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. Associations between Calcium Intake, Osteoporosis Knowledge and Osteoporosis Health Beliefs among young adult women in the Lower North Island, New Zealand. A thesis presented in partial fulfilment of the requirements for the degree of Master of Science in Human Nutrition At Massey University, Manawatu New Zealand Sarah Fekau 2021 i ABSTRACT Background/Aim: Osteoporosis is becoming the most prevalent bone disease in the world nowadays. While osteoporosis is often regarded as a disease of the elderly, however, maximising peak bone mass (PBM) during the adolescent and young adults’ stages is crucial to prevent or delay osteoporosis later in life. Osteoporosis is a preventable disease and making lifestyle changes or following health recommendations such as consuming adequate calcium is essential to prevent or delay osteoporosis. The first three decades of life is a crucial time to act, as it is the period to achieve optimal peak bone mass (PBM). Understanding an individual’s osteoporosis knowledge and health beliefs and factors influencing calcium intake may help to prevent osteoporosis. Therefore, this study aimed to examine the associations between calcium intake, osteoporosis knowledge and health beliefs among young female adults in the lower North Island in New Zealand. It also aimed to state the level of osteoporosis knowledge, osteoporosis health beliefs, calcium intake and validation of the food frequency questionnaire (FFQ). Study design: This was a secondary data analysis of 130 females (university students) between 18-25 years of age who voluntarily participated. Participant’s knowledge and health beliefs on osteoporosis were measured using the osteoporosis knowledge test (OKT) and osteoporosis health belief scale (OHBS). A FFQ was completed to estimate the calcium intake. Descriptive analysis, bivariate correlation and multiple regression were used to analyse the data of osteoporosis knowledge, health beliefs and associations with calcium intake. Validity was evaluated using the Spearman correlation coefficient (SCC), Wilcoxon signed rank test, Cross- classification, Weighted kappa statistics and Bland Altman analysis. Result: Findings show osteoporosis knowledge was significantly associated with calcium intake, susceptibility, calcium barriers, health motivation and one of the predictors of calcium intake as with perceived severity. In general, the university students had moderate mean knowledge on osteoporosis (16.9±3.7) and perceived moderate median susceptibility (15) and severity (19) to osteoporosis. The students perceived many benefits of taking calcium intake with lower calcium barriers and they were highly health motivated. Median daily calcium intake of 692mg (462, 10250) was below the estimated average requirement (EAR) of 1050mg (15-18 years old) and 840mg (19-30 years old) with acceptable findings on the validity analysis of the FFQ. Conclusion: Overall, the findings confirm the HBM theory that some perceptions such as severity and knowledge influence individual’s likelihood of engaging in health promoting behaviour (calcium intake) among this study population. Surprisingly most HBM constructs ii were not linked to behaviour (calcium intake). This is interesting but may not be causal. The findings show that increasing knowledge or improving awareness of osteoporosis especially related to physical activity, dairy and non-dairy calcium food sources and seriousness of getting osteoporosis may be the recommended preventive intervention for these university students. iii ACKNOWLEDGEMENT The completion of this thesis is made possible through the great support and contributions of these following people. First and foremost, I would like to give honour and glory to God Almighty for His grace and mercy towards my studies and completion of the thesis. Secondly, to my two Academic Supervisors Dr Janet Webber and Associate Professor Louise Brough for their in-depth knowledge, support, guidance, patience, encouragement and understanding throughout the journey. Your tireless support and contributions were the pillar of this thesis. Thirdly, a huge appreciation to Katie Schraders Elizabeth Reymonds for the primary data collection of the Love the Bone study (LTB). Also, acknowledging participants of LTB study. This study was made possible because of the availability of the data. Fourthly, I would like to extend my gratitude to the NZAid Scholarship team for the generous financial assistance (thesis allowance) given to support completion of the thesis. Finally, to my friends, family and dear husband for their continuous prayers, encouragement, moral support and understanding rendered towards the completion of this thesis. iv TABLE OF CONTENTS Abstract i Acknowledgement iii Table of Contents iv List of Tables vi List of Figures vii List of Abbreviations viii List of Appendices x Chapter 1: Introduction 1 1.1 : Background 1 1.2 : Purpose of the study 3 1.3 : Research question, aim and objectives 3 1.4 : Contributions of Researchers 4 Chapter 2: Literature Review 5 2.1 : Defining Osteoporosis 5 2.2 : Understanding bone health and osteoporosis 5 2.3 : Risk or prevention factors (non-modifiable & modifiable) 7 2.4 : Epidemiology and Burden of Osteoporosis 13 2.5 : Prevention of Osteoporosis with tools used to assess 15 Osteoporosis Knowledge, Health Beliefs and Calcium Intake 2.6 : Current Osteoporosis Knowledge, Health Beliefs and 20 Behaviour (calcium intake) 2.7 : Predictors of Calcium Intake 24 2.8 : Rationale of this study 25 Chapter 3: Methodology 35 3.1 : Study Design 35 3.2 : Sample Size 35 3.3 : Ethical Approval 35 v 3.4 : Subject Recruitment 35 3.5 : Data Collection 37 3.6 : Analysis of Calcium Intake from FFQ 39 3.7 : Data Management 42 Chapter 4: Results 44 4.1 : Demographic Characteristics 44 4.2 : Osteoporosis Knowledge Test (OKT) 45 4.3 : Osteoporosis Health Beliefs Scale (OHBS) 46 4.4 : Assessment of Calcium Intake 50 4.5 : Relationship of Calcium Intake with OKT and OHBS 56 4.6 : Multiple Regression (Investigating predictors of 58 calcium intake) Chapter 5: Discussion 59 5.1 : Osteoporosis Knowledge (OKT) 59 5.2 : Osteoporosis Health Beliefs (OHBS) 62 5.3 : Validation and Reproducibility of the FFQ 66 5.4 : Dietary Calcium Intake 67 5.5 : Summary of Calcium Intake, OKT and OHBS 68 5.6 : Relationships between Calcium intake, OKT and OHBS 68 5.7 : Predictors of Calcium Intake 69 5.8 : Strengths and Limitations 70 Chapter 6: Conclusion and Recommendations 72 6.1 : Conclusion 72 6.2 : Recommendations 73 References 74 Appendices 80 vi LIST OF TABLES Table 1.1: Contributions of Researchers 4 Table 2.1: Summary of studies using OKT, OHBS, Associations 26 between Calcium Intake, OKT and OHBS Table 3.1: Classification of food item into NZANS food group 40 Table 3.2: Statistical Test Cut-offs Criteria 42 Table 4.1: Demographics of the study participants 44 Table 4.2: Osteoporosis Knowledge Test (OKT) Scores mean (SD), 45 percentages & interpretation Tables 4.3: Osteoporosis Health Belief Scale mean (SD), median 47 (25th, 75th), percentage & interpretation Table 4.4: Calcium Intake from FFQ1 and FFQ2 & 3DDD 50 Table 4.5: Cross classification of FFQ1 and 3DDD 51 Table 4.6: 10 Dietary records with misclassification 52 Table 4.7: Milk and milk products serving per day and calcium daily 55 intake Table 4.8: Spearman rho’s matrix correlation for calcium intake and 56 independent variables Table 4.9: Pearson and Spearman’s Correlation between independent 57 variables Table 4.10: Simultaneous Multiple Regression analysis for variables 58 predicting Calcium intake vii LIST OF FIGURES Figure 2.1: Bone mass across lifespan 7 Figure 3.1: Flow diagram of recruitment and data collection process 36 Figure 4.1: Bland Altman 51 Figure 4.2: Percentage contribution of food item to dietary calcium 53 Figure 4.3: Contribution of Calcium intake by food group 53 Figure 4.4: Percentage of participants consuming food group at least once 54 or more per day & week viii LIST OF ABBREVIATIONS a BMD ANZHFR Areal Bone Mineral Density Australia & New Zealand Hip Fracture Registry BHQ Bone Health Questionnaire BMD Bone Mass Density BMC Bone Mineral Content DALYs Daily Adjustments in Life Years DEXA Dual-emission X-ray Absorptiometry DXA Dual X-ray Absorptiometry EAR Estimated Average Requirement FFQ Food Frequency Questionnaire FR Food Records HBM Health Belief Model IOF International Osteoporosis Federation LTB Love the Bone Study NHS National Health Survey NZANS New Zealand Adult Nutrition Survey NZ New Zealand OKAT Osteoporosis Knowledge Assessment Tool OKQ Osteoporosis Knowledge Questionnaire OKS Osteoporosis Knowledge Scale OKT Osteoporosis Knowledge Test OHBS Osteoporosis Health Belief Scale ix OPBS Osteoporosis Preventing Behaviours Survey OSES Osteoporosis Self -Efficacy Scale PTH Parathyroid Hormone PBM Peak Bone Mass QALYs Quality Adjusted Life in Years QUS Quantitative Ultrasound RAM Rapid Assessment Method RCT Randomized Controlled Trial RDA Recommended Dietary Allowance RDI Recommended Dietary Intake r-OKT Revised Osteoporosis Knowledge Test SD Standard Deviation USA United States of America WHI Women Health Initiative WHO World Health Organization 3DDD 3 Day Diet Diary x LIST OF APPENDICES Appendix 1: Ethic Approval 80 Appendix 2: Screening Form 81 Appendix 3: Information Sheet 83 Appendix 4: Consent Form 89 Appendix 5: OKT Questionnaire 90 Appendix 6: OHBS Questionnaire 96 Appendix 7: Bone Health Questionnaire Extracts 103 (Medical History and FFQ Section) Appendix 8: 3 Day Diet Diary (3DDD) 105 Appendix 9: Calcium Content per serving from FoodWorks 107 Appendix 10: OKT Complete Result 110 Appendix 11: OHBS Complete Result 112 1 CHAPTER 1: INTRODUCTION 1.1: Background Osteoporosis is the most prevalent bone disease in the world today. It affects predominantly post- menopausal women. It is also the most preventable bone disease (Fourie, Floyd, & Marshall, 2015). Globally approximately 200 million people are affected with osteoporosis (Al Anouti et al., 2019). Osteoporosis usually progresses asymptomatically and is often diagnosed after a fracture occurs. Among the Western population, particularly in North America, Europe, and Oceania, it has been reported that there has been a substantial increase in the incidence of osteoporosis fractures (Curtis, Moon, Harvey, & Cooper, 2017). The main fracture sites are the hip, spine, wrist, and humerus associated with high morbidity and mortality rates and massive socioeconomic burdens. Hip and spine fractures are the most serious, as higher immobility, morbidity, and mortality rates are related to these fractures (WHO, 2003). About 38% of women aged over 50 years are affected by hip and spine fracture globally (Wade, Strader, Fitzpatrick, Anthony, & O’Malley, 2014). Osteoporosis increases with increased age, so osteoporosis fracture consequences are expected to increase with our aging population (Brown, McNeill, Leung, Radwan, & Willingale, 2011). Osteoporosis is often referred to as a disease of the elderly however, maximising peak bone mass (PBM) during the stages of adolescence and young adulthood is crucial. Weaver et al. (2016) found that PBM timing varies according to the skeletal site, so skeletal health maintenance is paramount throughout puberty and the stage of young adulthood. Most healthy females attain 92% of their bone mineral content (BMC) at age 18 years and 99% by 26 years of age (Perez-Lopez, Chedraui, & Cuadros-Lopez, 2010). Adequate calcium intake is essential for adolescents and young adults to achieve PBM and maintain bone mineral density (BMD) later in life. Preventing osteoporosis and reducing the risk of osteoporosis can be achieved through nutritional measures, such as increasing calcium and vitamin D intake. It can also be accomplished by engaging in weight-bearing exercise (WBE) and addressing