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. 
 
Risk-based surveillance in animal health
A thesis presented in partial fulfillment of the requirements for the degree of
Doctor of Philosophy
at Massey University, Palmerston North,
New Zealand
by
Deborah Jayne Prattley
January 2009
Supervisors:
Associate Professor M.A. Stevenson
Professor R.S. Morris
Institute of Veterinary, Animal and Biomedical Sciences
Massey University
Palmerston North, New Zealand
i
Abstract
Animal health surveillance is an important part of animal health care, particularly in
countries dependent on livestock for food production and international trade. There are
two major issues related to the provision of effective surveillance activities. Firstly, for
good information to become available, the design and conduct of data collection activ-
ities should be carried out following sound statistical principles. In reality, constraints
such as imperfect tests and unavoidably-biased sampling strategies hinder straightfor-
ward analysis and interpretation of survey results. Risk-based surveillance is used to
target high-risk sub-populations to increase efficiency of disease detection; however,
biased datasets are generated.
This thesis develops methodologies to design risk-based surveillance systems and al-
low statistically valid analysis of the inherently biased data they generate. The first
example describes the development of a method to analyse surveillance data gathered
for bovine spongiform encephalopathy (BSE). The data are collected from four dif-
ferent surveillance streams of animals tested for BSE, with each stream containing
unavoidable biases and limitations. In the BSurvE model, these data are combined
with demographic information for each birth cohort to estimate the proportion of each
birth cohort infected with BSE. The prevalence of BSE in a national herd can then
be estimated using the method of moments, whereby the observed number of infected
animals is equated with the expected number. The upper 95% confidence limit for the
prevalence is estimated both for infected countries and for those where no BSE has
previously been detected.
A similar approach to that used in BSurvE is then applied to surveillance data for
trichinellosis, for which risk-based post-mortem testing is also performed. Negative
results from multiple species using different, imperfect tests are combined to give an
estimate of the upper 95% confidence limit of the national prevalence of trichinellosis
in a reference population. This method is used to provide support for freedom from
trichinellosis in Great Britain.
A different approach to risk-based surveillance is explored as the surveillance strategy
for detection of exotic causes of abortion in sheep and goats in New Zealand is examined.
Using a geographic information system (GIS) maps of disease risk factors were overlain
to produce a risk landscape for the lower North Island. This was used to demonstrate
ii
how areas of high- and low-risk of disease occurrence can be identified and used to
guide the design of a risk-based surveillance programme.
Secondly, within one surveillance objective there may be many ways in which the
available funds or human resources could be distributed. This thesis develops a method
to assess BSE surveillance programmes, and provides tools to facilitate BSE detection
on the basis of infection risk and to increase the efficiency of surveillance strategies.
A novel approach to allocation of resources is developed, where portfolio theory con-
cepts from finance are applied to animal health surveillance. The example of surveil-
lance for exotic causes of sheep and goat abortion is expanded upon. Risk of disease
occurrence is assessed for a population over different time periods and geographical
areas within a country, and portfolio theory used to allocate the number of tests to
be carried out within each of these boundaries. This method is shown to be more
likely to detect disease in a population when compared to proportional allocation of
the available resources.
The studies presented here show new approaches that allow better utilisation of
imperfect data and more efficient use of available resources. They allow development of
surveillance programmes containing an appropriate balance of scanning and targeted
surveillance activities. Application of these methods will enhance the implementation
and value of surveillance in animal health.
iii
Publications
Peer-reviewed papers
Prattley, D.J., Cannon, R.M., Wilesmith, J.W., Morris, R.S. and Stevenson, M.A. 2007.
A model (BSurvE) for estimating the prevalence of bovine spongiform encephalopathy
in a national herd. Preventive Veterinary Medicine 80 330-343.
Prattley, D.J., Morris, R.S., Cannon, R.M., Wilesmith, J.W. and Stevenson, M.A. 2007.
A model (BSurvE) for evaluating national surveillance programs for bovine spongiform
encephalopathy. Preventive Veterinary Medicine 81 225-235.
