. S., Leonardi, M. J., Karapondo, D. L., Malicky, E. S., Falkel, J. E., Hagerman, F. 
C., et al. (1991). Strength and skeletal muscle adaptations in heavy-resistance-trained 
women after detraining and retraining. Journal of Applied Physiology, 70(2), 631-640. 
Statistics New Zealand. (2006a). Demographic trends: 2005.   Retrieved 1 October 2009, 
2009, from 
http://www.stats.govt.nz/~/media/statistics/publications/population/demographic-
trends-2005/chapter4-deaths-life-expectancy.aspx 
Statistics New Zealand. (2006b). QuickStats about Income: 2006 Census.   Retrieved 7 
August 2009, 2009, from 
http://www.stats.govt.nz/~/media/Statistics/Publications/Census/2006%20Census%20r
eports/QuickStats%20About%20A%20Subject/QuickStats%20About%20Incomes/20
06-census-quickstats-about-incomes.ashx 
 References 
 203
Statistics New Zealand. (2006c). QuickStats about New Zealand's population and dwelling: 
2006 Census.   Retrieved 7 August 2009, 2009, from 
http://www.stats.govt.nz/~/media/Statistics/Publications/Census/2006%20Census%20r
eports/QuickStats%20About%20A%20Subject/QuickStats%20About%20NZs%20Pop
ulation%20and%20Dwellings/qstats-about-nzs-population-and-dwellings-2006-
census.ashx 
Statistics New Zealand. (2008). New Zealand life expectancy increases.   Retrieved 30 
September 2009, 2009, from 
http://www.stats.govt.nz/~/media/Statistics/Browse%20for%20stats/NZLifeTables/M
R05-07/NZLifeTables05-07MR.ashx 
Stedman's medical dictionary. (2000).  ( 27 ed.). Baltimore, Maryland: Lippincott Williams & 
Wilkins. 
Stumvoll, M., Mitrakou, A., Pimenta, W., Jenssen, T., Yki-Jarvinen, H., Van Haeften, T., et 
al. (2000). Use of the oral glucose tolerance test to assess insulin release and insulin 
sensitivity. Diabetes Care, 23(3), 295-301. 
Suminski, R. R., Mattern, C. O., & Devor, S. T. (2002). Influence of racial origin and skeletal 
muscle properties on disease prevalence and physical performance. Sports Medicine, 
32(11), 667-673. 
Summers, S. A., Garza, L. A., Zhou, H., & Birnbaum, M. J. (1998). Regulation of insulin-
stimulated glucose transporter GLUT4 translocation and Akt kinase activity by 
ceramide. Molecular and cellular biology, 18(9), 5457. 
Suter, E., Herzog, W., Sokolosky, J., Wiley, J. P., & Macintosh, B. R. (1993). Muscle fiber 
type distribution as estimated by Cybex testing and by muscle biopsy. Medicine & 
Science in Sports & Exercise, 25(3), 363. 
Suter, P. M., Hasler, E., & Vetter, W. (1997). Effects of alcohol on energy metabolism and 
body weight regulation: is alcohol a risk factor for obesity? Nutrition Reviews, 55(5), 
157-171. 
Svendsen, O. L., Haarbo, J., Hassager, C., & Christiansen, C. (1993). Accuracy of 
measurements of body composition by dual-energy x-ray absorptiometry in vivo. 
American Journal of Clinical Nutrition, 57(5), 605-608. 
Swinburn, B., Ley, S., Carmichael, H., & Palnk, L. (1999). Body size and composition in 
Polynesians. International Journal of Obesity, 23, 1178-1183. 
 References 
 204
Tang, H., Quertermous, T., Rodriguez, B., Kardia, S. L. R., Zhu, X., Brown, A., et al. (2005). 
Genetic structure, self-Identified race/ethnicity, and confounding in case-control 
association studies. The American Journal of Human Genetics, 76(2), 268-275. 
Tanner, C. J., Barakat, H. A., Dohm, G. L., Pories, W. J., MacDonald, K. G., Cunningham, P. 
R. G., et al. (2002). Muscle fiber type is associated with obesity and weight loss. The 
American Journal of Physiology Endocrinology and Metabolism, 282(6), E1191-1196. 
Thamer, C., Machann, J., Bachmann, O., Haap, M., Dahl, D., Wietek, B., et al. (2003). 
Intramyocellular lipids: anthropometric determinants and relationships with maximal 
aerobic capacity and insulin sensitivity. Journal of Clinical Endocrinology & 
Metabolism, 88(4), 1785-1791. 
Thomas, D. R. (2001). Assessing ethnicity in New Zealand health research. New Zealand 
Medical Journal, 114, 86-88. 
Thomas, S., Reading, J., & Shephard, R. J. (1992). Revision of the physical activity readiness 
questionnaire (PAR-Q). Canadian Journal of Sport Sciences, Journal Canadien des 
Sciences du Sport, 17(4), 338. 
Thorstensson, A., Grimby, G., & Karlsson, J. (1976). Force-velocity relations and fiber 
composition in human knee extensor muscles. Journal of Applied Physiology, 40(1), 
12-16. 
Thorstensson, A., & Karlsson, J. (1976). Fatiguability and fibre composition of human 
skeletal muscle. Acta Physiologica Scandinavica, 98(3), 318-322. 
Tipene-Leach, D., Pahau, H., Joseph, N., Coppell, K., McAuley, K., Booker, C., et al. (2004). 
Insulin resistance in a rural Māori community. The New Zealand Medical Journal, 
117(1207), U1208. 
Tobias, M., Paul, S., & Yea, L. (2006). Embodying Social Rank: How body fat varies with 
social status, gender and ethnicity in New Zealand. Wellington,NZ.: Ministry of 
Health. 
Toledo, F. G. S., Menshikova, E. V., Ritov, V. B., Azuma, K., Radikova, Z., DeLany, J., et al. 
(2007). Effects of physical activity and weight loss on skeletal muscle mitochondria 
and relationship with glucose control in type 2 diabetes. Diabetes, 56(8), 2142. 
Toledo, F. G. S., Watkins, S., & Kelley, D. E. (2006). Changes induced by physical activity 
and weight loss in the morphology of intermyofibrillar mitochondria in obese men and 
women. Journal of Clinical Endocrinology & Metabolism, 91(8), 3224. 
Tuomilehto, J., Lindstrom, J., Eriksson, J. G., Valle, T. T., Hamalainen, H., Ilanne-Parikka, 
P., et al. (2001). Prevention of type 2 diabetes mellitus by changes in lifestyle among 
 References 
 205
subjects with impaired glucose tolerance. New England Journal of Medicine, 344(18), 
1343-1350. 
Tuttle, C. R. (2002). Lessons learned from failed research among the New Zealand Māori. 
Nutritional Anthropology, 25(1), 13-19. 
Unger, R. H. (1995). Lipotoxicity in the pathogenesis of obesity-dependent NIDDM. Genetic 
and clinical implications. Diabetes, 44(8), 863-870. 
Unger, R. H., & Zhou, Y. T. (2001). Lipotoxicity of B-cells in obesity and in other causes of 
fatty acid spillover. Diabetes, 50(Suppl 1), S118-S121. 
Utter, J., Scragg, R., Schaaf, D., & Fitzgerald, E. (2006). Nutrition and physical activity 
behaviours among Māori, Pacific and NZ European children: identifying opportunities 
for population-based interventions. Australian and New Zealand Journal of Public 
Health, 30(1), 50-56. 
Uusitupa, M., Lindi, V., Louheranta, A., Salopuro, T., Lindstrom, J., & Tuomilehto, J. (2003). 
Long-term improvement in insulin sensitivity by changing lifestyles of people with 
impaired glucose tolerance: 4-year results from the Finnish Diabetes Prevention Study. 
Diabetes, 52(10), 2532-2538. 
van der Ploeg, G. E., Gunn, S. M., Withers, R. T., Modra, A. C., & Crockett, A. J. (2000). 
Comparison of two hydrodensitometric methods for estimating percent body fat. 
Journal of Applied Physiology, 88(4), 1175-1180. 
van Loon, L. J. C. (2004). Intramyocellular triacylglycerol as a substrate source during 
exercise. Proceedings of the Nutrition Society 63, 301-307. 
van Loon, L. J. C. (2004). Use of intramuscular triacylglycerol as a substrate source during 
exercise in humans. Journal of Applied Physiology, 97(4), 1170-1187. 
van Loon, L. J. C., Greenhaff, P. L., Constantin-Teodosiu, D., Saris, W. H. M., & 
Wagenmakers, A. J. M. (2001). The effects of increasing exercise intensity on muscle 
fuel utilisation in humans. The Journal of Physiology, 536(1), 295. 
Wade, A. J., Marbut, M. M., & Round, J. M. (1990). Muscle fibre type and aetiology of 
obesity. Lancet(British edition), 335(8693), 805-808. 
Wallace, T. M., & Matthews, D. R. (2002). The assessment of insulin resistance in man. 
Diabetic Medicine, 19(7), 527-534. 
Wang, J., Thornton, J. C., Russell, M., Burastero, S., Heymsfield, S., & Pierson Jr, R. N. 
(1994). Asians have lower body mass index (BMI) but higher percent body fat than do 
whites: comparisons of anthropometric measurements. American Journal of Clinical 
Nutrition, 60(1), 23. 
 References 
 206
Warram, J. H., Martin, B. C., Krolewski, A. S., Soeldner, J. S., & Kahn, C. R. (1990). Slow 
glucose removal rate and hyperinsulinemia precede the development of type II 
diabetes in the offspring of diabetic parents. Annals of Internal Medicine, 113(12), 
909. 
Wei, M., Gaskill, S. P., Haffner, S. M., & Stern, M. P. (1997). Waist circumference as the 
best predictor of noninsulin dependent diabetes mellitus (NIDDM) compared to body 
mass index, waist/hip ratio and other anthropometric measurements in Mexican 
Americans-a 7-year prospective study. Obesity research, 5(1), 16. 
Wei, M., Gibbons, L. W., Kampert, J. B., Nichaman, M. Z., & Blair, S. N. (2000). Low 
cardiorespiratory fitness and physical inactivity as predictors of mortality in men with 
Type 2 diabetes. Annals of Internal Medicine, 132(8), 605-611. 
Weir, G. C., & Bonner-Weir, S. (2004). Five stages of evolving beta-cell dysfunction during 
progression to diabetes. Diabetes, 53(suppl 3), S16. 
Welch, B. E., & Crisp, C. E. (1958). Effect of the level of expiration on body density 
measurement. Journal of Applied Physiology, 12(3), 399. 
Whitaker, R. C., Wright, J. A., Pepe, M. S., Seidel, K. D., & Dietz, W. H. (1997). Predicting 
obesity in young adulthood from childhood and parental obesity. New England 
Journal of Medicine, 337(13), 869-873. 
WHO. (1946). Preamble to the Constitution of the World Health Organization as adopted by 
the International Health Conference. New York: WHO. 
WHO. (1995). Physical status: the use and interpretation of anthropometry. Geneva: World 
Health Organisation. 
WHO. (1998). The World Health Report 1998. Life in the 21st Century - A Vision for All. 
Geneva: World Health Organisation. 
WHO. (2004). Appropriate body-mass index for Asian populations and its implications for 
policy and intervention strategies. Lancet, 363(9403), 157. 
Wilcox, G. (2005). Insulin and insulin resistance. The Clinical Biochemist Reviews, 26, 19-39. 
Wild, S. H., Roglic, G., Green, A., Sicree, R., & King, H. (2004). Global prevalence of 
diabetes: estimates for the year 2000 and projections for 2030. Diabetes care, 27(10), 
2569. 
Wilmore, J. H. (1969). A simplified method for determination of residual lung volumes. 
Journal of Applied Physiology, 27(1), 96-100. 
Wilmore, J. H., Costill, D. L., & Kenney, W. L. (2005). Physiology of sport and exercise. 
Champaign, Illinois: Human Kinetics. 
 References 
 207
Wilmore, J. H., Vodak, P. A., Parr, R. B., Girandola, R. N., & Billing, J. E. (1980). Further 
simplification of a method for determination of residual lung volume. Medicine and 
Science in Sports and Exercise, 12(3), 216-218. 
Windisch, A., Gundersen, K., Szabolcs, M. J., Gruber, H., & Lømo, T. (1998). Fast to slow 
transformation of denervated and electrically stimulated rat muscle. The Journal of 
Physiology, 510(2), 623. 
Winegrad, A. I., & Renold, A. E. (1958). Studies on rat adipose tissue in vitro. Journal of 
Biological Chemistry, 233(2), 267. 
Yeh, M. P., Adams, T. D., Gardner, R. M., & Yanowitz, F. G. (1987). Turbine flowmeter vs. 
Fleisch pneumotachometer: a comparative study for exercise testing. Journal of 
Applied Physiology, 63(3), 1289-1295. 
Yki-Jarvinen, H., & Koivisto, V. A. (1983). Effects of body composition on insulin 
sensitivity. Diabetes, 32(10), 965-969. 
Young, T. K., Reading, J., Elias, B., & O'Neil, J. D. (2000). Type 2 diabetes mellitus in 
Canada's First Nations: status of an epidemic in progress. Canadian Medical 
Association Journal, 163(5), 561-566. 
Zhu, S., Wang, Z., Heshka, S., Heo, M., Faith, M., & Heymsfield, S. (2002). Waist 
circumference and obesity-associated risk factors among whites in the third National 
Health and Nutrition Examination Survey: clinical action thresholds. American 
Journal of Clinical Nutrition, 76, 743-749. 
Zierath, J., & Hawley, J. (2004). Skeletal muscle fiber type: influence on contractile and 
metabolic properties. Public Library of Science Biology, 2(10), 1523-1527. 
Zimmet, P. (1979). Epidemiology of diabetes and its macrovascular manifestations in Pacific 
populations: the medical effects of social progress. Diabetes Care, 2(2), 144-153. 
 