lifestyle behaviours, such as smoking and alcohol drinking. An individual's knowledge and perceptions of their health issues have a significant impact on their lifestyle and their health. Therefore, adequate measures at an early age, such as raising awareness and modifying lifestyle, are crucial to prevent osteoporosis. University/College-age is the time when peak bone mass (PBM) is achieved and lifestyle behaviours are being developed and it is the perfect opportunity for adopting lifestyle changes (Khired et al., 2021). The Health Belief Model (HBM) has been used to try to help people change their lifestyle. It 2 consists of these four constructs: susceptibility, severity, benefits and barriers, which suggest a person’s perceptions towards a health problem (Rosenstock, Strecher, & Becker, 1988). For instance, susceptibility measures an individual’s perception on their risk of developing a health concern or its existence while severity measures the consequences of having that health concern. The HBM was later adapted as the Osteoporosis Health Belief Scale (OHBS) to measure an individual’s or people’s perceptions on these constructs: susceptibility, severity, benefits and barriers of health behaviours (calcium intake and physical activity) to related to osteoporosis (Edmonds, Turner, & Usdan, 2012). In 2007, New Zealand, recorded 70,631 people were diagnosed with osteoporosis which incurred a cost of $330 million (Brown, McNeill, Radwan, & Willingale, 2007). Ninety percent of them were females, and 28% of the cases were diagnosed following a fracture. The National Health Survey 2006/07 reported that osteoporosis prevalence increases with age and was higher among European females (Ministry of Health, 2008). Brown et al. (2007), further emphasised that osteoporosis is more common than breast and prostate cancers and is usually accompanied by fractures and premature deaths. It was highlighted as well that fractures caused by osteoporosis would increase from 84,000 in 2007 to nearly 100,000 in 2013 and 116,000 in 2020. 4906 hip fractures alone were reported in the Australian and New Zealand Hip fracture registry (ANZHFR) 2018 annual report (ANZHFR, 2018). Sadly, an increased rate of osteoporosis fractures is expected in the future. Hence, an increase in health and socioeconomic burdens is expected that had an estimated increased health cost of NZ 458 million in 2020 (Brown et al., 2011). Statistics show low calcium intake existed among females of all ages in the 1997 National Nutrition Survey (NNS) and 2008/09 New Zealand Adult Nutrition Survey (NZANS). In the NZANS 2008/09, the intake in the age groups 15-18 years (682mg) and 19-30 years (704mg) was below the estimated average requirement (EAR) for the respective age groups (1050mg and 840mg). Having calcium intake below EAR is of great concern as a risk factor for osteoporosis. F u r t h e r m o r e , a previous study, von Hurst and Wham (2007) found moderate osteoporosis knowledge (16.4 out of 26) among females aged 20-49 years living in Auckland. Having insufficient knowledge of osteoporosis risk and prevention behaviours, such as nutrition (calcium and vitamin D intake) or having these behaviours without knowing that they help bone health may limit preventive measures. Additionally, responses to the osteoporosis heath belief questionnaire von Hurst and Wham (2007) showed females perceived they had a low susceptibility to the severity of osteoporosis, thus osteoporosis is not regarded as a threat. A low perception of the barriers to calcium intake was noted, but the belief that calcium-rich 3 foods are high in cholesterol (fat) was regarded as a barrier to calcium intake by 75% of the study population. Similar barriers were found in previous studies in the early 2000s in New Zealand (Gulliver & Horwath, 2001; Wham & Worsley, 2003). In contrast, less than a fifth of 102 South Asian females living in Auckland agreed with the idea that calcium-rich foods are high in fats (Tsai, 2008). Tsai (2008) also found there was a low perception regarding the susceptibility and severity of osteoporosis and the barriers to calcium intake. A lack of osteoporosis knowledge and inadequate calcium intake among the study population was highlighted as well. Moreover, a review of the literature revealed that young adults from six countries (Canada, Iran, Nigeria, Pakistan, Sri Lanka, and Syria) had low osteoporosis knowledge and viewed osteoporosis as a disease of old age (Chan, Mohamed, Ima-Nirwana, & Chin, 2018). Also, low perceived susceptibility, low perceived severity, and not engaging in osteoporosis behaviours were highlighted in this population. Chan et al. (2018) also found that lack of knowledge and misconception greatly influenced behaviours to prevent osteoporosis. On the other hand, other researchers found knowledge or increased osteoporosis knowledge does not influence or alter behaviours (Althobiti, Naqshbandi &Mohamed, 2020). Therefore, further investigation on osteoporosis in terms of the knowledge, health beliefs, and calcium intake among young females is needed. 1.2: Purpose of the study The two studies (Tsai, 2008; von Hurst and Wham, 2008) previously conducted among women in New Zealand had a low proportion of young adults; therefore, this study will focus on only young female adults. To my knowledge, this will be the first study to assess knowledge, health beliefs regarding osteoporosis, and dietary calcium intake among young female adults or university-age females. The information will be valuable for responsible ministry, agency, and stakeholders to develop appropriate and effective interventions for the young adult population. Additionally, the findings will confirm or refute the previous findings and whether the implemented recommendations, such as education or the awareness of low-fat products was effective. The study’s research questions, aim and objectives are listed below. 1.3 : Research question, Aim and Objectives 1.3.1: Research Questions: Are osteoporosis knowledge and osteoporosis health beliefs predictors of calcium intake? 4 1.3.2: Research Aim: To examine associations between calcium intake, osteoporosis knowledge, and osteoporosis health beliefs (OHB) among young adult females in the lower North Island Region of New Zealand. 1.3.3: Research Objectives: 1. Describe the level of osteoporosis knowledge. 2. Describe the level of osteoporosis health beliefs. 3. Describe calcium intake including validation of a short food frequency questionnaire (FFQ). 4. Examine the associations between calcium intake, osteoporosis knowledge and health beliefs. 5. Examine the association between family history, osteoporosis knowledge and health beliefs. 6. Investigate the predictors of calcium intake. 1.4: Contributions of Researchers Table: 1.1: Contributions of Researchers Researcher Contribution Sarah Fekau MSc (Human Nutrition) Student Primary author of this thesis who is responsible for developing research questions, aims and objectives; sampling size; data processing, statistical analysing and thesis writing Dr Janet Weber Academic Supervisor (main supervisor) Responsible for supervision of the entire research process through final submission. Provide technical and academic assistance in all aspects. Dr Louise Brough Academic Supervisor (co-supervisor) Supervised and assisted with all statistical analyses and validation of the food frequency questionnaire (FFQ). Katie Schraders Researcher collecting primary data used for this secondary analysis Elizabeth Reymonds Researcher collecting primary data used for this secondary analysis 5 CHAPTER 2: LITERATURE REVIEW 2.1 : Defining Osteoporosis Osteoporosis is a term derived from two Greek words; osteon' (bone) and 'poros' (little hole), which means "porous bone" (Rachner, Khosla, & Hofbauer, 2011). Osteoporosis is also known as the 'silent disease' or 'silent thief' because it is usually asymptomatic until a fracture occurs (Gass & Dawson-Hughes, 2006; Soleymanian, Niknami, Hajizadeh, Shojaeizadeh, & Montazeri, 2014). The World Health Organization (WHO) defines osteoporosis by bone mineral density (BMD) based on a T-Score from the dual x-ray absorptiometry (DXA) scan. The DXA scan can predict bone strength, fracture risks and is regarded as the gold standard to measure bone mass measurements (Dell, Greene, Anderson, & Williams, 2009). Osteoporosis has two distinct types: primary or secondary. Primary osteoporosis includes involutional osteoporosis type 1 or postmenopausal osteoporosis and involutional osteoporosis type 2 or senile (aging) osteoporosis (Sozen et al., 2016). Primary osteoporosis is related to hormonal, and aging factors. In contrast, secondary osteoporosis is associated with pre-existing diseases such as diabetes, renal disease, congenital conditions (leukemia), and a drug-induced condition, e.g. hyperthyroidism. Secondary osteoporosis develops following those mentioned diseases, medications, and lifestyle changes (Sozen et al., 2016). Postmenopausal and aging osteoporosis are common due to the natural decline of estrogen levels and increased bone loss with advanced aging (Poole & Compston, 2006). Osteoporosis is a significant public health issue in developed countries, and the prevalence is expected to increase with the increasing elderly population (Svedbom, Ivergård, Hernlund, Rizzoli, & Kanis, 2014). The health and economic burden of osteoporosis is enormous and causes substantial disability adjustments in life years (DALYs). Studies detail that osteoporosis is preventable if peak bone mass (PBM) is maximised during the skeletal growth period at the adolescent and young adult stage and if the rate of bone loss is slow (NIH, 2001). 2.2 : Understanding Bone Health and Osteoporosis 2.2.1 : Bone Health Bone is a connective tissue that supports the human body with mechanical support and facilitates muscle action, locomotion, and protects internal organs (Prentice et al., 2003). Bones are made up of collagen and hydroxyapatite (calcium and phosphorus); the minerals are the bone mineral content (BMC). 6 The bone’s size, thickness and volume are the areal bone mineral density (aBMD). Production and maintenance of bones are functioned by three cells, osteoblasts, osteoclasts, and osteocytes. The osteoclast's role is to remove old bone (resorption) while osteoblasts replace the old bone with new ones (formation or deposition). This process of resorption and deposition is called bone remodeling and is controlled by the osteocytes. The process maintains bone size and strength and regulates bone density and calcium levels in the body (de Villiers, 2009; Nichols, Bonnick, & Sanborn, 2000). Bone remodeling is a continuing process in life, with a gain in bone density in early life followed by a gradual bone loss once PBM is reached. The rate of bone loss increases as people get older and following menopause in women (Nichols et al., 2000). Over time bones gradually become brittle due to low BMD, which increases bone fragility, risk of fracture and later osteoporosis. 2.2.2 : Peak Bone Mass (PBM) and Bone Loss Bones grow in length and width in the first two decades of life with more bone formation than bone resorption (Ilich & Kerstetter, 2000). During that period, a steady accumulation of bone mass is formed leading to PBM. The rate of bone mass increases at a different rate as per life stages. Teegarden stated that PBM occurs at a different age for different bone sites (Teegarden et al., 1995). Although the timing of achieving PBM is unclear, girls gained 85% of PBM by 18 years of age and boys gain 90% by the age of 20 (Weaver et al., 2016). For instance, Weaver et al. (2016), noted that hip PBM is achieved between age 16 and 19 and between 33 and 40 for the lumbar spine PBM among females. Once PBM is reached, bone loss begins and continues until the end of life (Ilich & Kerstetter, 2013). Gradual bone loss persists throughout adulthood, increases between 5 and 10 years post-menopause in women, and continues with advanced age in both genders. The evidence showed that a 10% increase in PBM reduces the risk of osteoporotic fractures by 50% and hip fractures by 30% (Bonjour, Chevalley, Ferrari, & Rizzoli, 2009). Therefore, it is crucial to achieve maximum bone mass during PBM attainment (Bono & Einhorn, 2003). 