Prattley, D., Morris, R.S., Stevenson, M.A. and Thornton, R. 2007. Application of
portfolio theory to risk-based allocation of surveillance resources in animal populations.
Preventive Veterinary Medicine 81 56-59.
Reports
Prattley, D., Morris, R.S., Sujau, M., Sauter-Louis, C., Cogger, N. and Cannon, R.M.C.
2007. Interpretation of Trichinella surveillance data from Great Britain. A report for
the United Kingdom’s Food Standards Agency.
Cogger, N., Morris, R.S. and Prattley, D. 2007. Farm-specific risk analysis system
to determine risk status for Trichinella. A report for the United Kingdom’s Food
Standards Agency.
Prattley, D. and Stevenson, M.A. 2006. Saleyard risk analysis. A report for the Ministry
of Agriculture and Fisheries.
Prattley, D. 2006. Scenario tree analysis of surveillance for vector-borne causes of
ovine abortion in New Zealand. A study conducted for the Ministry of Agriculture and
Forestry, New Zealand.
Prattley, D. 2005. Review on the use of serological surveillance for FMD in sheep. A
report for the National Centre for Disease Investigation, Biosecurity New Zealand.
iv
Prattley, D. 2005. Risk-based surveillance for causes of ovine and caprine abortion
exotic to New Zealand. A case study conducted for the Ministry of Agriculture and
Forestry, New Zealand.
Wilesmith, J.W., Morris, R.S., Stevenson, M.A., Cannon, R.M., Prattley, D.J. and Be-
nard, H. 2004. Development of a method for evaluation of national surveillance data and
optimization of national surveillance strategies for bovine spongiform encephalopathy.
A project conducted by the European Union TSE Community Reference Laboratory,
Veterinary Laboratories Agency Weybridge, United Kingdom.
Selected conference papers and presentations
Cogger, N., Morris, R. and Prattley, D. 2007. Solutions to a “Tricky” surveillance
problem. Australian and New Zealand Chapter of the Society for Risk Analysis.
Prattley, D., Morris, R.S., Stevenson, M.A. and Thornton, R. 2006. Matching risk
and resources in design of surveillance strategies. Presented as a lead paper at the
conference of the International Society of Veterinary Epidemiology and Economics.
Prattley, D., McIntyre, L., Morris, R.S., Stevenson, M.A. and Howe, M. 2006. Clinical
practice as a source of veterinary surveillance data - what is it worth? Presented at the
Annual Conference of the New Zealand Veterinary Assocation.
Prattley, D., Cannon, R.M., Wilesmith, J.W., Stevenson, M.A. and Morris, R.S. 2006.
Scoring points - an objective method to evaluate BSE testing data and optimise surveil-
lance activity. Presented at the Annual Conference of the New Zealand Veterinary
Assocation.
Prattley, D., Morris, R.S., Stevenson, M.A. and Thornton, R. 2006. Matching risk and
resources in design of surveillance strategies. Presented at the Annual Conference of
the New Zealand Veterinary Assocation.