 
  
I
Appendices  
10.5 Appendix A – Study 1 
Table 0-1 – Participant physical characteristics 
 Sydney Study Wellington Study  
Name Age Height Weight BMI % BF (Siri) LBM Age Height Weight BMI % BF (Siri) LBM 
LB 20 1.74 75.80 25.2 21.9 59.20 22 1.74 70.50 23.4 16.5 58.40 
RW 21 1.82 73.60 22.2 16.2 61.68 23 1.82 88.20 26.6 19.5 63.40 
TD 20 1.89 93.50 26.2 11.7 82.56 22 1.89 99.50 27.9 13.0 84.75 
SW 23 1.81 103.10 31.5 36.7 65.26 25 1.81 94.50 28.8 29.5 66.25 
NB 30 1.78 80.85 25.5 18.7 65.73 32 1.78 75.00 23.7 13.0 65.00 
RC 30 1.82 82.95 25.2 17.7 68.27 32 1.82 90.00 27.2 23.5 68.80 
CH 33 1.77 79.00 25.2 20.0 63.20 37 1.80 102.70 31.7 20.3 81.40 
BC 34 1.80 91.00 28.1 17.7 74.89 36 1.83 91.40 27.3 22.0 70.65 
JH 34 1.83 90.00 27.0 24.6 67.86 35 1.77 79.10 25.2 19.5 63.48 
TK 37 1.74 95.40 31.5 27.4 69.26 38 1.74 94.70 31.3 26.0 69.95 
JG 31 1.79 90.50 28.2 14.0 77.83 32 1.79 85.50 26.7 7.5 79.35 
JR 25 1.82 152.00 45.9 42.5 87.40 27 1.82 152.70 46.1 40.0 91.20 
Mean 28 2 92 28 22 70 30 1.80 93.65 28.8 20.9 71.89 
 App
endix
 A
 
  
II
Table 0-2 – Blood Pathology (Sydney) 
Name Ins pM Ins 2 hr Glucose 2 hr Glucose Total Chol HDL LDL Chol/LDL Ins iU HOMA 
LB 53 114 4.5 4.4 5.1 1.30 2.6 3.9 7.63 1.53 
RW 21 108 4.5 4.7 4.7 1.25 3.8 2.4 3.02 0.60 
TD 39 109 6.1 3.2 - - - - 5.62 1.52 
SW 128 34 4.6 2.8 5.5 1.42 3.6 3.9 18.43 3.77 
NB 67 136 5.5 5.4 6.1 1.27 3.7 4.8 9.65 2.36 
RC 53 14 4.8 3.8 8.6 1.54 6.3 5.6 7.63 1.63 
CH 71 191 5.5 5.2 4.0 1.02 2.7 3.9 10.22 2.50 
BC 52 29 5.4 2.7 3.5 1.22 2.0 2.9 7.49 1.80 
JH 46 114 5.0 4.3 4.7 1.13 3.2 4.2 6.62 1.47 
TK 43 249 4.9 4.1 5.6 1.01 4.1 5.5 6.19 1.35 
JG 65 - 5.0 3.3 5.0 1.00 2.8 4.9 9.36 2.08 
JR 264 184 5.7 4.3 5.8 1.37 3.8 4.2 38.01 9.63 
Mean 51 117 5 4 5 1 3 4 11 3 
 
App
endix
 A
 
  
III
Table 0-3 – Blood Pathology (Wellington) 
Name Ins pM Ins 2 hr Glucose 2 hr Glucose Total Chol HDL LDL Chol/LDL Ins iU HOMA 
LB 36 49 5.1 3.9 4.5 1.50 2.30 3.0 5.2 1.17 
RW 54 65 4.2 2.5 5.0 1.13 3.30 4.4 7.8 1.45 
TD 32 - 4.9 - 3.9 2.59 1.20 1.5 4.6 1.00 
SW 85 165 4.5 3.4 4.6 1.72 2.40 2.7 12.2 2.45 
NB 32 213 4.9 6.3 6.3 1.40 4.00 4.5 4.6 1.00 
RC 28 78 4.7 4.2 7.9 1.42 5.40 5.6 4.0 0.84 
CH 54 114 5.0 4.3 4.6 0.97 2.70 4.7 7.8 1.73 
BC 31 42 5.1 3.1 4.9 1.19 3.40 4.1 4.5 1.01 
JH 34 155 4.7 5.2 3.8 1.03 2.60 3.7 4.9 1.02 
TK 72 96 4.9 4.1 4.6 0.92 3.10 5.0 10.4 2.26 
JG 46 152 5.4 4.6 5.2 0.95 3.40 5.5 6.6 1.59 
JR 195 804 5.0 5.5 5.6 1.20 3.60 4.7 28.1 6.24 
Mean 58.3 176 4.9 4.3 5.1 1.34 3.12 4.1 8.4 1.81 
 