7 Fig 2.1: Bone mass across the lifespan for male & female (Source: Weaver et al., 2016) 2.3 : Risk or Prevention Factors Even though osteoporosis is often regarded as a disease that affects the elderly, it develops during a person’s lifespan (Cech, 2012). The Development of osteoporosis is associated with non-modifiable and modifiable risk factors that influence bone mass gain and bone loss. While factors such as genetics, gender, hormone, and aging are non-modifiable, a person’s lifestyle choices that consist of calcium intake, vitamin D, dietary pattern, physical activity, smoking, and alcohol are the modifiable factors. Knowing factors that have an impact on PBM and the rate of bone loss is essential to establish and sustain skeletal health. Non- Modifiable Factors 2.3.1 : Genetics/Ethnicity/ Family history Genetics significantly influences bone architecture, bone strength, and PBM; an estimated 60% to 80% of PBM is determined by genetics (WHO, 2003). Twin studies confirm that 80% of the PBM is controlled by genetic predisposition, while the remaining 20% by environmental factors (Sambrook et al.1996). Also, certain ethnicities such as Caucasians, Asians, Europeans, and Hispanics have been reported to have lower BMC, BMD, and bone size compared to African and Pacific people (Bachrach et al. 1999; Bhudhikanok et al.1996; Brown et al. 2013). Evidence also shows that daughters of postmenopausal women with osteoporosis have lower BMC at the lumbar spine and femoral neck than daughters of non-osteoporotic mothers (Percival, 1999; Seeman et al., 1989; WHO, 2003). However, the link can be either heredity or 8 influenced by family lifestyle as both contribute to bone health. 2.3.2 : Advanced Age Bone loss is a natural process that happens after PBM is reached and accompanies aging. Bone loses its strength and density as individual ages through the natural changes in bone regulating hormone and calcium metabolism. (Snelling, Crespo, Schaeffer, Smith, & Walbourn, 2001). At age 50 years, loss of bone mass occurs at 0.7-2% yearly for postmenopausal women and 0.5-0.7% for men (Bono & Einhorn, 2003; O'Keefe et al., 2016). International Osteoporosis Foundation (2007), reported that osteoporosis incidence increases by age, with 10% among women at 60 years, 20% at 70 years, 40% at 80 years, and 66% at 90 years. About 90% of hip fractures occur among women at over 50 years of age. Women have a longer life expectancy, so they are more likely to have osteoporosis due to continued bone loss over the years and during their advanced years. 2.3.3 : Gender-Female Women are prone to have osteoporosis earlier than men, and their chances are three times more over the lifetime (WHO, 2003). Females have less BMC, bone mass and aBMD than males during childhood and adolescence, and usually have rapid bone loss following menopause (Naganathan & Sambrook, 2003; Weaver et al., 2016). Although bone loss occurs in both men and women, women’s bone loss occurs at an earlier age and at a faster rate (Alswat, 2017). Alswat (2017), further noted that women also have high bone resorption marker and tend to have fractures five to ten years earlier than men. However, between 60 to 65 years of age, the rate of loss slows down to a similar rate with men. On average, women lose 30% of their bone mass while men between 45 and 75 years of age lose 15 percent. 2.3.4 : Reproductive and Hormonal Status Women tend to lose bone mass rapidly 5-10 years postmenopausal due to estrogen decline. As estrogen production declines, the estrogen level in the blood level also decreases and changes the bone remodeling cycle. As a result, less bone formation and greater bone resorption lead to bone loss. The timing of menarche is another determinant of BMD and BMC among females. Chang et al. (2017) found the age of menarche had a weak negative association with BMD on the lumbar spine among Korean females aged 20-50 years. Significant associations were found with those who had menarche at an older age (16-18 years) with BMD. BMD of those who had menarche at an older age (16-18 years) was lower compared to those who had menarche < 12 years, even after controlling for confounding factors such as age, family history, body mass 9 index (BMI), parity, and lifestyle behaviours (smoking, alcohol, and exercise). However, no associations of BMD with the age of menarche were found in the femur and femur neck regions among this study population of 5032 women. The authors also found taking oral contraceptives had a negative association with BMD. However, the findings contradict the other findings in which higher doses of contraceptives had positive associations with BMD (Hagemans et al., 2004) and no significant difference was found with the age of menarche (Sioka et al., 2010). Also, other factors like the number of pregnancies, breastfeeding, duration of fertility and early menopause influence BMD (Grizzo et al., 2020; Sioka et al., 2010). Modifiable Factors Dietary and lifestyle behaviours strongly influence PBM and the rate of bone loss (Compston, 2004; Percival, 1999; WHO, 2003). A recent review concluded that about 20-40% of adult PBM is influenced by nutritional and lifestyle factors (Weaver et al., 2016). The authors developed a four-grade (A-D) system to assess evidence-based studies. Factors with robust evidence are graded as A, moderate evidence as grade B, limited evidence as grade C, and inadequate evidence as grade D. Based on the grading system, lifestyle factors are graded as follow: grade A calcium intake and physical activity; grade B- vitamin D and dairy foods (food pattern); grade C- fiber, fruits, and vegetable intake, smoking (detrimental), coco-cola and caffeine beverages (detrimental) and grade D- alcohol and oral contraceptives (detrimental). The grading system is based on evidence on the influences of each factor on PBM or overall bone health. The mentioned factors will be discussed according to nutritional and lifestyle behavioural factors. Nutritional Factors 2.3.5 : Calcium Intake A recent systematic review reported that adequate calcium intake is essential during the human lifespan to ensure maximum PBM, maintain BMD, and reduce bone loss later in life (Weaver et al., 2016). About 99% of the calcium in the human body is found in bone in the form of hydroxyapatite (Ilich & Kerstetter, 2013). Calcium intake is crucial during skeletal growth to attain a higher PBM. In general, calcium intake, calcium absorption, and calcium excretion determine the calcium balance, impacting bone mass imbalance and bone loss. Bonjo et al., (1997) conducted a randomized control trial (RTC) on pre-pubertal girls for one year. The girls were given calcium enriched food and drinks equivalent to 850 mg of calcium intake per day for a year; BMD gains on all sites were noted in both placebo (similar food and drinks but not 10 fortified with calcium) and calcium groups but were higher among the calcium supplemented group. Significant positive effects of calcium on bone accrual and reduced bone loss were noted in a systematic review among premenopausal women. Of the six prospective studies, five showed additional calcium intake of 600mg-1500mg either from dietary sources or supplements provided bone mass maintenance and prevented bone loss (Anderson & Rondano, 1996). The review has confirmed that adequate calcium intake is associated with the accrual of bone mass and reduced bone loss. 2.3.6 : Vitamin D Vitamin D is a nutrient that functions as a steroid hormone essential for developing and maintaining bone health, serum calcium, neuromuscular function, and metabolic processes (Holick, 2007). The primary source of vitamin D is synthesis during exposure to sunlight, but it can also be found in a few natural and fortified foods. A deficiency of vitamin D will decrease intestinal calcium absorption, which increases the secretion of parathyroid hormone (PTH). Increased PTH triggers bone resorption by releasing calcium from bone, leading to a decline of BMD. Vitamin D given with a calcium supplement, was found to positively affect bone density in the Women's Health Initiative (WHI) study among women between 50 and 79 years of age. A dose of 400 IU vitamin D and 1000 mg calcium per day increased the hip bone density by 1.06% after seven years and delayed bone mass loss further compared to those on placebo (Brunner et al., 2008; Holick, 2007). A further sub-analysis of the WHI participants also reported that a 29% reduction of hip fracture was found among women who consumed at least 80% of calcium and vitamin D doses during the study period compared to non-compliant women. Also, reduction of fracture risk by 26% and 32% was found among participants with high doses of vitamin D at 700 and 800IU/day compared to those taking 400IU/day in a meta -analysis (Bischoff-Ferrari et al., 2004). Vitamin D deficiency also leads to muscle pain, muscle weakness, and wasting, as one of vitamin D's roles is to maintain neuromuscular function (Binkley, 2012). Muscle weakness and muscle loss have been associated with increased risk of falls. Falls are more prevalent among older people 65 years of age and above. Binkley (2012) reported falls are the leading cause of fractures, especially hip fractures which were 90% fall related. A meta-analysis of 5 RCTs found that vitamin D supplementation does reduce the risk of falls by 22% among 1237 elderly between 65 and 92 years of age (Bischoff-Ferrari et al., 2004). 11 2.3.7 : Dietary Pattern Although individual nutrients play an essential role in bone health and osteoporosis prevention, the overall diet or dietary pattern is also worth considering. Nutrients are obtained from foods and food groups in combination, not foods eaten alone, so they are better regarded as overall diet or dietary patterns. Studies on dietary patterns found associations with high and low BMD. Dietary patterns rich in fruits, vegetables, nutrient-dense, and Mediterranean characteristics are associated with higher BMD and reduced risk of fracture (Kontogianni et al., 2009; Tucker et al., 2002). The studies also note that lower BMD was associated with a high confectionary and energy-dense pattern dietary pattern. Similar findings of negative associations were found with dietary patterns high in processed foods among Australian women (McNaughton, Wattanapenpaiboon, Wark, & Nowson, 2011). Women with the dietary pattern of Mediterranean characteristics showed positive associations with a higher BMD at the hip (0.2%), spine (0.3%), and total body BMC (0.6%) compared to women with a high consumption of processed foods. The associations remained even after adjustment of confounding factors such as age, smoking status, and physical activity. Lifestyle Behavioural Factors 2.3.8 : Physical Activity Physical activity influences bone accrual during childhood and adolescence and helps reduce bone loss and the risk of falls and fracture (Kohrt, Bloomfield, Little, Nelson, & Yingling, 2004). Confirmation of positive association of physical activity and BMD was highlighted in a recent review, whereby 90% of the 20 longitudinal studies reviewed found a statistically significant higher BMD among the physically active than the physically inactive cohort (Weaver et al., 2016). Regular physical activity of more than 60 minutes with moderate – vigorous-intensity physical activity 3-5 times per week helps promote healthy and stronger bones (Carter & Hinton, 2014). Baxter-Jones, Kontulainen, Faulkner, and Bailey (2008) reported an increase of 1-5% of BMC and BMD among athletes engaged in weight bearing exercise (WBE) during training from seven months to two years. WBE are recommended and reported to be more beneficial to BMD than non- WBE. A comparison study of WBE and non-WBE in a cross-sectional study among young adults found higher BMD of athletes who participated in high-intensity forces, such as gymnastics, weightlifting, and bodybuilding compared to swimming athletes (Kohrt et al., 2004). 