v
When
you are
describing
A
shape
or
sound
or tint
Don’t
state
the
matter
plainly
But
put
it in
a
hint
And
learn
to
look
at
all
things
With
a
sort
of
mental
squint
Lewis Carroll
vi
Contents
Abstract ii
Publications iv
Contents xi
List of Tables xiii
List of Figures xvi
1 Introduction 1
2 Literature review 3
2.1 Introduction to animal health surveillance . . . . . . . . . . . . . . . . . 3
2.2 Classification of surveillance activities . . . . . . . . . . . . . . . . . . . 5
2.3 Data analysis in risk-based surveillance . . . . . . . . . . . . . . . . . . . 10
2.4 Combining data sources . . . . . . . . . . . . . . . . . . . . . . . . . . . 11
2.5 Evaluation of alternative surveillance strategies . . . . . . . . . . . . . . 16
2.6 Allocation of resources . . . . . . . . . . . . . . . . . . . . . . . . . . . . 21
2.7 References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 24
3 BSE prevalence estimation 33
3.1 Abstract . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 33
3.2 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 33
3.3 Methods . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 35
vii
3.3.1 Input data . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 35
3.3.2 BSE status assessment . . . . . . . . . . . . . . . . . . . . . . . . 36
3.4 Results . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 40
3.4.1 Input data . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 40
3.4.2 BSE status assessment . . . . . . . . . . . . . . . . . . . . . . . . 43
3.5 Discussion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 43
3.6 Conclusions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 46
3.7 References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 46
4 Evaluating national surveillance programmes for BSE 49
4.1 Abstract . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 49
4.2 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 49
4.3 Methods . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 51
4.3.1 BSE surveillance assessment . . . . . . . . . . . . . . . . . . . . . 51
4.3.2 Surveillance resource allocation . . . . . . . . . . . . . . . . . . . 52
4.4 Results . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 54
4.4.1 BSE surveillance assessment . . . . . . . . . . . . . . . . . . . . . 54
4.4.2 Surveillance resource allocation . . . . . . . . . . . . . . . . . . . 54
4.5 Discussion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 57
4.6 Conclusions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 59
4.7 References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 59
5 Integration of diverse surveillance data 61
5.1 Abstract . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 61
5.2 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 62
5.3 Methods . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 63
5.3.1 Fox data . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 63
5.3.2 Horse data . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 63
5.3.3 Pig data . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 64
viii
5.3.4 Species and sub-population weighting . . . . . . . . . . . . . . . 66
5.3.5 Theory for estimating prevalence . . . . . . . . . . . . . . . . . . 67
5.3.6 Analysis . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 68
5.4 Results . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 68
5.4.1 Sampling effort . . . . . . . . . . . . . . . . . . . . . . . . . . . . 69
5.4.2 Upper confidence limit (95%) for the estimate of prevalence . . . 71
5.5 Discussion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 72
5.6 Conclusions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 77
5.7 References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 77
5.8 Appendix 1: Additional figures . . . . . . . . . . . . . . . . . . . . . . . 79
5.9 Appendix 2: Sample weightings . . . . . . . . . . . . . . . . . . . . . . . 96
5.9.1 References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 105
6 Risk-based surveillance for exotic ovine abortion 109
6.1 Abstract . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 109
6.2 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 109
6.3 Methods . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 110
6.3.1 Surveillance zones . . . . . . . . . . . . . . . . . . . . . . . . . . 110
6.3.2 Incursion zones . . . . . . . . . . . . . . . . . . . . . . . . . . . . 110
6.3.3 Sheep and goat population data . . . . . . . . . . . . . . . . . . 111
6.3.4 Abortion data . . . . . . . . . . . . . . . . . . . . . . . . . . . . 111
6.3.5 Estimation of abortion and case submission rates . . . . . . . . . 111