App
endix
 A
 
  
IV
Table 0-4 – Aerobic capacity 
 Sydney Wellington 
Name VO2max VO2max/kg VO2max/LBM VO2max Vo2max/kg VO2max/LBM 
LB 3550 46.83 59.96 3269 46.37 55.97 
RW 3833 52.09 62.15 3571 40.49 56.32 
TD 4567 48.85 55.32 3748 37.67 44.22 
SW 4202 40.75 64.38 3641 38.53 54.96 
NB 3796 46.95 57.75 3161 42.15 48.63 
RC 4409 53.15 64.58 3363 37.37 48.88 
CH 3165 40.06 50.08 4151 40.42 51.00 
BC 4979 54.71 66.48 4104 44.90 58.09 
JH 4375 48.61 64.47 3862 48.82 60.84 
TK 3843 40.28 55.49 2802 29.59 40.06 
JG 4212 46.54 65.09 3762 44.00 47.41 
JR 4075 26.81 46.62 3340 21.87 36.62 
Mean 4084 45.47 59.36 3564 39.35 50.25 
 
App
endix
 A
 
  
V
Table 0-5 – Multiple regression analysis (Sydney) 
Dependant variable Independent variable r r² t P 
HOMA-IR 
 0.847 0.717  0.003 
% body fat   3.417 0.008 
VO2maxLBM   -1.797 0.106 
Fasting serum insulin 
 0.861 0.741  0.002 
% body fat   3.952 0.003 
VO2maxLBM   -1.452 0.181 
Fasting serum glucose 
 0.65 0.423  0.084 
% body fat   -1.287 0.230 
VO2maxLBM   -2.544 0.032 
2hr serum insulin 
 0.781 0.610  0.023 
% body fat   -0.060 0.954 
VO2maxLBM   -3.366 0.010 
2hr serum glucose 
 0.543 0.295  0.207 
% body fat   -0.774 0.459 
VO2maxLBM   -1.942 0.084 
 
App
endix
 A
 
  
V
I
Table 0-6 – Multiple regression analysis (Wellington only) 
Dependant variable Independent variable r r² t P 
HOMA-IR 
 0.873 0.763  0.002 
% body fat   3.829 0.004 
VO2maxLBM   -2.388 0.041 
Fasting serum insulin 
 0.879 0.772  0.001 
% body fat   4.116 0.003 
VO2maxLBM   -2.214 0.054 
Fasting serum glucose 
 0.508 0.258  0.261 
% body fat   -1.391 0.198 
VO2maxLBM   -1.473 0.175 
2hr serum insulin 
 0.764 0.584  0.030 
% body fat   1.824 0.106 
VO2maxLBM   -1.794 0.111 
2hr serum glucose 
 0.496 0.246  0.324 
% body fat   -0.801 0.446 
VO2maxLBM   -1.608 0.146 
 
App
endix
 A
 
  
V
II
10.6 Appendix B – Study 2 
Table 0-7 – Participant physical characteristics (non-Māori) 
Name Age Height Weight BMI % BF (Siri) LBM 
W 32 1.81 69.3 21.2 16.3 57.8 
WR 25 1.94 75.8 20.1 13.5 65.5 
NC 23 1.74 86.7 28.6 21.5 68.1 
SW 24 1.83 90.9 27.1 20.5 72.3 
SC 25 1.82 87.6 26.4 19.0 70.9 
GE 32 1.67 62.4 22.4 8.0 57.2 
DM 35 1.84 83.2 24.6 22.0 64.4 
CM 29 1.81 84.5 25.8 14.0 72.4 
MV 24 1.79 81.9 25.6 13.0 71.0 
AR 35 1.75 76.0 24.8 22.0 59.5 
KM 18 1.78 66.2 20.9 5.5 62.4 
JG 28 1.81 70.7 21.6 15.0 53.5 
TM 28 1.67 62.8 22.5 15.0 59.8 
BG 28 1.75 72.0 23.5 18.0 59.0 
CB 20 1.85 98.2 28.7 25.0 73.2 
KM 19 1.73 72.5 24.2 17.0 60.1 
JM 28 1.78 68.6 21.7 16.0 56.7 
Mean 28 1.81 80.1 24.5 16.9 66.1 
 
App
endix
 B
 
  
V
III
Table 0-8 – Participant physical characteristics (Māori) 
Name Age Height Weight BMI % BF (Siri) LBM 
RP 26 1.78 80.0 25.2 24.5 60.6 
ES 25 1.81 95.2 29.1 21.7 74.3 
AW 32 1.84 115.4 34.1 28.7 82.3 
KT 36 1.81 97.8 29.9 30.7 68.2 
WM 28 1.79 94.7 29.6 25.3 71.0 
KT 37 1.81 101.2 30.9 11.7 89.1 
PT 26 1.84 83.4 24.6 15.0 71.0 
CT 23 1.88 94.6 26.8 21.0 74.7 
BE 34 1.82 85.4 25.8 20.0 68.3 
IW 25 1.95 110.2 29.0 28.0 79.7 
LR 25 1.70 111.3 38.5 33.0 74.4 
ND 30 1.70 84.8 29.3 23.3 65.0 
LT 28 1.87 118.6 33.9 30.5 82.2 
ST 23 1.81 82.2 25.1 15.5 71.8 
CW 22 1.83 88.6 26.5 22.0 68.8 
CW 23 1.79 100.0 31.2 21.0 79.7 
NP 32 1.59 61.0 24.1 14.0 52.6 
JC 30 1.87 107.3 30.7 22.0 83.9 
HM 18 1.74 69.4 22.9 8.0 63.7 
Mean 28 1.80 94.4 29.0 22.7 72.6 
 
App
endix
 B
 
  
IX
Table 0-9 – Blood Pathology (Māori) 
Name Ins pM Ins 2 hr Glucose 2 hr Glucose Total Chol HDL LDL Chol/LDL Ins iU HOMA 
RP 75 118 4.0 5.7 - - - - 10.8 1.92 
ES 32 32 4.7 5.0 - - - - 4.6 0.96 
AW 111 308 5.1 5.3 5.8 0.9 3.8 6.2 16.0 3.62 
KT 46 99 4.1 5.1 3.9 1.4 2.2 2.8 6.6 1.21 
WM 53 68 4.9 2.6 - - - - 7.6 1.66 
KT 138 329 5.6 4.6 - - - - 19.9 4.95 
PT 40 41 4.5 3.5 - - - - 5.8 1.15 
CT 35 127 4.4 - - - - - 5.0 0.99 
BE 35 61 4.2 5.6 - - - - 5.0 0.94 
IW 33 39 3.6 5.3 6.6 1.1 4.6 5.8 4.8 0.76 
LR 103 453 5.1 6.5 4.0 0.8 2.6 4.9 14.8 3.36 
ND 52 80 5.1 3.3 - - - - 7.5 1.70 
LT 105 - 5.4 4.5 - - - - 15.1 3.63 
ST 41 - 4.6 2.7 - - - - 5.9 1.21 
CW 106 - 4.8 4.2 - - - - 15.3 3.26 
CW 76  - 4.7 -  - - - - 10.9 2.29 
NP 12 136 4.6 4.2 5.8 1.1 4.4 5.3 1.7 0.35 
JC 40 202 5.3 - 5.5 0.8 3.9 6.8 5.8 1.36 
HM 33 156 5.2 4.1 3.4 0.4 2.7 8.0 4.8 1.10 
Mean 63.6 147.8 4.7 4.5 5.2 1.1 3.5 5.0 9.2 2.00 
 
App
endix
 B
 
  
X
Table 0-10 – Blood Pathology (non-Māori) 
Name Ins pM Ins 2 hr Glucose 2 hr Glucose Total Chol HDL LDL Chol/LDL Ins iU HOMA 
W 33 - 5.0 - - - - - 4.8 1.07 
WR 81 90 4.7 4.5 3.9 1.0 2.5 4.1 11.7 2.44 
NC 81 264 4.6 5.0 4.8 1.1 3.2 4.4 11.7 2.38 
SW 41 17 4.9 6.1 5.5 1.2 3.7 4.6 5.9 1.29 
SC 47 158 4.8 4.0 4.8 0.7 3.5 7.2 6.8 1.44 
GE 41 138 5.0 4.2 4.3 1.0 3.0 4.5 5.9 1.31 
DM 50 42 4.8 2.7 4.2 1.6 2.3 2.6 7.2 1.54 
CM 165 - 5.2 3.8 3.3 0.8 2.1 4.3 23.8 5.49 
MV 42 - 4.8 4.3 5.2 0.9 4.1 6.2 6.0 1.29 
AR 8 32 4.5 1.7 4.4 1.0 3.1 4.4 1.2 0.23 
KM 29 199 4.5 4.5 4.5 1.0 3.1 4.4 4.2 0.84 
JG 16 57 4.5 3.8 4.4 1.1 3.0 4.0 2.3 0.46 
TM 13 54 4.7 3.5 4.3 1.3 2.6 3.4 1.9 0.39 
BG 49 209 5.2 4.1 4.1 0.7 3.0 5.5 7.1 1.63 
CB 56 49 5.0 3.2 4.8 1.0 3.3 4.9 8.1 1.79 
KM 22 113 4.9 4.7 4.5 1.0 2.9 4.7 3.2 0.69 
JM 17 122 4.5 4.1 4.6 1.1 3.2 4.4 2.4 0.49 
Mean 47 110 4.8 4.0 4.5 1.0 3.0 4.6 6.7 1.52 
 
App
endix
 B
 
  
X
I
Table 0-11 – Aerobic capacity (Māori) 
Name VO2max Vo2max/kg VO2max/LBM 
RP 3520 44.00 58.09 
ES 3852 40.46 51.84 
AW 4674 40.50 56.79 
KT 4108 42.00 60.23 
WM 4728 49.93 66.59 
KT 4007 39.59 44.97 
PT 3460 41.49 48.73 
CT 3649 38.57 48.85 
BE 3868 45.29 56.63 
IW 4506 40.89 56.54 
LR 3667 32.95 49.29 
ND 3940 46.46 60.62 
LT 3808 32.11 46.33 
ST 3792 46.13 52.81 
CW 4000 45.15 58.14 
CW 4139 41.39 51.93 
NP 3267 53.59 62.11 
JC 5136 47.87 61.22 
HM 3075 44.30 48.27 
Mean 3982 42.70 54.55 
 