12 2.3.9 : Smoking Although the association varied between studies, with some showing decreases and others not, a meta-analysis of the data (94 prospective and cross-sectional studies) found current smokers’ bone mass was 10% lower than non- smokers in all bone sites (lumbar, spine, hip and forearm). (Ward & Klesges, 2001). Additionally, smoking is associated with fracture risk in all sites, but a greater risk in the hip site with men at higher risk. In all, smoking was found to be inversely associated with low bone mass and an increase in fracture risks. Similarly, a recent review by Weaver et al. (2016), confirmed smoking is associated with low BMD at specific skeletal sites, and smokers are prone to stress fractures among young adults. These findings were based on 13 studies comprised of prospective (6) and cross-sectional (7) studies published since 2000. Other risk behaviours, such as inadequate dietary calcium intake, physical activity, and heavy alcohol consumption, were noted common among smokers. These behavioural factors are confounding factors which were not adjusted in some of the studies in the review. 2.3.10 : Alcohol consumption Another proposed risk factor for osteoporosis is alcohol consumption and it has been shown to have both positive and negative effects on BMD. In a meta-analysis and systematic review, the findings showed the consumption of a 0.5-1 standard drink of alcohol had lower fracture risk while a higher fracture risk for the consumption of 2 or more standard drinks as compared to abstainers (Berg et al., 2008). Also, positive association between alcohol consumption and BMD was noted however, evidence of level of alcohol consumption (moderate & heavy) was not sufficient to compare. Apart from fracture risks, associations between alcohol intake and osteoporosis risk were found in a recent systematic review. Cheraghi et al. (2019) found that the relative risk of developing osteoporosis varies among alcohol consumers as compared to abstainers. Those consuming 0.5-1 standard drink per day have 1.38 times risk of developing osteoporosis, 1.34 risk times for 1-2 standard drinks and 1.63 times for 2 or more standard drinks per day. The risk is higher among consumers who consume 2 or more standard drinks per day. There is a positive relationship between alcohol intake and osteoporosis regardless of the amount consumed however, risk is much higher with higher intake (≥ 2 standard drinks). 13 2.3.11 : Summary of Risk Factors Even though osteoporosis is a severe disease it is preventable as well. Certain risk factors such as genetics/family history/race, advanced age, gender, hormonal changes are non-modifiable which does influence bone especially in the attainment of PBM and bone loss. Other risk factors are modifiable which can be controlled such as adequate intake of calcium vitamin D, a nutrient-dense dietary pattern, engaging in WBE, taking alcohol in moderation, and if possible, it helps to quit smoking. 2.4 : Epidemiology and Burden of Osteoporosis 2.4.1 : Prevalence - globally Osteoporosis is a growing public health issue worldwide but is more prevalent in the following regions: North America, Europe, South-east Asia, and Oceania (Australia & New Zealand). In 2010, 10.3 million adults in the United States of America (USA) had osteoporosis, while 43.3 million had low bone mass. About 54% of the adult population ≥50 years had osteoporosis or osteopenia (Wright et al. 2014). The majority with osteoporosis were non-Hispanic White, followed by Mexican American and non-Hispanic Black. High osteoporosis prevalence has been reported from 27 countries in Europe. Almost 28 million osteoporosis cases were documented in 2010 among men and women between the ages of 50 and 84 in European countries where 80% of the cases were women (Hernlund et al., 2013). In the Asian region, approximately 70 million adults (≥50years) had osteoporosis, and 50% were women in China. Despite recording high numbers, the actual cases were underreported due to under-diagnosis (IOF, 2003). For the Oceania region, in year 2001, 2 million people were reported with osteoporosis in Australia and 4.7 million in 2012 (IOF, 2007; Tatangelo et al., 2018). An estimated 652,500 people over age 50 years reported being diagnosed with osteoporosis between 2011 and 2012 in Australia, and 88% were women. Tatangelo et al. (2018) also highlight an issue with under-diagnosis, which does not reflect the real picture of the actual osteoporosis prevalence rate and its burden in Australia. In terms of predicting the incidence of osteoporosis, an increase of 31% is predicted for 2022 in Australia. According to the New Zealand National Health Survey (2002/03), the osteoporosis prevalence rate was 2.4%, meaning 1 in 42 adults had osteoporosis. Osteoporosis was more prevalent among females and the non-Maori population. In 2007, 70,631cases were reported based on the NZ survey 2002/03 and NZ census among adults aged 50+ years, and 90% of the patients 14 were women (Brown et al., 2007). Reported cases were diagnosed either by fractures or from a DXA scan. It was noted that cases were underdiagnosed and underestimated due to the diagnostic process and costing. Public health funding does not cover asymptomatic diagnosis and preventative treatment and the cost of the diagnosis is not reimbursed (Brown et al. 2011). Hence, the real prevalence is expected to be much higher as elderly population increases. 2.4.2 : Incidences of Fractures and impacts on morbidity, mortality, and quality of life Fracture is one of the utmost consequences of osteoporosis as it increases morbidity and mortality risk (WHO, 2003). The most common osteoporotic fracture sites are the hip, vertebrae, and forearm (Poole & Compston, 2006). These osteoporotic fractures impose massive burdens on a country's health and economy. Hip fracture is the most serious one as it is so painful and requires extended hospitalization and care (WHO, 2003). Furthermore, it comes with a higher degree of immobility, morbidity, and mortality, creating huge health burdens. Statistics show a trend of 20% of patients with hip fractures died within a year following fracture (Harvey, Dennison & Cooper, 2010). In general, fractures affect an individual physically and psychologically in terms of the subsequent loss of mobility, ability to make daily activities, loss of productivity, loss of independence, and reduced quality of life (Harvey, Dennison, & Cooper, 2010). IOF (2007) reported approximately 9 million fractures in the year 2000 globally, with hip fracture the highest at 1.6million, followed by forearms at 1.7million and 1.4 million vertebral fractures. These osteoporotic fractures occurred more frequently among women (1 in 3) than men (1in 5) of adults over 50 years. Globally, an osteoporotic fracture occurs every 3 seconds, and incidence rates are projected to increase 2-3-fold by 2050. Burge et al. (2017) reported more than 2 million osteoporotic fractures in the USA in 2005 with a prediction of a 50% increase in fractures that is expected by the year 2025. Similarly, a prediction of an increase of 4.8 million fractures by 2025 was made for Europe (IOF, 2003). The increasing trend of fractures was also reported in China and Australia, with a high incidence of hip fracture. China reported 68,700 annually while 20 000 were reported in Australia (IOF, 2003). An increase incidence of hip fractures (411,000) was later reported in 2015 for China (Si, Winzenberg, Jiang, Chen, & Palmer, 2015). No new record for Australia was reported in the IOF page however, 18,424 hip fractures were reported in 2017 from hospital registrations following the establishment of the ANZHFR in 2015 (ANZHFR, 2018). In New Zealand, more than 84,000 osteoporotic fractures were reported in year 2007 and an estimated increase of 37% between 2007 and 2020 is predicted. Two-thirds of the osteoporotic 15 fractures reported in 2007 occurred among women, as expected. The most frequent fractures were vertebral (33%), rib (25%), forearm (14%), and hip (5%). Although hip fractures were only 5% of the total fractures, they require more attentive care with higher costs. Fractures come with quality adjustment in life years (QALY) losses and huge health and economic burdens. Brown et al. (2007), estimated nearly 11250 QALYs lost in 2007 to osteoporotic fractures and predicted an increase to 15100 in 2020. Health costs were estimated at more than NZ 211 million per year for fracture treatment. As expected, vertebral and hip fractures incurred more expenses, with NZ 55 million costs for fracture care and rehabilitation and NZ 330 million for treatment and management were incurred. In total, approximately NZ 330 million per annum has been spent on osteoporosis-related costs in 2007 with an expected increase in the future. High predictions of the prevalence, incidences, osteoporotic fractures, and the loss of QALYs with an increase of NZ 458 million for health cost is predicted for the year 2020 and beyond. 2.5 : Prevention of Osteoporosis with Tools used to assess Osteoporosis Knowledge, Health Beliefs and Calcium Intake Failure to reach optimal PBM before skeletal maturity and excessive bone loss after PBM contributes to the development of osteoporosis. By 30 years of age, PBM is achieved; thus, identifying controllable factors, such as calcium intake and exercise in the first three decades of life is essential to reduce osteoporosis later in life (Gammage, Gasparotto, Diane, Mack & Klentrou, 2012). Although the science of reducing risk is relatively straightforward, it is not being followed as the high rates of osteoporosis show. Therefore, understanding why and developing strategies to promote bone health behaviours and reduce risky behaviours among young adult females are essential. Specific tools, such as the health belief model (HBM), osteoporosis knowledge test (OKT), osteoporosis health belief scale (OHBS) and osteoporosis self-efficacy scale (OSES) have been developed and validated to assess and understand the level of osteoporosis knowledge, predict osteoporosis health behaviours and to determine an individual’s level of confidence to take action. 2.5.1 : Health Belief Model (HBM) The Health Belief Model (HBM) was developed by a group of social psychologists to understand people's failures to participate in the early detection and prevention of diseases (Hochbaum, Rosenstock, & Kegels, 1952). This model was derived from the psychological and behavioural theory that believes health-related actions are stimulated by the following factors: the existence of health concern, vulnerability to serious health conditions, and having high 16 health benefits that outweigh barriers to health actions (Rosenstock, Strecher, & Becker, 1988). These factors are expressed in five dimensions as below: 1. Perceived susceptibility- refers to an individual's feelings of risk or vulnerability to contracting a disease. 2. Perceived severity- refers to an individual's feelings of seriousness on the implications of the disease or consequences of the disease. 3. Perceived benefits- refers to an individual's belief in particular behaviours that are effective in reducing the threat of a disease. 4. Barriers- refers to individuals' belief in potential obstacles that impede undertaking recommended behaviours to prevent or minimize disease. 