6.3.6 Number of cases to test to detect disease . . . . . . . . . . . . . 113
6.3.7 Surveillance system design . . . . . . . . . . . . . . . . . . . . . . 113
6.4 Results . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 114
6.4.1 Surveillance zones and population data . . . . . . . . . . . . . . 114
6.4.2 Incursion zones . . . . . . . . . . . . . . . . . . . . . . . . . . . . 115
6.4.3 Abortion data . . . . . . . . . . . . . . . . . . . . . . . . . . . . 117
ix
6.4.4 Estimation of abortion and case submission rates . . . . . . . . . 124
6.4.5 Number of cases to test to detect disease . . . . . . . . . . . . . 127
6.4.6 Surveillance system design . . . . . . . . . . . . . . . . . . . . . . 128
6.4.7 Incursion zone surveillance . . . . . . . . . . . . . . . . . . . . . 130
6.5 Discussion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 133
6.6 Conclusions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 137
6.7 References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 138
7 Application of portfolio theory to risk-based surveillance 140
7.1 Abstract . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 140
7.2 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 141
7.3 Methods . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 142
7.3.1 Characteristics of a portfolio . . . . . . . . . . . . . . . . . . . . 142
7.3.2 Portfolio allocation . . . . . . . . . . . . . . . . . . . . . . . . . . 143
7.3.3 The surveillance portfolio . . . . . . . . . . . . . . . . . . . . . . 144
7.3.4 Evaluation of resource allocation methods . . . . . . . . . . . . . 147
7.4 Results . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 147
7.4.1 Safety first: multiple exotic diseases . . . . . . . . . . . . . . . . 147
7.4.2 Safety first: regional and temporal allocation . . . . . . . . . . . 147
7.4.3 SIM: allocation of laboratory tests . . . . . . . . . . . . . . . . . 150
7.4.4 Evaluation of resource allocation methods . . . . . . . . . . . . . 151
7.5 Discussion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 151
7.6 Conclusions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 154
7.7 References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 155
8 General discussion 157
8.1 A review of the studies in this thesis . . . . . . . . . . . . . . . . . . . . 157
8.2 Future research . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 163
8.3 Conclusions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 165
x
8.4 References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 165
Bibliography 167
xi
List of Tables
3.1 The BSurvE age data entry table for country A . . . . . . . . . . . . . . 41
3.2 Observed number of animals tested and testing positive . . . . . . . . . 44
4.1 The observed number of animals tested and testing positive . . . . . . . 52
4.2 Calculated number of points required . . . . . . . . . . . . . . . . . . . . 56
4.3 Summary of the number of points gained . . . . . . . . . . . . . . . . . 57
4.4 Calculated time to achieve point target . . . . . . . . . . . . . . . . . . . 57
5.1 Categorisation of high- and low-risk accommodation for pigs . . . . . . . 65
5.2 Weighting values assigned to surveillance tests . . . . . . . . . . . . . . 69
5.3 Estimated number of pigs present by sub-population . . . . . . . . . . . 70
5.4 Estimated number of pigs slaughtered per year . . . . . . . . . . . . . . 71
5.5 Estimated number of pigs tested in 2005 . . . . . . . . . . . . . . . . . . 71
5.6 95% upper confidence limit for the prevalence estimate . . . . . . . . . . 74
5.7 Results of studies investigating prevalence in the red fox . . . . . . . . . 97
5.8 Results of testing of horsemeat in various countries . . . . . . . . . . . . 99
5.9 Results of studies investigating infection in domestic pigs . . . . . . . . 102
6.1 Composition of surveillance zones . . . . . . . . . . . . . . . . . . . . . . 110
6.2 Characteristics of surveillance zones . . . . . . . . . . . . . . . . . . . . 115
6.3 Stock numbers by surveillance zone . . . . . . . . . . . . . . . . . . . . . 115
6.4 Seaports: stock numbers by incursion zone . . . . . . . . . . . . . . . . . 116
6.5 International airports: stock numbers by incursion zone . . . . . . . . . 117
xii
6.6 Seaports 2003: vessel arrivals and international cargo weights . . . . . . 117
6.7 Airports 2003: international aircraft arrivals . . . . . . . . . . . . . . . . 118
6.8 Descriptive statistics for actual and simulated flock sizes . . . . . . . . . 125
6.9 Within-flock incidence risk of abortion . . . . . . . . . . . . . . . . . . . 126
6.10 Number of flocks with high estimated abortion incidence risk . . . . . . 127
