CABI is a trading name of CAB International CABI CABI Nosworthy Way 745 Atlantic Avenue Wallingford 8th Floor Oxfordshire OX10 8DE Boston, MA 02111 UK USA Tel: +44 (0)1491 832111 Tel: +1 (617)682-9015 Fax: +44 (0)1491 833508 E-mail: cabi-nao@cabi.org E-mail: info@cabi.org Website: www.cabi.org © Copyright International Livestock Research Institute and CAB International 2020. Co-published by CABI and the International Livestock Research Institute 2020. This publication is copyrighted by the International Livestock Research Institute (ILRI). It is licensed for use under the Creative Commons Attribution 4.0 Inter- national Licence. To view this licence, visit https://creativecommons.org/ licenses/by/4.0. Unless otherwise noted, you are free to share (copy and redistrib- ute the material in any medium or format), adapt (remix, transform, and build upon the material) for any purpose, even commercially, under the following conditions: ATTRIBUTION. The work must be attributed, but not in any way that suggests endorsement by ILRI or the author(s). NOTICE: For any reuse or distribution, the licence terms of this work must be made clear to others. Any of the above conditions can be waived if permission is obtained from the copyright holder. Nothing in this licence impairs or restricts the author’s moral rights. Fair dealing and other rights are in no way affected by the above. The parts used must not misrepresent the meaning of the publication. ILRI would appreciate being sent a copy of any materials in which text, photos etc. have been used. ISBN-13: 978 1 78924 185 3 (hardback) ILRI ISBN: 92-9146-586-3 (hardback) CABI Commissioning editor: Alexandra Lainsbury CABI Editorial assistant: Lauren Davies CABI Production editor: James Bishop Typeset by SPi, Pondicherry, India Printed and bound in the UK by Bell and Bain Ltd, Glasgow 5 Veterinary Epidemiology at ILRAD and ILRI, 1987–2018 Brian Perry1, Bernard Bett2, Eric Fèvre2, Delia Grace3 and Thomas Fitz Randolph2 1University of Oxford and University of Edinburgh, UK; 2International Livestock Research Institute, Nairobi, Kenya; 3International Livestock Research Institute, Nairobi, Kenya and University of Greenwich, UK Contents Executive Summary 209 The problem 209 ILRI’s role in the global context 209 Impact of ILRI’s research 210 Scientific impacts 210 Economic impact assessment 210 Developmental impacts 210 Capacity development 210 Partnerships 210 Impacts on human resources capacity in veterinary epidemiology 211 Impacts on national animal health departments and services 211 Impacts on animal health constraints in developing countries 211 Impacts on ILRI’s research and strategy 211 Introduction 211 The introduction of veterinary epidemiology and economics at ILRAD 212 Field studies in Kenya 212 Tick-borne disease dynamics in eastern and southern Africa 214 The heartwater studies in Zimbabwe 215 Economic impact assessments of tick-borne diseases 216 Tick and tick-borne disease distribution modelling 217 Modelling the infection dynamics of vector-borne diseases 218 Impacts of Trypanosomiasis and its Control 218 Economic impact of trypanosomiasis 218 The epidemiology of resistance to trypanocides 219 The development of a modelling technique for evaluating control options 219 Sustainable trypanosomiasis control in Uganda 220 Sustainable trypanosomiasis control in the Ghibe Valley of Ethiopia 220 Spatial modelling of tsetse distributions 220 Preventing and containing trypanocide resistance in the cotton zone of West Africa 220 Rabies Research: A Networking and Capacity-building Role in Africa 221 © International Livestock Research Institute 2020. The Impact of the International 208 Livestock Research Institute (eds J. McIntire and D. Grace) Veterinary Epidemiology at ILRAD and ILRI, 1987–2018 209 The Economic Impacts of Rinderpest Control 222 Applying Economic Impact Assessment Tools to Foot-and-mouth disease Control 223 The southern Africa FMD economic impact study 224 Economic Impacts of FMD in Peru, Colombia and India 224 Economic impacts of FMD control in endemic settings in low- and middle-income countries 224 The Global Foot-and-Mouth Disease Research Alliance (GFRA) 224 Rift Valley Fever 225 Economic impact assessment of control options and calculation of disability-adjusted life years (DALYs) 226 RVF risk maps for eastern Africa 226 Land-use change and RVF infection and disease dynamics 226 Epidemiology of Gastrointestinal Parasites 227 Priorities in Animal Health Research for Poverty Reduction 227 The Wellcome Trust Epidemiology Initiatives 228 The