App
endix
 B
 
  
X
II
Table 0-12 – Aerobic capacity (non-Māori) 
Name VO2max Vo2max/kg VO2max/LBM 
W 3846 55.50 66.54 
WR 3408 44.96 52.03 
NC 4213 48.59 61.86 
SW 4672 51.40 64.62 
SC 4001 45.67 56.43 
GE 3221 51.62 56.31 
DM 3481 41.84 54.05 
CM 4059 48.04 56.06 
MV 3704 45.23 52.17 
AR 3572 47.00 60.03 
KM 2723 41.13 43.64 
JG 4591 64.92 85.81 
TM 4244 67.58 70.97 
BG 3521 48.90 59.68 
CB 4647 47.32 63.48 
KM 3788 52.25 63.03 
JM 3802 55.42 67.05 
Mean 3853 50.43 60.81 
 
 
 
App
endix
 B
 
  
X
III
10.7 Appendix C – Study 3 
Table 0-13 – non-Māori participants 
Name Age Height Weight BMI % BF UWW LBM % BF BIA BIA Lean Mass 
W 32 1.81 69.3 21.2 16.3 57.8 14.0 59.6 
WR 25 1.94 75.8 20.1 13.5 65.5 9.5 68.6 
NC 23 1.74 86.7 28.6 21.5 68.1 17.6 71.2 
SW 24 1.83 90.9 27.1 20.5 72.3 14.6 77.4 
SC 25 1.82 87.6 26.4 19.0 70.9 18.7 71.2 
GE 32 1.67 62.4 22.4 8.0 57.2 12.0 54.6 
DM 35 1.84 83.2 24.6 22.0 64.4 15.2 70.7 
CM 29 1.81 84.5 25.8 14.0 72.4 12.4 72.2 
MV 24 1.79 81.9 25.6 13.0 71.0 13.1 72.8 
AR 35 1.75 76.0 24.8 22.0 59.5 9.6 65.2 
KM 18 1.78 66.2 20.9 5.5 62.4 10.8 58.3 
JG 28 1.81 70.7 21.6 15.0 53.5 8.4 64.2 
TM 28 1.67 62.8 22.5 15.0 59.8 13.5 54.1 
BG 28 1.75 72.0 23.5 18.0 59.0 21.4 57.0 
DN 26 1.82 71.5 21.6 21.3 56.3 13.1 62.1 
JM 28 1.78 68.6 21.7 16.0 56.7 17.3 56.8 
AW 26 1.82 78.1 23.6 8.5 71.3 10.6 69.8 
KE 23 1.85 88.8 25.9 19.5 71.5 17.3 73.4 
KC 21 1.78 71.6 22.6 9.7 68.4 7.1 66.5 
JH 28 1.81 86.8 26.5 19.7 70.0 15.9 73.0 
Mean 27 1.79 76.8 23.9 15.9 64.4 13.6 65.9 
 
App
endix
 C
 
  
X
IV
Table 0-14 – Māori participants 
Name Age Height Weight BMI % BF UWW LBM % BF BIA BIA Lean Mass 
RP 26 1.78 80.0 25.2 24.5 60.6 16.7 66.6 
ES 25 1.81 95.2 29.1 21.7 74.3 22.4 73.9 
AW 32 1.84 115.4 34.1 28.7 82.3 24.8 86.8 
KT 36 1.81 97.8 29.9 30.7 68.2 25.2 73.1 
WM 28 1.79 94.7 29.6 25.3 71.0 22.8 73.2 
KT 37 1.81 101.2 30.9 11.7 89.1 29.2 71.7 
PT 26 1.84 83.4 24.6 15.0 71.0 10.9 74.3 
CT 23 1.88 94.6 26.8 21.0 74.7 14.7 80.7 
BE 34 1.82 85.4 25.8 20.0 68.3 12.8 74.4 
PH 28 1.84 97.1 28.7 20.7 76.9 14.7 82.8 
AR 20 1.76 112.0 36.2 25.5 83.8 26.3 82.5 
IW 25 1.95 110.2 29.0 28.0 79.7 18.0 90.1 
ND 30 1.70 84.8 29.3 23.3 65.0 20.0 68.0 
LT 28 1.87 118.6 33.9 30.5 82.2 24.2 89.6 
ST 23 1.81 82.2 25.1 15.5 71.8 11.2 72.7 
CW 22 1.83 88.6 26.5 22.0 68.8 17.1 72.8 
CW 23 1.79 100.0 31.2 21.0 79.7 21.4 78.0 
NP 32 1.59 61.0 24.1 14.0 52.6 10.3 54.3 
HM 18 1.74 69.4 22.9 8.0 63.7 10.5 61.4 
SB 22 1.77 99.1 31.6 31.3 67.9 31.2 68.2 
DA 26 1.79 118.0 36.8 29.5 83.6 27.0 86.7 
JE 20 1.76 107.5 34.7 28.5 77.1 32.0 73.0 
DP 27 1.88 84.8 24.0 18.0 69.6 13.7 73.2 
LB 22 1.74 70.50 23.4 16.5 58.40 16.6 58.9 
RW 23 1.82 88.20 26.6 19.5 63.40 16.1 72.0 
SW 25 1.81 94.50 28.8 29.5 66.25 23.3 72.0 
BC 36 1.83 91.40 27.3 22.0 70.65 19.9 72.9 
JR 27 1.82 152.70 46.1 40.0 91.20 44.4 84.6 
Mean 27 1.80 95.7 29.4 22.9 72.6 20.6 74.6 App
endix
 C
 
  
X
V
10.8 Appendix D – Study 4 
Table 0-15 – Participant physical characteristcs 
Name Age Height Weight BMI % BF  
NP 32 1.59 61.0 24.1 14.0 
HM 18 1.74 69.4 22.9 8.0 
WR 25 1.94 75.8 20.1 13.5 
JM 28 1.78 68.6 21.7 16.0 
DN 26 1.82 71.5 21.6 21.3 
IW 25 1.95 110.2 29.0 28.0 
GE 32 1.67 62.4 22.4 8.0 
CM 29 1.81 84.5 25.8 14.0 
TM 28 1.67 62.8 22.5 15.0 
Mean 27 1.77 74.0 23.3 15.3 
 
App
endix
 D
 
  
X
V
I
Table 0-16 – T50 & TPF  
  Day 1 Day 2 Day 3 
Name Twitch# T50 (Lillioja et al.) TPF (Lillioja et al.) T50 TPF T50 TPF 
CM 1 0.065 0.109 0.061 0.101 0.064 0.104 
 2 0.065 0.109 0.061 0.101 0.063 0.105 
 3 0.065 0.109 0.061 0.102 0.063 0.105 
 4 0.065 0.108 0.061 0.101 0.063 0.104 
 5 0.064 0.107 0.061 0.101 0.063 0.105 
GE 1 0.062 0.104 0.060 0.106 0.062 0.101 
 2 0.062 0.104 0.058 0.100 0.062 0.101 
 3 0.062 0.103 0.058 0.100 0.062 0.101 
 4 0.062 0.104 0.059 0.099 0.061 0.100 
 5 0.061 0.103 0.058 0.100 0.061 0.101 
TM 1 0.064 0.105 0.065 0.108 0.066 0.109 
 2 0.064 0.104 0.064 0.107 0.066 0.111 
 3 0.064 0.105 0.065 0.107 0.066 0.109 
 4 0.064 0.104 0.065 0.108 0.066 0.109 
 5 0.065 0.107 0.065 0.107 0.065 0.106 
DN 1 0.066 0.106 0.073 0.106 0.068 0.108 
 2 0.066 0.104 0.074 0.105 0.067 0.106 
 3 0.066 0.102 0.073 0.105 0.067 0.105 
 4 0.066 0.106 0.074 0.106 0.066 0.106 
 5 0.066 0.106 0.074 0.106 0.066 0.106 
HM 1 0.056 0.098 0.059 0.097 0.060 0.100 
 2 0.058 0.102 0.060 0.101 0.060 0.101 
 3 0.061 0.106 0.061 0.101 0.060 0.100 
 4 0.058 0.108 0.062 0.100 0.060 0.102 
 5 0.058 0.107 0.060 0.099 0.060 0.100 
 
App
endix
 D
 
  
X
V
II
Table 0-17 – T50 & TPF (continued) 
  Day 1 Day 2 Day 3 
Name Twitch# T50 (Lillioja et al.) TPF (Lillioja et al.) T50 TPF T50 TPF 
WR 1 0.062 0.100 0.059 0.099 0.065 0.104 
 2 0.062 0.099 0.058 0.097 0.065 0.105 
 3 0.061 0.098 0.059 0.098 0.065 0.104 
 4 0.062 0.100 0.058 0.099 0.065 0.106 
 5 0.062 0.099 0.058 0.098 0.065 0.105 
NP 1 0.062 0.134 0.061 0.138 0.062 0.135 
 2 0.062 0.133 0.063 0.141 0.062 0.136 
 3 0.062 0.134 0.063 0.142 0.062 0.136 
 4 0.062 0.135 0.062 0.141 0.062 0.135 
 5 0.062 0.134 0.062 0.136 0.062 0.133 
JM 1 0.060 0.106 0.062 0.106 0.062 0.105 
 2 0.060 0.106 0.062 0.108 0.062 0.105 
 3 0.060 0.106 0.062 0.109 0.062 0.105 
 4 0.060 0.106 0.062 0.110 0.062 0.107 
 5 0.060 0.108 0.062 0.109 0.062 0.103 
IW 1 0.065 0.115 0.064 0.115 0.064 0.113 
 2 0.064 0.113 0.066 0.113 0.064 0.112 
 3 0.066 0.115 0.066 0.113 0.064 0.113 
 4 0.066 0.113 0.065 0.113 0.064 0.113 
 5 0.065 0.114 0.065 0.113 0.064 0.112 
 