6.11 Number of flocks to be tested to detect disease . . . . . . . . . . . . . . 127
6.12 Maximum number of infected flocks . . . . . . . . . . . . . . . . . . . . 128
6.13 Number of flocks required to be tested to detect disease . . . . . . . . . 136
6.14 Seaports: total number of animals to test to detect disease . . . . . . . . 139
6.15 Airports: total number of stock to test to detect disease . . . . . . . . . 139
7.1 The number of surveillance tests required to detect one positive flock . . 148
7.2 The simulated mean risk score for each surveillance area . . . . . . . . . 149
xiii
List of Figures
3.1 The distribution of exiting uninfected animals . . . . . . . . . . . . . . . 41
3.2 The distribution of exiting BSE-infected animals . . . . . . . . . . . . . 42
3.3 Infection status of BSE-infected cattle . . . . . . . . . . . . . . . . . . . 42
3.4 Predicted true BSE prevalences for each birth cohort . . . . . . . . . . . 43
4.1 Proportion of each age group tested in the casualty slaughter stream . . 54
4.2 Points associated with the value of testing an animal . . . . . . . . . . . 55
4.3 Calculated number of points required . . . . . . . . . . . . . . . . . . . . 55
5.1 95% UCL for the prevalence of Trichinella in low-risk grower pigs . . . . 73
5.2 Sampling effort for foxes tested . . . . . . . . . . . . . . . . . . . . . . . 79
5.3 Sampling effort for horses . . . . . . . . . . . . . . . . . . . . . . . . . . 80
5.4 Sampling effort for low-risk grower pigs . . . . . . . . . . . . . . . . . . 81
5.5 Sampling effort for high-risk grower pigs . . . . . . . . . . . . . . . . . . 82
5.6 Sampling effort for low-risk breeder pigs . . . . . . . . . . . . . . . . . . 83
5.7 Sampling effort for high-risk breeder pigs . . . . . . . . . . . . . . . . . 84
5.8 Upper 95% confidence limit for foxes (unweighted) . . . . . . . . . . . . 85
5.9 Upper 95% confidence limit for horses (unweighted) . . . . . . . . . . . 86
5.10 Upper 95% confidence limit for low-risk grower pigs (unweighted) . . . . 87
5.11 Upper 95% confidence limit for high-risk grower pigs (unweighted) . . . 88
5.12 Upper 95% confidence limit for low-risk breeder pigs (unweighted) . . . 89
5.13 Upper 95% confidence limit for high-risk breeder pigs (unweighted) . . . 90
xiv
5.14 Upper 95% confidence limit for foxes . . . . . . . . . . . . . . . . . . . . 91
5.15 Upper 95% confidence limit for horses . . . . . . . . . . . . . . . . . . . 92
5.16 Upper 95% confidence limit for high-risk grower pigs . . . . . . . . . . . 93
5.17 Upper 95% confidence limit for low-risk breeder pigs . . . . . . . . . . . 94
5.18 Upper 95% confidence limit for high-risk breeder pigs . . . . . . . . . . . 95
6.1 Simulation of abortion incidence risk . . . . . . . . . . . . . . . . . . . . 112
6.2 Surveillance zones in New Zealand and breeding ewe density . . . . . . . 114
6.3 The main incursion zones within New Zealand . . . . . . . . . . . . . . 116
6.4 Ewe population size compared to abortion submissions . . . . . . . . . . 118
6.5 Total number of abortion submissions to each laboratory . . . . . . . . . 119
6.6 Number of surveillance cases . . . . . . . . . . . . . . . . . . . . . . . . 120
6.7 Total number of abortion case submissions . . . . . . . . . . . . . . . . . 120
6.8 Diagnosis of laboratory submissions . . . . . . . . . . . . . . . . . . . . 121
6.9 Diagnoses by surveillance zone in 2003 . . . . . . . . . . . . . . . . . . . 122
6.10 Submissions suitable for use as surveillance cases . . . . . . . . . . . . . 122
6.11 Diagnosis of surveillance cases . . . . . . . . . . . . . . . . . . . . . . . . 123
6.12 Proportion of cases undiagnosed at each laboratory . . . . . . . . . . . . 123
6.13 Simulated within-flock abortion incidence risk . . . . . . . . . . . . . . . 124
6.14 Histogram of simulated flock sizes for Southland . . . . . . . . . . . . . 125
6.15 Simulated incidence risk of abortions in Southland flocks . . . . . . . . . 126
6.16 Areas where disease conditions were met in January . . . . . . . . . . . 129
6.17 Areas where disease conditions were met in April . . . . . . . . . . . . . 130
6.18 Areas where disease conditions were met in July . . . . . . . . . . . . . 131
6.19 Areas where disease conditions were met in October . . . . . . . . . . . 132
6.20 Number of cases submitted from each surveillance zone in 2003 . . . . . 136
7.1 Median risk score and interquartile range of risk scores for each disease . 148
7.2 Number of surveillance tests to be carried out in each SA by month . . 150
xv
7.3 The number of tests allocated to each SA . . . . . . . . . . . . . . . . . 151
7.4 The maximum prevalence of diseased ewe flocks to be missed . . . . . . 152
xvi