5. Cue of Actions- a trigger necessary to promote engagement in preventive behaviours. 2.5.2 : Osteoporosis Health Belief Scale (OHBS) The Osteoporosis health belief scale (OHBS) was developed following the HBM concept to assess health beliefs about osteoporosis and identify the relationship between health beliefs and preventive behaviours (Kim, Horan, Gendler, & Patel, 1991). The two most critical preventive behaviours are calcium intake and exercise, which are incorporated in the OHBS. Knowing and understanding their benefits and barriers is the key to address the prevention and delaying of osteoporosis. This OHBS has been tested for reliability and validity among 150 elderly participants and used in various countries and age populations. The OHBS has seven constructs based on the HBM, emphasizing the benefits and barriers of exercise and calcium intake behaviours. 1. Perceived Susceptibility- refers to an individual's vulnerability to developing osteoporosis. 2. Perceived severity (seriousness) - refers to an individual's beliefs of the harmful effects or consequences of developing osteoporosis. 3. Perceived Exercise Benefits- individual's beliefs of the benefits of exercise to prevent and minimize the threat of osteoporosis. 4. Perceived Calcium Benefits- individual's beliefs of the benefits of calcium intake to prevent and minimize the threat of osteoporosis. 17 5. Perceived Exercise Barriers- refers to the individual's barriers that hinder one's ability to do exercise. 6. Perceived Calcium Barriers- refers to an individual's barriers to having adequate calcium intake. 7. Perceived Health Motivation- refers to one's ability to seek and take health-related actions concerning osteoporosis. Each subscale has six statements with Likert-scale responses as strongly disagree (1); disagree (2); neutral (3); agree (4) and strongly agree (5). The total range of scores for each subscale is 6-30 and 35-210 for the overall score of the OHBS. High perceived susceptibility and severity indicate a threat to physical health, social life, and daily activities that typically stimulate health behaviours. Other positive triggers to behavioural change are the high perceived benefits of exercise and calcium intake behaviours and positive health motivation. In contrast, high perceived barriers are expected to hinder behavioural change. 2.5.3 : Osteoporosis Self-Efficacy Scale (OSES) Following the OHBS, the osteoporosis self-efficacy scale (OSES) was developed to measure the confidence level in actioning behavioural change in exercise and calcium intake (Horan, Kim, Gendler, Froman, & Patel, 1998). This self-efficacy concept was derived from Bandura's social cognitive theory based on two factors that motivate actions: favourable outcomes and individuals' confidence level to execute change. OSES, when added with OHBS, is referred to as the expanded OHBS (Gammage & Klentrou, 2011). The OSES has two sets of scales: 21 items and 12 items; either one can be used. Both scales have a confidence scale range from 0- 100, with a low score indicating not at all confident and very confident for the high score. The OHBS and OSES were tested for validity and reliability and used among different ethnicities, ages, and gender populations (Horan et al., 1998; McLeod & Johnson, 2011). Each tool can either be used separately or together. 2.5.4 : Osteoporosis Knowledge Test (OKT) The Osteoporosis knowledge test (OKT) is another tool developed in the early 1990s to measure osteoporosis knowledge among studied populations (Kim, Horan, & Gendler, 1991). Having knowledge on osteoporosis (risk factors and preventions) is important to make informed choices about the prevention of osteoporosis. Although knowledge is essential it is believed to be mediated by health beliefs to influence behaviour change (Piaseu, Schepp, & 18 Belza, 2002). Supporting findings were found and confirmed by Chang (2008) and Sedlak, Doheny, and Jones (2000) that osteoporosis knowledge alone does not influence behavioural change, but it will along with health beliefs and self-efficacy. The OKT is comprised of three domains: osteoporosis risks, calcium food sources, and exercises to prevent osteoporosis. The OKT consists of 24 items: 9 on osteoporosis risks, 7 on exercise, and 8 on calcium intake. The OKT was validated and used in various countries (China, United States of America, Thailand, and New Zealand). It has been used with modifications such as translation, food choices and wordings to accommodate the study population (Chen, Liu & Cai, 2005; Doheny, Sedlak, Estok & Zellar, 2007; von Hurst & Wham, 2007). The OKT was revised in 2011 due to new evidence on bone health, such as bone accretion, bone remodeling, the role of vitamin D, and other associated risk factors (C. Gendler, Martin, Kim, Von Hurst, 2011). The revised OKT (r-OKT) included an extra domain labelled as general consisting of three questions: one each on peak bone mass (PBM), diagnosis, and treatment. Additionally, five risk factors (elderly, smoking, alcohol use, overweight & eating disorders) and three vitamin D questions were added in the risk and nutrition domains, respectively. Revision was based on previous studies' evidence and recommendations (Gendler et al., 2015). For instance, the vitamin D questions were added by von Hurst and Wham (2007) in New Zealand and smoking and alcohol were added by Chen, Lui, and Cai (2005) in China. The r- OKT consists of 32 items, of which 14 were retained unchanged from the original OKT. There are 11 risk domain questions, 6 exercise domain questions, 12 nutrition domain questions, and 3 general domain questions. Similar to the original OKT it can be measured as the total OKT score, nutrition subscale and exercise subscale. This r-OKT has been tested for reliability and validity among a group of adults of both genders of white people. Gendler et al., (2015) stated that the r-OKT has more strengths as compared to the other knowledge questionnaires such as the osteoporosis knowledge assessment tool (OKAT), facts on osteoporosis quiz (FOOQ), and osteoporosis knowledge questionnaire (OKQ) as it incorporates evidence-based changes and offers ‘don't know’ responses which help to avoid guessing. However, an evaluation of its effectiveness adapting or modifying to suit different cultures and language translation is encouraged. 2.5.5 : Calcium Intake Tool A variety of tools have been used to measure calcium intake among study populations. Calcium-specific food frequency questionnaires (FFQ), food records and rapid assessment 19 method (RAM) for calcium intake are the common dietary tools used to measure estimated calcium intake. The calcium specific FFQ is a shorter version of the FFQ, where only food items with a significant amount of calcium are included. There are many versions of a calcium specific FFQ depending on the numbers of calcium food items included. The calcium specific FFQ can be either short, medium, or long. Frequency of consumption of food item as per serving are asked in daily, weekly, monthly, yearly, and never/infrequently. The number of serve(s) and frequency are calculated to estimate calcium intake per day. The food record is a record where all foods and drinks consumed throughout the day are recorded with serving sizes. In the food record, the serving sizes are recorded based on actual or estimated food item consumed. While the number of days of a record can vary, it is mostly between 3-4 days. Other specific information such as food brands, serving sizes, ingredients and recipes are also recorded alongside the meal and time. Estimated calcium intake is calculated based on all food and drink items or meals consumed unlike the FFQ, only from calcium rich foods. Nutrient intake estimation is more accurate and less biased with a food record compared to using FFQ (Park et al., 2018). Another tool used is the RAM method which consists of 4-6 food categories to measure calcium estimates. The food categories are milk-cheese-yoghurt; fruits-vegetables; bread-cereals-rice- pasta; meat-fish-poultry-dry nuts/seeds; calcium enriched orange juice and calcium supplements with fixed serving sizes (Gammage & Klentrou, 2011; Wallace, 2002). Respondents are asked about the number of times food items are consumed per day or week and to answer yes or no for calcium supplements. The number of servings is then converted into calcium intake and reported as the number of servings per day or week. The RAM tool is less time consuming and quick to gather information in terms of a larger-scale study. Comparison of the calcium estimates from the three tools show that RAM has the tendency to over-estimate calcium intake and not appropriate for young adult population as most would not be taking calcium supplements (Gammage & Klentrou, 2011; Moore, Braid, Falk, & Klentrou, 2007). A previous study using RAM has found only 10% of young adults took calcium supplements (Edmonds, 2009). While FFQ has been known to overestimate/underestimate nutrients intakes compared to a food record, it is beneficial to detect longer consumption habits (Narruz-Varli, Kose, Tatar, Arslan & Koksal, 2018). FFQ usually underestimates calcium intake estimates as compared to food records (Ong et al., 2017; Söderberg et al., 2017). Regardless of its limitations, FFQ is quick and less expensive to use in a large-scale study. 20 2.6.0 : Current Osteoporosis Knowledge, Health Beliefs, and Behaviours (Calcium intake) A recent systematic review reported poor knowledge about osteoporosis among adolescents and young adults (Chan et al., 2018). Thirty-four articles published from 2008 to May 2018 which studied college and university students were analyzed in the review. This is used/referred to in the following paragraphs plus older/newer studies in New Zealand and in other countries. Articles using OKT and OHBS are summarized in Table 1. 2.6.1 : Level of Osteoporosis Knowledge The review aimed to provide findings on osteoporosis knowledge, beliefs, and practices among young adults on bone health. Of the thirty-four articles, thirteen measured osteoporosis knowledge, whereby nine reported having poor knowledge while four with good knowledge scores based on knowledge score’s percentage. While all studies used a cross-sectional study design, different tools including an osteoporosis knowledge questionnaire (OKQ), OKT, r- OKT, osteoporosis knowledge scale (OKS), osteoporosis knowledge assessment tool (OKAT) were used to measure the knowledge. The review also found osteoporosis knowledge varies among age groups, gender, and education level but may not be comparable as different knowledge questionnaires were used. Of the 13 studies that measured osteoporosis knowledge, only 4 had used OKT and r-OKT as tools to measure osteoporosis knowledge. These four will be examined further to compare their level of osteoporosis. Amre, Safadi, Jarrah, Al‐Amer, and Froelicher (2008) found a poor level of osteoporosis knowledge (12.5 out of 23) among 85 nursing students in Jordon. The inadequate knowledge was found throughout the domains of the modified OKT used. Surprisingly these final year nursing students had limited osteoporosis knowledge which was not expected. Similarly, Gammage, Gasparotto, Mack, and Klentrou (2012) found generally poor osteoporosis knowledge among Canadian College students, but women have better knowledge than men. Low-moderate osteoporosis knowledge was reported in a comparable study with 408 American and 409 Chinese college students in separate countries (Ford, Bass, Zhao, Bai, & Zhao, 2011). American students (14.5 out of 24 or 60%) have better osteoporosis knowledge than the Chinese students (11.8 out of 24 or 49%). Interestingly Nguyen and Wang (2012) found higher osteoporosis knowledge among healthcare students in Colombia which was different from the Jordanian nursing students. The r-OKT was used in this study