App
endix
 D
 
  
X
V
III
Table 0-18 – ANOVA table for TPF & T50 
ANOVA for TTP 
Source of variation Degrees of freedom SSQ MSQ F Significance (p) 
Between subjects 8 0.0145758 0.0018220 61.1409 <0.001 
Between days 2 0.0000279 0.0000140 0.4698 NS 
Subject/Day Error (1) 16 0.0004772 0.0000298 17.5294 <0.001 
Between trials 26 0.0150809 0.0005800   
Test 4 0.0000061 0.0000015 0.8824 NS 
Subject/Test 32 0.0001018 0.0000032 1.8824 NS 
Day/Test 8 0.0000164 0.0000020 1.1765 NS 
Error (2) 64 0.0001098 0.0000017   
Total 134 0.0153151    
ANOVA for T50 
Between subjects 8 0.0009534 0.0001192 5.2743 <0.01 
Between days 2 0.0000297 0.0000148 0.6549 NS 
Subject/Day Error (1) 16 0.0003621 0.0000226 75.3333 <0.001 
Between trials 26 0.0013452 0.0000517   
Test 4 0.0000021 0.0000005 1.6667 NS 
Subject/Test 32 0.0000142 0.0000004 1.3333 NS 
Day/Test 8 0.0000040 0.0000005 1.6667 NS 
Error (2) 64 0.0000217 0.0000003   
Total 134     
  
App
endix
 D
 
  
X
IX
10.9  Appendix E – Study 5 
Table 0-19 – Participant physical characteristics 
Name Age Height Weight BMI % BF  
WM 28 1.79 94.7 29.6 25.3 
PT 26 1.84 83.4 24.6 15.0 
CT 23 1.88 94.6 26.8 21.0 
LT 28 1.87 118.6 33.9 30.5 
ST 23 1.81 82.2 25.1 15.5 
NP 32 1.59 61.0 24.1 14.0 
HM 18 1.74 69.4 22.9 8.0 
WR 25 1.94 75.8 20.1 13.5 
JM 28 1.78 68.6 21.7 16.0 
DN 26 1.82 71.5 21.6 21.3 
KT 36 1.81 97.8 29.9 30.7 
KT 37 1.81 101.2 30.9 11.7 
BE 34 1.82 85.4 25.8 20.0 
IW 25 1.95 110.2 29.0 28.0 
LR 25 1.70 111.3 38.5 33.0 
GE 32 1.67 62.4 22.4 8.0 
CM 29 1.81 84.5 25.8 14.0 
MV 24 1.79 81.9 25.6 13.0 
KM 18 1.78 66.2 20.9 5.5 
TM 28 1.67 62.8 22.5 15.0 
BG 29 1.75 72.0 23.5 18.0 
Mean 27 1.79 83.6 26.0 18.0 
 
App
endix
 E
 
  
X
X
Table 0-20 – Blood pathology 
Name Ins pM Ins 2 hr Glucose 2 hr Gluc HOMA 
WM 53 68 4.9 2.6 1.66 
PT 40 41 4.5 3.5 1.15 
CT 35 127 4.4 - 0.99 
LT 105 - 5.4 4.5 3.63 
ST 41 - 4.6 2.7 1.21 
NP 12 136 4.6 4.2 0.35 
HM 33 156 5.2 4.1 1.10 
WR 81 90 4.7 4.5 2.44 
JM 17 122 4.5 4.1 0.49 
DN 56 271 5.5 5.0 1.97 
KT 46 99 4.1 5.1 1.21 
KT 138 329 5.6 4.6 4.95 
BE 35 61 4.2 5.6 0.94 
IW 33 39 3.6 5.3 0.76 
LR 103 453 5.1 6.5 3.36 
GE 41 138 5.0 4.2 1.31 
CM 165 - 5.2 3.8 5.49 
MV 42 - 4.8 4.3 1.29 
KM 29 199 4.5 4.5 0.84 
TM 13 54 4.7 3.5 0.39 
BG 49 209 5.2 4.1 1.63 
Mean 56 152 4.8 4.3 1.77 
 
 
 
 
App
endix
 E
 
  
X
X
I
Table 0-21 – Aerobic capacity, fatigue test and twitch data 
Name VO2lbm TTP TT50% Fmax Fmean of 56-60 (T60) T60 as percentage of max (%) Diff (Fmax-T60) Percentage decline 
WM 66.59 0.115 0.065 215.5 92.9 43.1 122.6 56.9 
PT 48.73 0.101 0.060 231.7 62.0 26.7 169.7 73.3 
CT 48.85 0.106 0.064 233.4 87.8 37.6 145.6 62.4 
LT 46.33 0.127 0.068 288.2 129.6 45.0 158.6 55.0 
ST 52.81 0.106 0.060 211.9 97.6 46.0 114.3 54.0 
NP 62.11 0.134 0.062 144.4 56.5 39.2 87.9 60.8 
HM 48.27 0.104 0.058 178.3 83.7 46.9 94.6 53.1 
WR 52.03 0.099 0.062 161.0 77.9 48.4 83.1 51.6 
JM 67.05 0.106 0.060 187.9 81.8 43.6 106.1 56.4 
DN 64.69 0.106 0.066 167.5 38.0 22.7 129.5 77.3 
KT 60.23 0.116 0.067 266.6 105.9 39.7 160.7 60.3 
KT 44.97 0.110 0.066 280.3 91.3 32.6 189.0 67.4 
BE 56.63 0.111 0.064 201.8 75.1 37.2 126.7 62.8 
IW 56.54 0.114 0.065 233.9 82.8 35.4 151.1 64.6 
LR 49.29 0.113 0.071 294.0 77.5 26.3 216.5 73.7 
GE 56.31 0.104 0.062 182.4 80.6 44.2 101.8 55.8 
CM 56.06 0.108 0.065 218.7 82.0 37.5 136.7 62.5 
MV 52.17 0.101 0.063 201.0 75.6 37.6 125.4 62.4 
KM 43.64 0.104 0.063 156.3 44.7 28.6 111.6 71.4 
TM 70.97 0.105 0.064 171.7 104.9 61.1 66.8 38.9 
BG 59.68 0.107 0.065 153.5 69.5 45.3 84.0 54.7 
Mean 55.43 0.109 0.064 208.6 80.8 39.3 127.7 60.7 
App
endix
 E
 
Appendix F 
 XXII
10.10 Appendix F – Participant Information Forms & Health 
Questionairres 
 
10.11 Pre-Exercise Health Screening Questionnaire 
 
Name:______________________________________________________________ 
 
Address: ____________________________________________________________ 
 
Phone: _______________________________ 
 v 
Age: ________________ 
 
Please read the following questions carefully.  If you have any difficulty, please advise the medical practitioner, 
nurse or exercise specialist who is conducting the exercise test.  
 
Please answer all of the following questions by ticking only one box for each question: 
 
This questionnaire has been designed to identify the small number of persons (15-69 years of age) for whom physical activity 
might be inappropriate.  The questions are based upon the Physical Activity Readiness Questionnaire (PAR-Q), originally 
devised by the British Columbia Dept of Health (Canada), as revised by 1Thomas et al. (1992) and 2Cardinal et al. (1996), 
and with added requirements of the Massey University Human Ethics Committee.  The information provided by you on this 
form will be treated with the strictest confidentiality.  
 
Qu 1. Has your doctor ever said that you have a heart condition and that you should only do physical activity 
recommended by a doctor? 
 Yes  No 
 
Qu 2. Do you feel a pain in your chest when you do physical activity? 
Yes  No 
 
Qu 3. In the past month have you had chest pain when you were not doing physical activity? 
 Yes  No 
 
Qu 4. Do you lose your balance because of dizziness or do you ever lose consciousness? 
 Yes  No 
Appendix F 
 XXIII 
Qu 5. Is your doctor currently prescribing drugs (for example, water pills) for your blood pressure or heart 
condition? 
 Yes  No 
 
Qu 6. Do you have a bone or joint problem that could be made worse by vigorous exercise? 
 Yes  No 
 
Qu 7. Do you know of any other reason why you should not do physical activity? 
 Yes  No 
Qu 8. Have any immediate family had heart problems prior to the age of 60? 
 Yes  No 
 
Qu 9. Have you been hospitalised recently? 
 Yes No 
 
Qu 11. Do you have any infectious disease that may be transmitted in blood? 
 Yes No 
 
 
 
You should be aware that even amongst healthy persons who undertake regular physical activity there is a risk of sudden 
death during exercise.  Though extremely rare, such cases can occur in people with an undiagnosed heart condition.  If you 
have any reason to suspect that you may have a heart condition that will put you at risk during exercise, you should seek 
advice from a medical practitioner before undertaking an exercise test.  
 
 
I have read, understood and completed this questionnaire.  
 