with a 21 total score of 32. Nguyen and Wang found a higher mean of 24 out of 32 (76%). Findings from the four studies were inconsistent as two studies reported low knowledge, one with moderate and another with high osteoporosis knowledge were found among University or College students from China, United States of America (USA), Canada and Colombia. Similar results on moderate osteoporosis knowledge were reported in two New Zealand Studies. von Hurst and Wham (2007) used a modified OKT (26) among women aged 20-49 years living in Auckland. Two additional vitamin D questions were included in the questionnaire; the total score was 26. The participants were selected through snowballing sampling whereby six hundred and twenty- two women participated. Younger women (20-29 years) had the lowest total mean score of 15.8 or 58% compared to 17.3 (67%) in the 40-49 years’ group. The researchers also found older women had good knowledge of the risk and nutrition domains as compared to young adult women. Likewise, a moderate level of osteoporosis knowledge was found among 102 South Asian women between 20 and 49 years of age who were living in Auckland, with a mean score of 15.1 out of 26 or 58% even though the women were highly educated (Tsai, 2008). The same modified OKT tool used above was also used. 2.6.2 : Level of Osteoporosis Health Beliefs 2.6.2.1: Perceived Susceptibility and Severity The osteoporosis health belief scale (OHBS) was used in six of the studies reviewed by Chan et al. (2018). According to the HBM, high perceived susceptibility and severity indicate a threat that increases the likelihood of action. Findings from the six studies found low perceptions of susceptibility and severity is common among young adult population. In a comparison study conducted among American and Chinese students found both study population perceived low susceptibility to osteoporosis using the original OHBS (Ford et al., 2011). Each study population has a mean score of 13.4 out of 30 (45%) and 12.9 out of 30 (43%), respectively. Similarly, low mean perceived severity scores were reported, both groups measured 14.4 out of 30 or 48 percent. These findings confirm that college students from both countries perceived low susceptibility and severity to osteoporosis. Comparable findings were found among 353 nursing students in Damascus in a cross-sectional study. Again, the original OHBS was used in this study with a mean score of 13.4 out of 30 (45%) for low susceptibility of osteoporosis and 17.1 out of 30 (57%) showing moderate perception of severity of osteoporosis (Sayed- Hassan, Bashour, & Koudsi, 2013). 22 Findings similar to Sayed-Hassan et al. (2013) were found among Canadian university students however, a modified OHBS was used (Gammage et al., 2012). Each subscale or construct has 5 items instead of 6 with different scoring from the original OHBS. Participants, both males (1.7 out of 5 or 34%) and females (2.4 out of 5 or 48%), reported having low perceptions of the susceptibility of osteoporosis. Despite having low perceptions of susceptibility, these participants perceived moderate severity of osteoporosis. Males have a mean score of 3 out of 5 or 60% for severity and 3.4 out of 5 or 68% for females. Comparable findings were found in another study conducted in Canada among women and men, between 18 and 25 years of age, between 30 and 50 years of age and 50+ years of age, also using a modified OHBS (Shanthi Johnson, McLeod, Kennedy, & McLeod, 2008). Young adults (18- 25 years old) had low perceived susceptibility, with scores of 10.9 out of 25 (44%) and 8.6 out of 25 or 34% for females and males respectively. Both young males and females have moderate mean scores of severities; 15.6 out of 25 (62%) and 13.8 out of 25 (55%), respectively. Further, the other two studies (Bilal et al., 2017; de Silva et al., 2014) in the review did report low perceived susceptibility of osteoporosis among medical students in Pakistan and Sri Lanka from the modified OHBS. Unfortunately, as both studies did not report the mean score of severity, so they cannot be compared. In addition to articles in the review, other studies among university and college students also found related findings to the above studies. Edmonds, Turner, and Usdan (2012) found low perceived susceptibility and moderate perceived severity among 792 college students between 17 and 31 years of age in the USA using OHBS. A similar finding of low perceived susceptibility was found among university students in Malaysia (Chiang, 2020). Although students perceived low susceptibility, osteoporosis was perceived as a serious disease with a severity mean score of 20. In contrast, Mostafa, Mohtasham, Sakineh, and Mona (2016) reported moderate mean scores of both susceptibility (15.1 out of 30) and severity (18.2 out of 30) among university students in Iran. On the same note, two studies in New Zealand, one reported low perceptions of susceptibility and moderate perceptions of osteoporosis severity among women between 20 and 49 years of age living in Auckland while the other perceived moderate susceptibility and severity (Tsai, 2008; von Hurst & Wham, 2007). In all, the studies found common findings as young adults perceived low susceptibility and moderate severity to osteoporosis. 23 2.6.2.2: Perceived Benefits and Barriers of Calcium Intake Benefits and barriers to calcium intake are hypothesized to influence and predict the intake of calcium foods. Tsai (2008) reported perceived high benefits (23.5 out of 30) of calcium intake and low barriers (13.4 out of 30) of calcium intake among South Asian women in New Zealand. Although von Hurst and Wham (2007) did not report mean scores, 91% of the participants agreed that "taking enough calcium prevents problems with osteoporosis," and 77% agreed with "calcium-rich foods have too much cholesterol." These statements highlight the benefits and barriers of calcium intake among the study population. They found older women (40- 49years) are more likely to agree with the statement of calcium barriers than younger women. Among the European population the benefits of high perceptions of calcium intake were reported (Edmonds et al., 2012; Ford et al., 2011; Gammage et al., 2012). Two USA studies reported having high perceptions of calcium benefits at 22.3 out of 30 (74%) (Edmonds et al., 2012) and 23.2 out of 30 (77%) (Ford et al., 2011). Similar high perceptions of benefits were found among Canadian university students (4.2 out of 5) (Gammage et al., 2012). Studies of young women in two Asian countries also reported participants having high perception scores on the benefits of calcium intake. A study conducted among Medical students between 18 and 40 years of age reported a mean score of 23.1 out of 30 for benefits of calcium intake (Chiang (2020). The study by Aree-Ue and Petlamul (2013) found both younger (between 20 and 35 years of age) and older (≥60 years of age) women perceived high calcium benefits despite living in rural Thailand. Young adults perceived low-moderate barriers to calcium intake as reported by the mean scores of the seven studies mentioned above. According to the responses, these were the common barriers: the high cost of calcium-rich foods, which was reported by Chinese students in (Ford et al., 2011), availability and accessibility reported by Malaysian students (Chiang, 2020), food preferences such as taste, cultural foods, dislike and (Aree-Ue & Petlamul, 2013; Bilal et al., 2017; de Silva et al., 2014) giving up other foods in order to consume calcium foods (Tsai, 2008) and calcium-rich foods are high in cholesterol (von Hurst & Wham, 2007). The findings indicate that barriers to calcium intake vary by countries among young adults. 24 2.6.2.3. Health Motivation Heath motivation measures a person’s self-motivation to seek or practice health-seeking behaviours. All six studies that measured susceptibility and severity in the review reported having moderate-high perceptions of health motivation (Bilal et al., 2017; de Silva et al., 2014; Ford et al., 2011; Gammage et al., 2012; Sayed-Hassan et al., 2013; Shanthi Johnson et al., 2008). On the same note the other four studies also found similar findings (Chiang, 2020; Edmonds et al., 2012; Mostafa et al., 2016; Tsai, 2008). These findings show young adults are health motivated despite having different nationalities and cultural backgrounds. 2.7 : Predictors of Calcium intake (relationship between osteoporosis knowledge, health beliefs and calcium intake) Osteoporosis knowledge and osteoporosis health beliefs may influence a person's calcium intake, as highlighted earlier in the health belief model. Since not all studies reported on earlier measured the predictors of calcium intake, a few studies will be discussed here. Of the two reported studies in New Zealand, only Tsai's study measured the calcium intake of participants using food records. The participants from Tsai's study had a median calcium intake of 685 mg/day, with calcium barriers and health motivations, reported significant predictors of calcium intake in the multivariate analysis. Similar findings were reported among college students in the USA between 17 and 31 years of age for both genders (Edmonds et al., 2012). The participants also had inadequate calcium intake with calcium barriers and health motivations as the predictors of calcium intake. In terms of calcium barriers, a consistent finding was found among female university students in Iran. Calcium barriers were the only significant predictors of calcium intake (Mostafa et al., 2016). In contrast, a study conducted among younger females in grades 8-11 in Canada found osteoporosis knowledge and calcium self-efficacy in addition to calcium barriers as predictors of calcium intake from a simultaneous regression (Gammage & Klentrou, 2011). Osteoporosis knowledge was reported as one predictor of calcium intake along with severity and health motivations among 333 Malaysian university students (Chiang, 2020). Participants had moderate osteoporosis knowledge, moderate susceptibility, severity, calcium barriers, and health motivation scores but perceived high calcium benefits. An inadequate intake of dairy products was reported, as 61.5% did not consume adequate dairy products based on four questions (RAM). The findings revealed the common predictors of calcium intakes were 25 calcium barriers, health motivation and osteoporosis knowledge. Perceived severity of osteoporosis and calcium self -efficacy were found but were not consistent in the studies. 2.8 : Rationale of this study Osteoporosis is a growing public health problem around the world and in New Zealand. It is a preventable disease and making lifestyle changes or following health recommendations is essential to prevent or delay osteoporosis. Additionally, the predictions of increasing fractures, health, and socioeconomic burdens of osteoporosis highlight the need for urgent actions. The first three decades of life is a crucial time to act as it is the period to achieve optimal PBM. Understanding young female adults' knowledge and health beliefs on osteoporosis may help to prevent osteoporosis. Previous studies in New Zealand have identified females have moderate knowledge of osteoporosis and a perceived low threat of developing osteoporosis. Also, females between 15 to 30 years of age tend to have low calcium intake below the EAR as reported in surveys in 1997 and 2009 which is the crucial time to attain PBM. Additionally, the high-fat content of calcium-rich foods was highlighted as a main barrier to the consumption of calcium-rich foods. This study aims to determine the association between calcium intake, osteoporosis knowledge, and osteoporosis health beliefs. The findings of this study will provide insight into the areas to address in terms of preventive behaviours especially for young adults. 26 Table 2.1: Summary of Studies using OKT, OHBS and Associations between Calcium Intake, OKT, and OHBS Reference Study Characteristics Methodology Findings Studies using OKT, OHBS and associations between calcium intake KEY: +ve= positive; -ve= negative; ↓=increase/high; ↑= decrease/low Amre et al., 2008 85 - Nursing University students Male= 58, Female= 27 Age =19- 32 yrs. Study Country- Jordan Objective To explore baccalaureate nursing student’s knowledge of osteoporosis for beginning practice in the community Tool Modified version of OKT and OKQ (out of 23) Overall poor knowledge of osteoporosis – 12.6 out of 23 (54.9%) - graduating student nurses have limited knowledge to perform in community (health promotion and diseases prevention). (Aree-Ue & Petlamul, 2013) 187 females Age-20-35years &≥60 years Study Country- Thailand Objective To compare knowledge, attitude towards preventive behaviours between younger & older women Tools OKT OHBS OSES Statistical analysis T-test and bivariate Education level- younger women had formal education & reached a higher level as compared to older women OKT- 11.54±3.78 (young); 7.71±3.96 (older). The statistically significant difference that older women had lower knowledge score OHBS- - susceptibility – 16.86±4.79 (young); 13.77±6.04(older) Statistically proven young women had greater susceptibility - Severity -22.66±3.89(young);23.34±6.06 (older) - Calcium benefits-23.20±3.75 (young); 22.25±4.34 (older) - **Calcium barriers- 15.29±3.34 (young); 17.63±4.20 (older) - Health motivation-23.21±3.34 (young); 23.69±4.57 (older) OSES- calcium intake- 43.35±10.12(young);38.80±14.48 (older) 27 Correlation - Knowledge has a positive correlation to osteoporosis preventive behaviours in young women - Positive & significant correlations were found with preventive actions and overall health beliefs in older women - all other relationships not significant Bilal et al., 2017 400 university female medical students Mean age 19.4 Study Country- Pakistan Objective - To assess osteoporosis knowledge, beliefs & practices among university students Tools - OKAT- osteoporosis knowledge - OHBS- osteoporosis health beliefs - FFQ- calcium intake Statistics Analysis - Descriptive analysis - Chi Square tests Calcium Intake= average 510mg/day (low calcium intake) -only 29% met calcium RDA OKAT= 8% had good knowledge score, 49% average & 43% poor OHBS - Susceptibility – believe chances of getting osteoporosis is high (14%) - Severity-osteoporosis would made daily living challenging (81%) - Calcium barriers- calcium food is difficult to eat (30%) - Health Motivation- take steps to improve health (62%) Chang, 2006 265 women 25-45years old Study country- Taiwan Objective -Examine associations between demographics knowledge (15 questions on risk & prevention), health beliefs, and calcium intake Tools Self-designed questionnaire - Demographics - Knowledge (15 questions on risk & prevention) - OHBS (4 subscales- susceptibility, severity, calcium benefits & barriers). Calcium intake- based on the authors' previous study & literature Statistical analysis - T-test and ANOVA test Calcium intake= 454.7±66 (low calcium intake) Knowledge (out of 15) = 12.1±2.8 or 80.6% (high score) OHBS (out of 5)- - susceptibility=2.8±0.5 (moderate score) - Severity=2.2±0.7 (low score) - calcium benefits=1.9±0.4(low score) - calcium barriers=2.9±0.7 (moderate score) Correlations between calcium intake & variables -Women who had high calcium intake are likely to be less knowledgeable, (-0.324) perceived greater susceptibility (0.252), severity (0.292), perceived fewer barriers (-0.293) are older (0.258), have a family history (4.730), graduated from high school (6.210) and self-rated health (0.253) 28 - Bivariate correlations Stepwise regression analysis Predictors of calcium intake Knowledge, number of children self -rated health, level of education, body mass index (Chiang, 2020) University students Both gender (95males & 238 females) 18-40years old Study country- Malaysia - Objective - Examine the association between osteoporosis knowledge, health beliefs & calcium intake among medical sciences students Tools - OKT (16 items) - OHBS (30 items) - 6-13.9 (low); 14-21.9 (mod); 22-30 (high) - OPBS- dietary calcium questions used to measure calcium intake - Calcium servings modified to meet the daily requirement in the local context (Malaysia) - Adequate = ≥7 servings daily - Inadequate= < 7 servings daily Statistical analysis - Descriptive - Correlations - Multiple linear regression - OKT (total score of out of 100) - Mean total score=50.4±16.48 (moderate score) - 4.5% (high score); 55.9% (moderate); 39.6% (low) OHBS (out of 30) - susceptibility – 14.2±4.02 (moderate score) - severity – 20.4±4.67 (moderate score) - calcium benefits -23.1±3.94 (high score) - calcium barriers- 14.4±3.99 (moderate score) - health motivation -21.6±3.79(moderate score) Calcium intake -61.5% had inadequate dairy products Correlations -Calcium intake has not significant correlations with OKT and OHBS constructs - Positive correlations between OKT and health motivation(r=0.173), calcium benefits(r=0.127); and negative correlations with calcium barriers (r=-0.208) -Positive correlations between age and OKT (r=0.355) and health motivation (r=0.149) but negative correlations with calcium barriers (r=0.120) Predictors of dairy products -OKT (Beta= -0.175) perceived severity (Beta= -0.122) and health motivation (Beta=0.171) are predictors of dairy intake p value <0.001 with R square of 0.060. de Silva et al., 2014 186 female students medical School entrants - Mean age 20.7 Objectives - Determine the knowledge, beliefs, and practice regarding Osteoporosis among young female entering Medical Schools in Sri Lanka Calcium Intake =528mg per day (only 18.5% met RDA) OKAT= 51.6% average score, 40.8% poor score (overall lacked knowledge on risk and protective factors 29 Study country- Sri Lanka Tools -OKAT- 20 question. - The modified OHBS (3 questions per construct) - FFQ for assessing calcium intake. Statistical Analysis: - Descriptive analysis - Pearson Chi-Square OHBS -Susceptibility = low as only 13.9% (n=26) of women agreeing that they had higher chances of getting osteoporosis. - Severity= high as 83.3% felt having osteoporosis would make daily life difficult. Calcium barriers- low as 15% can’t tolerate and dislike calcium rich foods Health Motivation- moderate as 59% were motivated to improve health (Edmonds et al., 2012) - 792 College students - 17-31 years old - Both gender Study country - the USA Objectives -Examine the relationship between osteoporosis knowledge, health beliefs, and calcium intake Tools Osteoporosis knowledge test (OKT) - 24 items (total score is 24) - 1-point score per correct response Osteoporosis health belief scale (OHBS) - 42 items with 7 constructs. - Rated using a Likert scale (1-5) - Each construct has a possible score of 6-30 - High scores indicate high perception except for barriers. Barriers high scores indicate low perceptions Osteoporosis preventing behaviour survey (OPBS) has questions to measure calcium intake - 39 items consisting of questions on exercise, calcium intake, risk factors - Calcium intake is measured based on4 questions of the frequency of consumption of calcium-rich foods & intake of calcium supplements - Food is converted into servings classified as follow - Inadequate - < 4 servings/week - Moderate 5 servings /week or 1 serving/day - Adequate- 2-3 servings/day Mean age – 20.6 years Female (53.6%), male (46.45) Ethnicity- white (64.5%), remaining others OKT- (total score is out of 24) * No mean score of knowledge given but low scores, especially in the risk domain and few calcium questions. Lack of knowledge OHBS (total score is out of 30) * susceptibility =13.64 ±5.09 (low score) * severity =17.34± 4.37 (moderate score) * exercise & calcium benefits =23.23± 5.33, 22.26± 4.61 (high scores for both) *exercise & calcium barriers =24.27±4.62, 22.82± 4.57 (low scores for both) * health motivation =19.87±4.34 (high moderate score) Calcium Intake * 62.5% had an inadequate calcium intake Correlations & Multiple Regression *Significant positive correlation between health motivation & calcium intake (r= 0.204, p=0.000). High calcium intake, high health motivation * No significant correlations with other variables Predictors 30 Statistical analysis - Descriptive - Bivariate correlation Multiple regression * Health motivation, perceived barriers (calcium & exercise), age, ethnicity, and physical activity were predictors of calcium intake. All at a significant value  Low calcium barriers yet inadequate calcium intake Ford et al., 2011 774 university students 408 from the USA & 409 from China Study country- USA & China Objective Investigate differences in osteoporosis knowledge, self-efficacy, and health beliefs among Chinese and American students. Tools -OKT -OHBS -OSES (only for exercise so won’t be reporting in findings) Statistics Analysis Chi-Square Analysis OKT Osteoporosis differences were noted (US=14.52, Chinese=11.82) OHBS Susceptibility – low scores for both countries (US=13.4, Chinese= 12.9) Severity- both had mean score of 14- Calcium benefit- high for US (23) and moderate for China (18) Calcium barriers- low for US (13) and moderate for China (15)- cost was the main barrier to calcium intake among Chinese students Health Motivation- both had higher mean score of 21 and 20 respectively. (Gammage & Klentrou, 2011) - 510 grade 9- 12 females - From 8 different schools Study country- Canada Objective To investigate if the expanded health belief model (EHBM) could predict calcium intake & exercise Tools - OKT (out of 24) - OHBS (out of 5 for each constructs) - Osteoporosis self-efficacy scale (OSES) 21 items which measure confidence of preventive behaviours (calcium intake & exercise)- 11 (calcium) 10 (exercise) - Rapid assessment method (RAM) for calcium intake- mg /day (6 food categories) OKT (out of 24-total scores) - Mean =11.68±3.51 (low score) OHBS (out of 5) susceptibility =2.39±.74 (low score) severity =3.27±.69 (moderate score) calcium benefit =3.56±.58 (high score) calcium barrier =2.24±.68 (low score) exercise benefits =3.56±.59 (high score) exercise barriers =2.61±.70 (moderate score) health motivation =3.51±.68 (high score) Calcium Intake – mean = 1315.18±479.18 – adequate, meets recommendation for age group 31 - Physical activity questionnaire adolescent (PAQ_A) Statistical Analysis - Descriptive - Bivariate correlations Simultaneous regression – calcium intake Correlation & Regression *Calcium intake negative associated with susceptibility, calcium & exercise barriers * Calcium intake positive associated with severity, health motivation, knowledge &self-efficacy (calcium & exercise). *Correlations were in small-medium size *Predictors of calcium intake Calcium barriers (-ve), calcium self-efficacy (+ve), and knowledge (+ve) Gammage et al., 2012 - Objectives Gender differences in osteoporosis-related knowledge and beliefs Tool OKT (24)- measured as percentage Modified OHBS (35)- each construct out of 5 OSES (10 calcium intake & 11 exercise) Statistical Analysis Descriptive analysis Multiple Regression- predictors physical activity OKT Male = 57.7; female =61.4 OHBS Susceptibility - Male-=1.69; female=2.42 Severity- Male= 3.07; female=3.4 Calcium benefits -Male= 4.18; female=4.23 Calcium barriers- Male =1.76; female=2.14 Health Motivation-Male=3.93; female=3.77 Females are more susceptible to osteoporosis than males Nguyen & Wang 2012 206 Nursing students Age 21 to 27 yrs. Majority are female and Caucasian Study country- Colombia Objective Investigated osteoporosis knowledge in students from different health disciplines students Tools - Revised OKT (out of 32) Statistical Analysis Descriptive analysis OKT= High at 24.4 out of 34 (76.3%) - Osteoporosis knowledge discrepancies were found between students from health disciplines and year of class - Dietetics students have higher scores (Mostafa et al., 2016) 239 University Female students 18-44years old Mean age 22.17±2.66 Objective -to assess the determinants of calcium intake based on the health belief model Tools -Self- designed questionnaire consist of Calcium intake means= 945.63±629.19mg/day- below Recommended Daily Allowance (RDA) OHBS- all scores out of 30. (Low score in barriers indicate high perception while high for the other) 32 91.6% married Study country- Iran - Demography (9 items) - OHBS (24items- susceptibility, severity, calcium benefits & calcium barrier-6 questions each) - Food frequency-FF (19 items)- mg/day - susceptibility 15.19±4.45 (moderate score) - Severity 18.19±4.45 (moderate score) - calcium benefits 13.79±2.72 (low score) - calcium barriers 13.49±3.96 (moderate score) Statistical analysis Independent t-test, ANOVA - Pearson's Correlation Correlation- - Perceived susceptibility (-0.201**) - perceived severity ( -0.15*) - perceived barriers (-0.206**) Multivariate regression - Calcium barriers the strongest determinant of calcium intake (beta= -0.14, p =0.000) Sayed-Hassan & Hyam Bashour & Abir Koudsi, 2013 353 nursing students Study Country- Damascus (Syria) Objective Determine the level of osteoporosis knowledge and beliefs among nursing college students in Damascus Tools OKAT (out of 20) OHBS OKAT Mean knowledge score – 7.9 out of 20 (39. 