Signature: ________________________________________  Date: ___________________ 
 
 
 
 
 
 
 
References 
1. Thomas S, Reading J and Shephard RJ. Revision of the Physical Activity Readiness Questionnaire (PAR-Q). Can J 
Sport Sci 17(4): 338-345.  
2. Cardinal BJ, Esters J and Cardinal MK. Evaluation of the revised physical activity readiness questionnaire in older 
adults. Med Sci Sports Exerc 28(4): 468-472 
Appendix F 
 XXIV
 
Participant Information Summary 
Insulin resistance in Māori males. 
 
Who are the researchers and where can we be contacted? 
Dr Stephen Stannard  
Institute of Food Nutrition 
& Human Health  
Massey University 
Palmerston North 
Prof. Chris Cunningham  
Research Centre for Māori Health & 
Development,  
Massey University,  
Wellington 
Isaac Warbrick 
Research Centre for Māori Health & 
Development,  
Massey University,  
Wellington 
What is this study about?  This study is about type-2 diabetes which affects Māori at a greater rate 
than non-Māori.  Specifically we will investigate whether the increased incidence of type-2 diabetes 
seen in the Maori population may involve elevated muscle lipid concentrations.  Muscle lipid is an 
important ‘fuel’ for exercising muscle and a better understanding of its role in producing insulin 
resistance may enable the development of more effective strategies for preventing type-2 diabetes.   
 
If you choose to participate in this study, you will: 
• Complete a health questionnaire 
• Complete an exercise test of your oxygen consumption on a treadmill. 
• Perform an oral glucose tolerance test. 
• Provide two 10 ml venous blood samples. 
• Have your body composition assessed via underwater weighing and Bioelectrical Impedance 
(Full details of these procedures are given in the attached pages). 
 
What are the benefits of the research? 
This research will give you the opportunity to have an input into a study which is designed to 
understand the risks Māori have in relation to type-2 diabetes. 
 
 
Appendix F 
 XXV
Your Rights as Participants  
You are under no obligation to accept this invitation.   If you decide to participate, you have the right 
to: 
• decline to answer any particular question; 
• withdraw from the study; 
• ask any questions about the study at any time during participation; 
• provide information on the understanding that your name will not be used unless you give 
permission to the researcher; 
• be given access to a summary of the project findings when it  is concluded. 
 
Project Contacts 
Please contact any of the researchers if you have any concerns or require further information. 
 
Appendix F 
 XXVI
 
Participant Full Information Sheet 
Insulin resistance in Māori males. 
 
Aims of this study 
This research will use techniques to measure the oxidative capacity and insulin resistance for young 
Māori men.  Blood samples will be taken to estimate insulin sensitivity, blood lipid profile, glucose 
tolerance, C-reactive protein and other markers of insulin resistance.   
 
Participation 
Participants in this study must be male Māori who are physically active, aged between 18 and 45.  You 
will be fully informed of your involvement in this study and will complete a Medical History 
questionnaire and a Physical Activity Readiness questionnaire (PAR Q) to assess your readiness to 
participate in the study.  Following these, you will be required to sign a consent form after reading this 
“Participant Information Statement” and clarifying any matters that remain unclear.   
 
You will be asked to undergo one maximal exercise bout, two measures of body composition, an oral 
glucose tolerance test, and to give two venous blood samples.  It should be emphasised that you may 
withdraw from this study any time you wish.  If you choose to do so, there will be no compromise in 
the relationship between you and the investigators.  If you are a Massey University student, 
participation in, or withdrawal from the study will not influence your academic progress in any 
manner.  In deciding to partake in this study, you will be subjecting yourself to the following 
requirements: 
 
1.  Screening Procedures 
You will answer a demographic, Medical History and Physical Activity Readiness questionnaire.   
 
2.  Testing Procedures 
Body Composition:  You will attend the RCMHD Research Laboratory at Massey University, 
Wellington Campus and have your body composition assessed in an underwater weighing 
tank.  This will require you to dress down to a pair of swimming shorts only and sit 
immerse yourself in a warm water tank.  If you feel uncomfortable, you can exit the 
tank anytime during the procedure at your own request.   
Appendix F 
 XXVII 
 
VO2max test: After the underwater weighing measurement, you will undergo an exercise test on a 
treadmill to measure your maximal oxygen uptake (VO2max). You will begin the test by walking 
on the treadmill at a slow speed.  Each two minutes the treadmill speed and gradient will be 
increased until you are unable to continue.   
During exercise, your oxygen consumption will be measured by collected expired gases from 
your mouth using the Moxus gas analysis system.  This equipment measures volumes of air and 
oxygen passing into and out of your lungs and from this we can determine the rate at which you 
use oxygen.  Heart rate is measured electronically using the commonly available Polar chest 
band and wristwatch combination.  All exercise testing will be performed on the research 
quality Woodway treadmill located in the RCMHD Research Laboratory at Massey University, 
Wellington Campus.   
 
Oral glucose tolerance test and venous blood sample:  Prior to this test, a resting 10 ml blood 
sample will be taken from a vein in your arm by an experienced staff member.  The sample will 
be frozen and later analysed for blood lipids, glucose, C-reactive protein and insulin.  The 
glucose tolerance test involves drinking a sugary (glucose) solution before breakfast.  One 
further blood sample will be taken exactly two hours after drinking the solution to test your 
blood glucose and insulin levels again.  Following analysis, any remaining blood will be 
returned to the earth by our kaumätua following due tikanga. 
 
Potential Risks and Discomforts 
As with any research of this nature, there are some potential risks and discomforts, which you should 
be aware of.  The researchers will attempt to minimise these through careful, consistent monitoring of 
your response to the testing procedures.  Every effort will be made by the researchers to ensure your 
safety, comfort and familiarity with the testing procedures.   
 
a. Blood test and oral glucose tolerance test:  Sampling of venous blood may cause a degree of 
transient discomfort, and some bruising may occur at the point of insertion up to 48 hours afterwards.  
This discomfort is of no lasting consequence.  There is also a slight risk of fainting during blood 
sampling.  The risks and discomforts will be minimised, as the procedure will be performed under 
sterile conditions by highly experienced staff.   
 
b. Exercise procedure:  The exercise procedure during the VO2max test will not be easy, but the time of 
exercise at high intensity is short (less than 2 minutes) and you are able to terminate the test at any 
time at which you feel fatigued.  This test is often performed on both athletes and sedentary 
Appendix F 
 XXVIII
individuals to measure aerobic fitness.  As in any physical activity, there is a very small possibility of 
injuries that include, but are not restricted to, muscle, ligament or tendon damage, breathing 
irregularities and dizziness.  However, the exercise protocols are commonly performed in exercise 
physiology laboratories and potential risks to participants have been minimised.   
 
c. Body composition.  There are no risks associated with these techniques.  During the underwater 
weighing you will be fully submerged in a warm water bath, and must hold your breath for about 20 
second whilst the measurements are taken.  This is a little disconcerting and if you feel uncomfortable 
you are free to exit the water at any time.  During this procedure you will need to wear a swimming 
costume.   
 
Right of withdrawal 
Your participation in this study is completely voluntary.  You may choose to withdraw from the study 
at any time for any reason.  By signing the consent form you are indicating that the tests and 
procedures of this study have been explained to you and understood by you.   
 
Inquiries 
All inquiries regarding requirements and procedures used in this study are encouraged.  Please contact 
either of the investigators below if you have any questions.   
 
 
Mr Isaac Warbrick (06) 350 5799 ext 5264 I.Warbrick@massey.ac.nz 
Dr Stephen Stannard: (06) 350 5799 ext 7465 S.Stannard@massey.ac.nz 
Prof. Chris Cunningham: (04) 801 5799 ext 6027        C.W.Cunningham@massey.ac.nz 
 
 
 
 
 
This project has been reviewed and approved by the Massey University Human Ethics Committee, PN Application 04/110.  
If you have any concerns about the conduct of this research, please contact Professor John O’Neill, Chair, Massey University 
Campus Human Ethics Committee: Palmerston North, telephone 06 350 5249, email humanethicspn@massey.ac.nz. 
Appendix F 
 XXIX
 
Research Centre for Māori Health & Development 
College of Humanities & Social Sciences 
 
Participant Information Summary 
Insulin resistance in Maori & non-Maori men  
Who are the researchers and where can we be contacted? 
Mr Isaac Warbrick 
Research Centre for Māori 
Health & Development,  
Massey University,  
Wellington 
Prof. Chris Cunningham  
Research Centre for Māori Health & 
Development,  
Massey University,  
Wellington 
 Dr Stephen Stannard  
Institute of Food Nutrition & 
Human Health  
Massey University 
Palmerston North 
 
What is this study about?  This study is about type-2 diabetes which affects Māori at a greater rate 
than non-Māori.  Specifically we will investigate whether protection from type-2 diabetes seen in the 
Maori population may involve good aerobic fitness of your lean body mass which is mostly muscle, 
compared to non-Maori.  Muscle has an important role in the development of insulin resistance and 
ensuring aerobic fitness of muscle may be an important strategy to prevent type-2 diabetes.   
 
If you choose to participate in this study, you will: 
• Complete a health questionnaire 
• Complete an exercise test of your oxygen consumption on a treadmill. 
• Perform an oral glucose tolerance test. 
• Provide two 10 ml venous blood samples. 
• Have your body composition assessed via underwater weighing and Bioelectrical Impedance 
(Full details of these procedures are given in the attached pages). 
 
 
 
Appendix F 
 XXX
What are the benefits of the research? 
This research will give you the opportunity to have an input into a study which is designed to 
understand the risks Māori have in relation to type-2 diabetes and why they are more susceptible to the 
disease than non-Maori. 
 