6%) Low osteoporosis knowledge OHBS Susceptibility - low score at 13.2 (44%) Seriousness – high score than susceptibility at 17.1 (57%) Health Motivation- high score than susceptibility at 18.4 (61%) Statistical Analysis Descriptive Statistics Chi square test ANOVA Univariate linear regression Univariate linear regression OKAT domain on “knowledge of preventive factors” was a strong predictor of three OHBS subscales namely “benefits of exercise”, “benefits of calcium intake,” and “barriers to exercises’’. Shanti -Johnson et al 2008 Sample of 300 3 age groups - 18 -25 years - 30 years 50 - 50 plus Both gender (Male & Female) Study country- Canada Objective Compare osteoporosis health beliefs among different age and gender groups. Tools Modified OHBS (3)- susceptibility, severity & health motivation Statistical Analysis - Descriptive statistics OHBS Mean score by young adult of both gender on susceptibility was lower. Men and Women in the age groups, women had a high score in susceptibility in each group. No significant differences in seriousness and health motivation 33 - ANOVA - MANCOVA scores between the age group and gender. Tsai 2008 Sample of South Asian women 102 women aged 20- 49years old Study Country- New Zealand Objective Determine osteoporosis knowledge, health beliefs & dietary calcium intake among South Asian women Tool OKT (out of 26) OHBS (out of 30) Food Record Statistics Analysis Descriptive analysis Bivariate Correlation Multiple Regression Calcium intake= 685mg /day – lower than calcium EAR OKT- Average knowledge score at 15.1 out of 26 (58%) OHBS Susceptibility – low score at 17.0 Severity- high score at 19.3 Calcium Benefits- perceived higher benefits as 3 statements reported having >80% agreed (no mean score) Calcium Barriers- low score at 13.4 Health Motivation- high score at 22.4 Predictors of calcium Intake Calcium barriers, health motivation and use of dietary supplements (von Hurst & Wham, 2007) 622 women 20-49years Study country -New Zealand Objective - determine osteoporosis knowledge & health beliefs & attitude towards preventive behaviours - if age is the predictor factor to knowledge & health beliefs Tools - modified OKT (26 items) - OHBS (42 items) Statistical Analysis - Descriptive analysis - Univariate linear regression analysis OKT- moderate knowledge in total (16.4 or 63%) - 15.8±3.9 (20-29); 16.4±4.1 (30-39); 17.3±4.0 (40-49) - All fall in the moderate score - young women have the lowest knowledge mean score OHBS - susceptibility - low score -no significant difference in all 3 age groups - severity – low and no significant difference found - benefits of calcium & exercise had high scores with no significant difference - high barriers of barriers in young women and least in older women. No difference in scores of calcium barriers. Calcium foods rich in cholesterol (77%) is one of the obstacles to calcium intake - Older women have greater health motivation scores. Correlations 34 - Significant correlation between knowledge & health motivation - Benefits of exercise & calcium were predictors of health motivation, while barriers were the negative predictors. - No significant relations between susceptibility, severity, and health motivation. Wallace 2002 273 non -traditional Objective Calcium intake – 66.7% had calcium intake (<1200mg) College women - Examine associations between EHBM & -25% consumed half of RDA or less. Computer class protective behaviours -21.5% used calcium supplements students Tools 17-64 years of age - Facts on osteoporosis quiz (FOQ)- 26 items Knowledge – 16.83±5.25 (65%)- mod knowledge - OHBS Study country- USA - OSES (12 item) OHBS & SE-calcium- univariate - RAM- calcium intake - Adequate calcium >1200mg.day ↓Ca↓Ex ↑Ca↑Ex ↓Ca↑Ex ↑Ca↓Ex - Inadequate calcium < 1200mg Sus-18.26±4.59 17.61±4.56 16.04±4.76 16.32±4.59 - Weight bearing exercise Sev -18.80±4.22 20.22±4.23 18.99±4.02 18.37± 4.8 - Adequate = ≥90min/week ben - 24.32±3.12 25.53±3.15 25.23±3.01 24.61±3.92 - Inadequate =< 90min/week bar -14.42±3.56 10.68 ±4.14 12.03±3.70 12.71 ±3.67 H. Mo- 20.90 ±3.74 23.95± 3.64 23.24±4.14 20.53±4.36 Statistical analysis SE - 60.64± 22.18 79.93±18.55 70.55 ±22.09 67.11 Univariate ±24.09 Pearson correlation Stepwise multiple regression Predictors - Perceived susceptibility is one of the predictors of the Ca/Ex group. 35 3.1 : Study Design CHAPTER 3: METHODOLOGY This study is a secondary analysis of the Love the Bones (LTB) data. The LTB was a longitudinal study conducted among young female adults in 2017. The aim of the original study was to evaluate change on osteoporosis knowledge, beliefs and practices among young women who participated in a novel osteoporosis prevention. The study has two phases with two planned follow-ups at six months and two years. This present analysis focuses on the data collected in phase one and the food frequency questionnaire (FFQ) in phase two three months later. The analysis aimed to determine the association between calcium intake, osteoporosis knowledge, and osteoporosis health beliefs. 3.2 : Sample Size The sample size for this analysis was calculated using G*power version 3.1. The linear multiple regression fixed model was used as a statistical test with A priori as a power analysis. A moderate effect size of 0.15, α err probability of 0.05, power (1-β err prob) of 0.80 and 8 predictors for the regression analysis. The calculated sample size of 109, which was present in the data set. 3.3 : Ethical Approval and Consideration The Massey University Human Ethics Committee granted ethical approval (SOA 17/06) to conduct the study. All participants were given an information sheet, and written consent was obtained before obtaining the data. Identification numbers were assigned to the participants for confidentiality purposes. Furthermore, participation was voluntary, with acceptable withdrawal at any time during the study period. Copies of ethical approval, information sheet, and consent are available in Appendix 1,3 and 4. 3.4 : Subject Recruitment Subjects were recruited through flyers posted on university campuses, social media, and print media and by word of mouth in the lower North Island of New Zealand. Females the age of 18 and 25 were requested to complete a health screening form to check for participation eligibility. Conditions that might affect bone health/density were reasons for exclusions from the study, as stated in the flow chart below and in the screening form (appendix 2). 36 Take Home 5. 3-Day Diet Diary/ Food records (3DDD/FR) Phase One: Primary data Onsite 1. Osteoporosis Health Belief Scale (OHBS)* 2. Osteoporosis Knowledge Test (OKT)* 3. Bone Health Questionnaire (BHQ)* - FFQ 4. QUS Achilles measurement of non- dominant Recruitment Criteria Eligibility Criteria (Primary data)  Healthy women  18-25 years  Not pregnant  Living in Palmerston North & Wellington Regions  Have not been diagnosed with the following:  Juvenile Rheumatoid Arthritis  Diabetes  Uncontrolled Thyroid Disease  Inflammatory Bowel Disease (e.g. Chron’s, Ulcerative Colitis, Celiac Disease)  Chronic Renal Disease  Clinically significant Liver Disease  Not medically treated with the following  Corticosteroid tablets (prednisone, cortisone) daily for > 3months previously Health Screening Questionnaire Consent Form Data Collection Figure 3.1: Flow diagram of recruitment and data collection for primary data & secondary analysis Phase Two –Onsite Primary data 1. BHQ- FFQ2 Completed data: Secondary Analysis  130 OKT  130 OHBS  129 BHQ)  93 completed 3- DDD record/Food records  91 completed FFQ 2 37 3.5 : Data Collection Participants who met the eligibility criteria were asked to visit the Human Nutrition Research Unit (HNRU) at Massey University, Palmerston North campus. Each participant completed the three Survey Monkey questionnaires via computer on-site. A 3 day-diet diary (3DDD) was provided as a take-home paper booklet with instructions explained verbally by researchers and written instructions on the record sheet. The FFQ 2 was taken from the BHQ during phase two and it was done onsite. Data Collection Tools 3.5.1 : Revised Osteoporosis Knowledge Test (r-OKT) Questionnaire. The original Osteoporosis Knowledge test (OKT) was developed by Kim, Horan, and Gendler (1991) and was later revised in the study by Gendler et al., (2015). The revised OKT (r-OKT) has 32 questions with four domains: risk, exercise, nutrition, and general. Following pre-testing of the questionnaire, a modified r-OKT with 29 questions was used in this present study. Three questions were omitted due to difficulty and confusion to answer questions such as Calcium and Vitamin D’s recommended daily intake (RDI). Another omitted question is overweight as a risk factor of osteoporosis. The modified 29 questionnaire has 10 risk domain statements on genetic and environmental risk factors for osteoporosis with four choices of response regarding how these factors influence risk (more likely, less likely, neutral, and don't know). The other three domains have multiple choices questions with four possible answers. The exercise domain has six questions, 10 questions for nutrition, and three questions for the general domain. Questions for the exercise domain are on duration and frequency of exercise and the best types of exercise to reduce osteoporosis risk. The nutrition questions are mostly about food sources of calcium, with three about vitamin D. The general domain has three questions, one question each on peak bone mass (PBM), osteoporosis diagnosis, and osteoporosis treatment. Each correct response scores one, with zero score for incorrect or don't know answers. The scores are reported either as total OKT score or as exercise subscale and nutrition subscale. The maximum score for OKT is 29, with a high score defined as 21-29 (≥72%), moderate from 15-20 (50%- 69%), and ≤14 (<50%) as low based on percentage cut-off used by Evenson and Sanders (2016) and Nguyen and Wang (2012). The osteoporosis risk (10) and general (3) items are included in both subscales. This gives the exercise subscale the total score of 19 (13 from risk and general knowledge + 6 exercise) and 23 for the nutrition subscale (13 from risk and general knowledge + 38 10 nutrition). Permission to reproduce and use the questionnaire was obtained before the study. The questionnaire is included in Appendix 5. 3.5.2 : Osteoporosis Health Belief Scale (OHBS) Questionnaire The Osteoporosis He