Your Rights as Participants  
You are under no obligation to accept this invitation. If you decide to participate, you have the right to: 
• decline to answer any particular question; 
• withdraw from the study; 
• ask any questions about the study at any time during participation; 
• provide information on the understanding that your name will not be used unless you give 
permission to the researcher; 
• be given access to a summary of the project findings when it  is concluded. 
 
Project Contacts 
Please contact any of the researchers if you have any concerns or require further information. 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
This project has been reviewed and approved by the Massey University Human Ethics Committee, PN Application 07/60.  If 
you have any concerns about the conduct of this research, please contact Professor John O’Neill, Chair, Massey University 
Campus Human Ethics Committee: Palmerston North, telephone 06 350 5249, email humanethicspn@massey.ac.nz. 
 
Appendix F 
 XXXI
 
Research Centre for Māori Health & Development 
College of Humanities & Social Sciences 
 
Participant Full Information Sheet 
Insulin resistance in Māori & non-Māori men  
 
Aims of this study 
This research will use techniques to measure aerobic capacity and insulin sensitivity in a cohort of 
young Māori and non-Maori men.  Blood samples will be taken to estimate insulin sensitivity, blood 
lipid profile, glucose tolerance, C-reactive protein and other markers of insulin resistance.  A treadmill 
exercise test will be used to assess aerobic fitness.   
 
Participation 
Participants in this study will be part of two groups i) Maori and ii) non-Maori. Participants will be 
male who are physically active, aged between 18 and 45.  You will be fully informed of your 
involvement in this study and will complete a Medical History questionnaire and a Physical Activity 
Readiness questionnaire (PAR Q) to assess your readiness to participate in the study.  Following these, 
you will be required to sign a consent form after reading this “Participant Information Statement” and 
clarifying any matters that remain unclear.   
 
You will be asked to undergo one maximal exercise test on a treadmill, two measures of body 
composition, an oral glucose tolerance test, and to give two venous blood samples.  It should be 
emphasised that you may withdraw from this study any time you wish.  If you choose to do so, there 
will be no compromise in the relationship between you and the investigators.  If you are a Massey 
University student, participation in, or withdrawal from the study will not influence your academic 
progress in any manner.  In deciding to partake in this study, you will be subjecting yourself to the 
following requirements: 
 
1.  Screening Procedures 
You will answer a demographic, Medical History and Physical Activity Readiness questionnaire.   
 
2. Testing Procedures 
Appendix F 
 XXXII 
 
Venous blood sample and oral glucose tolerance test:  Prior to eating breakfast in the morning 
you will attend the Medlab Clinic to provide a 10 ml blood sample.  This will be taken from a 
vein in your arm by an experienced venepuncturist.  The sample will be frozen and later 
analysed for blood lipids, glucose, C-reactive protein and insulin.  The glucose tolerance test 
will then be performed with involves drinking a sugary (glucose) solution.  One further blood 
sample will be taken exactly two hours after drinking the solution to test your blood glucose and 
insulin levels again.  We will then provide you with breakfast.  Following analysis, any 
remaining blood will be returned to the earth by our kaumätua following due tikanga.   
 
Body Composition:  You will attend the RCMHD Research Laboratory at Massey University, 
Wellington Campus and have your body composition assessed in an underwater weighing 
tank.  This will require you to dress down to a pair of swimming shorts only and sit 
immerse yourself in a warm water tank.  You will be asked to breathe out then bend 
forward to fully submerge your body for approximately twenty seconds whilst we 
measure your body weight underwater.  If you feel uncomfortable, you can exit the tank 
anytime during the procedure at your own request.   
 
VO2max test: After the underwater weighing measurement, you will undergo an exercise test on a 
treadmill to measure your maximal oxygen uptake (VO2max).   
 
You will begin the test by walking on the treadmill at a slow speed.  Each two minutes the 
treadmill speed and gradient will be increased until you are unable to continue.   
 
During exercise, your oxygen consumption will be measured by collecting expired gases from 
your mouth using the Moxus gas analysis system.  This equipment measures volumes of air and 
oxygen passing into and out of your lungs and from this we can determine the rate at which you 
use oxygen.  It will require you to breathe through a mouthpiece during the test and have a clip 
on your nose.  During the test we will measure your heart rate using the commonly available 
Polar chest band and wristwatch combination.  All exercise testing will be performed on the 
research quality Woodway treadmill located in the RCMHD Research Laboratory at Massey 
University, Wellington Campus.   
 
Potential Risks and Discomforts 
As with any research of this nature, there are some potential risks and discomforts, which you should 
be aware of.  The researchers will attempt to minimise these through careful, consistent monitoring of 
Appendix F 
 XXXIII
your response to the testing procedures.  Every effort will be made by the researchers to ensure your 
safety, comfort and familiarity with the testing procedures.   
 
a. Blood test and oral glucose tolerance test:  Sampling of venous blood may cause a degree of 
transient discomfort, and some bruising may occur at the point of insertion up to 48 hours afterwards.  
This discomfort is of no lasting consequence.  There is also a slight risk of fainting during blood 
sampling.  The risks and discomforts will be minimised, as the procedure will be performed under 
sterile conditions by people highly experienced in this procedure.   
 
b. Exercise procedure:  The exercise procedure during the VO2max test will not be easy, but the time of 
exercise at high intensity is short (less than 3 minutes) and you are able to terminate the test at any 
time at which you feel fatigued.  This test is often performed on both athletes and sedentary 
individuals to measure aerobic fitness.  As in any physical activity, there is a very small possibility of 
injuries that include, but are not restricted to, muscle, ligament or tendon damage, breathing 
irregularities and dizziness.  However, the exercise protocols are commonly performed in exercise 
physiology laboratories and potential risks to participants have been minimised.   
 
c. Body composition.  There are no risks associated with these techniques.  During the underwater 
weighing you will be fully submerged in a warm water bath, and must hold your breath for about 20 
second whilst the measurements are taken.  This is a little disconcerting and if you feel uncomfortable 
you are free to exit the water at any time.  During this procedure you will need to wear a swimming 
costume.   
 
 
Right of withdrawal 
Your participation in this study is completely voluntary.  You may choose to withdraw from the study 
at any time for any reason.  By signing the consent form you are indicating that the tests and 
procedures of this study have been explained to you and understood by you.   
 
Inquiries 
All inquiries regarding requirements and procedures used in this study are encouraged.  Please contact 
either of the investigators below if you have any questions.   
 
Mr Isaac Warbrick (06) 350 5799 ext 5264 I.Warbrick@massey.ac.nz 
Dr Stephen Stannard: (06) 350 5799 ext 7465 S.Stannard@massey.ac.nz 
Prof. Chris Cunningham: (04) 801 5799 ext 6027       C.W.Cunningham@massey.ac.nz 
Appendix F 
 XXXIV 
 
10.12 Muscle contraction properties as predictors of insulin 
sensitivity in Māori and non-Māori men 
10.13 Pre-Exercise Health Screening Questionnaire 
 
Name:______________________________________________________________ 
 
Address: ____________________________________________________________ 
 
Phone: _______________________________ 
 
Age: ________________ 
 
Please read the following questions carefully.  If you have any difficulty, please advise the medical practitioner, 
nurse or exercise specialist who is conducting the exercise test.  
 
Please answer all of the following questions by ticking only one box for each question: 
 
This questionnaire has been designed to identify the small number of persons (15-69 years of age) for whom physical activity 
might be inappropriate.  The questions are based upon the Physical Activity Readiness Questionnaire (PAR-Q), originally 
devised by the British Columbia Dept of Health (Canada), as revised by 1Thomas et al. (1992) and 2Cardinal et al. (1996), 
and with added requirements of the Massey University Human Ethics Committee.  The information provided by you on this 
form will be treated with the strictest confidentiality.  
 
Qu 1. Has your doctor ever said that you have a heart condition and that you should only do physical activity 
recommended by a doctor? 
 Yes  No 
 
Qu 2. Has a member of your family died below the age of fifty (50) as a result of a heart condition? 
Yes  No 
 
Qu 3. In the past month have you had chest pain when you were not doing physical activity? 
 Yes  No 
 
Qu 4. Do you have any current injury, bone or joint problem in your legs, or previous injury that could be affected by 
resistance exercise? 
 Yes  No 
 
Appendix F 
 XXXV
 
Qu 5. Are you taking any prescribed medication? 
 Yes  No 
 
Qu 6. Do you know of any other reason why you should not do physical activity and/or resistance exercise? 
 Yes  No 
 
Qu7. Have you ever had persistent or regular back pain? 
 Yes  No 
 
Qu 8. Do you have a bleeding disorder? 
 Yes No 
 
Qu9. Do you currently have, or have you ever had renal and/or hepatic disease? 
 Yes No 
 
Qu 10. Are you diabetic? 
 Yes No 
 
Qu 11. Have you had/do you have a medical condition that may effect your ability to sense pain or discomfort? 
 Yes No 
 
Qu 12. Do you use a pacemaker? 
 Yes No 
 
Qu 12. Do you have a skin allergy or condition that may be put yourself and others at risk during exercise testing? 
 Yes No 
 
You should be aware that even amongst healthy persons who undertake regular physical activity there is a risk of sudden 
death during exercise.  Though extremely rare, such cases can occur in people with an undiagnosed heart condition.  If you 
have any reason to suspect that you may have a heart condition that will put you at risk during exercise, you should seek 
advice from a medical practitioner before undertaking an exercise test.  
 
 
I have read, understood and completed this questionnaire.  
 
Signature: ________________________________________  Date: ___________________ 
 
 
 
References 
3. Thomas S, Reading J and Shephard RJ. Revision of the Physical Activity Readiness Questionnaire (PAR-Q). Can J 
Sport Sci 17(4): 338-345.  
4. Cardinal BJ, Esters J and Cardinal MK. Evaluation of the revised physical activity readiness questionnaire in older 
adults. Med Sci Sports Exerc 28(4): 468-472 
Appendix F 
 XXXVI 
 
Participant Information Sheet 
Muscle contraction properties as predictors of insulin sensitivity in Māori and 
non-Māori men 
 
Who are the researchers and where can we be contacted? 
Dr Stephen Stannard  
Institute of Food Nutrition 
& Human Health  
Massey University 
Palmerston North 
Prof. Chris Cunningham  
Research Centre for Māori Health & 
Development,  
Massey University,  
Wellington 
Isaac Warbrick 
Research Centre for Māori Health & 
Development,  
Massey University,  
Wellington 
What is this study about? This study is part of a doctoral thesis for the primary researcher, Isaac 
Warbrick.  His doctoral research focuses on the development of insulin resistance and type-2 diabetes; 
a condition which affects Māori at a greater rate than non-Māori. Because insulin effectiveness in 
muscle is linked to skeletal muscle function we will investigate the relationship between blood markers 
of insulin resistance (a condition which leads to diabetes), and the peak power and contraction 
characteristics which are related to a persons muscle fibre type.  Measures of peak muscle power will 
be obtained through a number of leg extensions performed on the Biodex, while twitch characteristics 
will be obtained through electrically stimulated contractions on specially designed apparatus.  We also 
hope to compare our findings between Maori and non-Maori. Little is known regarding these 
relationships in Maori or whether they differ between ethnicities, but better understanding them may 
enable the development of more effective strategies for preventing type-2 diabetes in both groups.  
 
Ethnicity of participants in this study will be self-identified by each individual participant. 
 
If you choose to participate in this study, you will be invited to: 
• Complete a health questionnaire 
Appendix F 
 XXXVII 
• Complete 3-5 leg extensions against resistance on the Biodex.  
• Participate in an oral glucose tolerance test. 
• Provide two 10 ml venous blood samples. 
• Complete 3-5 electrically stimulated leg extensions. 
(Full details of these procedures are given in the attached pages). 
 
If any of the following apply:  
• You have a known heart or cardiovascular condition or if a member of your family died below 
the age of fifty (50) as a result of a heart condition. 
• You have any current or previous injury to your legs  
• You have had an injury or medical condition that you think may affect your ability to sense 
pain or discomfort. 
• You have ever had persistent or regular lower back pain. 
• You are taking prescribed medication. 
• You have cultural or religious sensitivities about human body measurements. 
• You have any other reason to consider that you are not in good health and of average, or 
better than average, fitness. 
• You have a bone or joint problem in your legs that could be made worse by resistance 
exercise 
• You are diabetic. 
• You have a bleeding disorder. 
• If you currently have or have had renal and/or hepatic disease 
• You use a pacemaker 
• You have a skin allergy or condition that may put yourself or others at risk 
…you should NOT participate in this project. 
 
What are the benefits of the research? 
This research will give you the opportunity to have an input into a study which is designed to 
understand the risks Māori and non-Maori have in relation to type-2 diabetes.  In completing the study 
we will also be able to give you an indication of your (quadriceps) muscle fibre type.   
 
Your Rights as a Participant 
You are under no obligation to accept this invitation.   If you decide to participate, you have the right 
to: 
• decline to answer any particular question; 
• withdraw from the study; 
• ask any questions about the study at any time during participation; 
• provide information on the understanding that your name will not be used unless you give 
permission to the researcher; 
• be given access to a summary of the project findings when it  is concluded. 
 
Project Contacts 
Please contact any of the researchers if you have any concerns or require further information 
Appendix F 
 XXXVIII
STUDY DETAILS 
 
Aims of this study 
This research will use laboratory techniques to measure the peak muscular force production and twitch 
characteristics during a leg extension, and how these measures relate to insulin resistance in both 
young Māori men and New Zealand non-Māori men.  Fasting blood samples will be taken to estimate 
insulin sensitivity.   
 
Participation 
Participants in this study will be male Māori and non-Maori who are physically active, aged between 
18 and 45.  You will be fully informed of your involvement in this study and will be invited to 
complete a Medical History questionnaire and a Physical Activity Readiness questionnaire (PAR Q) to 
assess your readiness to participate in the study.  Following these, you will be invited to sign a consent 
form after reading this “Participant Information Statement” and clarifying any matters that remain 
unclear.   
 
You will be asked to undergo one bout of leg extensions on the Biodex, one bout of electrically 
stimulated contractions on the electrical stimulation rig, an oral glucose tolerance test, and to give two 
venous blood samples.  It should be emphasised that you may withdraw from this study any time you 
wish.  If you choose to do so, there will be no compromise in the relationship between you and the 
investigators.  If you are a Massey University student, participation in, or withdrawal from the study 
will not influence your academic progress in any manner.  In deciding to partake in this study, you will 
be subjecting yourself to the following requirements: 
 
1.  Screening Procedures 
You will be invited to answer each of a Demographic, Medical History, and Physical Activity 
Readiness questionnaire.  
 
2.  Testing Procedures 
 
Muscular Fatigue:  Muscular Fatigue will be measured on the Biodex isometric dynamometer. 
You will be invited to sit in a chair while your lower leg is strapped to a mechanical arm that 
measures the velocity and force that your thigh muscle (quadriceps) is able to produce when you 
straighten your leg at maximal force against a known resistance for 60 repetitions.  This is very 
similar to using a leg extension machine found in most gyms.   
 
Muscle twitch characteristics: The twitch characteristics of the knee extensors (quadriceps) will 
be measured while you are seated in an adjustable straight-backed chair. During the test your 
ankle is strapped and the straps are connected to a load cell which records the forces produced 
by the quadriceps muscle during each electrical impulse. Electrical stimulation will be achieved 
by placing two large pads covered with electrode gel onto the skin on the belly of the 
Quadriceps muscles (front of thigh). A commercially available electrical stimulator will then 
deliver the impulses to the electrodes on your leg causing the quadriceps to contract. A suitable 
current will be obtained by increasing the current slowly until you find the sensation to be just 
tolerable. If at anytime, you feel uncomfortable you may stop the testing. 
 
Oral glucose tolerance test and venous blood sample:  One morning, before breakfrast, you 
will be asked to attend an accredited blood testing laboratory where a resting 10 ml blood 
sample will be taken from a vein in your arm by an experienced phlebotomist.  You will then be 
asked to undertake a glucose tolerance test which involves drinking a sugary (glucose) solution.  
One further blood sample will be taken exactly two hours after drinking the solution to test your 
blood glucose and insulin levels again.  The blood samples taken will be analysed for glucose 
and insulin concentrations to estimate your insulin sensitivity.  
 
Appendix F 
 XXXIX 
Note: The researchers are not medically trained to offer advice on treatment of illnesses that may be 
identified with the blood analysis.  However, if abnormalities are observed in the blood results, you 
will be notified by phone by Mr. Warbrick and advised to see a physician. The results of the tests will 
be made available to you.   
 
 
Potential Risks and Discomforts 
As with any research of this nature, there are some potential risks and discomforts, which you should 
be aware of.  The researchers will attempt to minimise these through careful, consistent monitoring of 
your response to the testing procedures.  Every effort will be made by the researchers to ensure your 
safety, comfort and familiarity with the testing procedures.   
 
 
a. Blood test and oral glucose tolerance test:  Sampling of venous blood may cause a degree of 
transient discomfort, and some bruising may occur at the point of insertion up to 48 hours afterwards.  
This discomfort is of no lasting consequence.  There is also a slight risk of fainting during blood 
sampling.  The risks and discomforts will be minimised, as the procedure will be performed under 
sterile conditions by highly experienced staff.   
 
 
b. Muscular Contractions on the Biodex:  You are likely to experience the fatigue of the thigh 
muscles associated with strenuous exercise during the contractions on the Biodex.  Nevertheless, as in 
any physical activity, there is a very small possibility of injuries that include, but are not restricted to, 
muscle, ligament or tendon damage and dizziness.  However, all protocols are commonly performed in 
exercise physiology laboratories and potential risks to participants have been minimized.  
 
c. Electrically stimulated muscle contractions:  Electrical Stimulation (ES) provides a small electrical 
stimulus to the belly of the muscle. At low frequencies (20Hz) participants may feel a ‘tingling’ in the 
muscle at higher frequencies (80Hz) participants may feel a ‘tugging’ in the muscle.  
 
Right of withdrawal 
Your participation in this study is completely voluntary.  You may choose to withdraw from the study 
at any time for any reason.  By signing the consent form you are indicating that the tests and 
procedures of this study have been explained to you and understood by you.  
 
Inquiries 
All inquiries regarding requirements and procedures used in this study are encouraged.  Please contact 
either of the investigators below if you have any questions.   
 
 
 
Mr Isaac Warbrick (06) 350 5799 ext 5264 I.Warbrick@massey.ac.nz 
Dr Stephen Stannard: (06) 350 5799 ext 7465 S.Stannard@massey.ac.nz 
Prof. Chris Cunningham: (04) 801 5799 ext 6027        C.W.Cunningham@massey.ac.nz 
 
 
 
This project has been reviewed and approved by the Massey University Human Ethics Committee: Southern A, Application 
08/56.  If you have any concerns about the conduct of this research, please contact Professor John O’Neill, Chair, Massey 
University Human Ethics Committee: Southern A, telephone 06 350 5799 x 8771, email humanethicsoutha@massey.ac.nz