Broader Economic Impact Contributions 228 The Responses to Highly Pathogenic Avian Influenza 229 The ISVEE Experience 230 The Role of Epidemiology in ILRAD and ILRI 230 The Impacts of ILRAD and ILRI’s Epidemiology 231 Capacity development in veterinary epidemiology and impact assessment 231 Impacts on national animal health departments and services 231 Impacts on animal health constraints in developing countries 231 Impacts on ILRI’s research and strategy 231 References 232 Executive Summary to national to global. Furthermore, veterinary epidemiological and economic impact sciences The problem are key components in a number of the global grand challenges relating to disease control, The effectiveness of detecting and controlling climate change and food security. animal diseases is dependent on a solid under- standing of their dynamics and impacts through scientifically sound qualitative and quantita- ILRI’s role in the global context tive methods by trained personnel. Veterinary epidemiology is the systematic characterization The International Livestock Research Institute and explanation of patterns of animal diseases (ILRI) and its predecessor, the International and the use of this information in the reso- Laboratory for Research on Animal Diseases lution of animal and human health problems. (ILRAD), have played an important international This discipline exploits an increasing inventory role in identifying and developing epidemio- of tools for effective data gathering, assembly logical tools for the investigation and resolution and analysis, modelling and reporting, all of animal health constraints to livestock pro- targeted at decision making by producers, duction. When the group was formed in 1987, it governments and international development was charged with addressing the two diseases agencies. Furthermore, the integration of considered the priority in Africa, namely East epidemiology with agricultural economics Coast fever (ECF) and trypanosomiasis, to sup- and other social sciences provides a uniquely port the development of vaccines against these effective tool for evaluating disease as a con- two diseases, which was the mandate of ILRAD. straint to broader development agendas, for Following the transition from ILRAD to ILRI, assessing the absolute and relative economic the institution has been a leader in exploring importance of diseases, and for evaluating the new epidemiological approaches, and in widen- costs and benefits of alternative intervention ing the disciplinary spectrum of epidemiological options, at different levels ranging from farm investigations. However, arguably most important 210 B. Perry et al. of all, ILRI has played a facilitating role in collab- around the world and has contributed substan- orating with countries, institutions and organ- tially to research and development networking izations in Africa, Asia and Latin America to re- in many areas of animal health, including tick- spond to requests for both short- and long-t erm borne diseases, rabies, and the fundamental tools partnerships and support at international, re- of epidemiology and socio-economic impacts. gional, national and local levels, and in exten- sive capacity building in epidemiological tools, Scientific impacts techniques and approaches. Initially, ILRAD and ILRI focused almost The group brought the science of structured exclusively on the dynamics and impact of tick- epidemiological analysis to the institute in meth- borne pathogens of livestock in Africa, but during odological terms (particularly in observational the 1990s, the geographical focus, disciplinary field studies), and also expanded the disciplinary make-up and range of tools used broadened sub- contributions to include social, economic and stantially, tackling multiple diseases in Africa, environmental considerations in the analysis Asia and Latin America and building capacity in of disease impacts, and the impacts of disease epidemiological and economic impact assess- control interventions. ment techniques. For a period of 15 years (1987–2002) ILRAD/ Economic impact assessment ILRI’s epidemiology and socio-economic impact assessment capacity was assembled in one team Most of the earlier economic impact assessments serving a range of institutional and externally carried out by ILRAD and ILRI were disease commissioned needs and was recognized inter- specific, such as on ECF, rinderpest and FMD, nationally for its focus of health issues affecting and built on evidence derived from underlying development and poverty reduction. Through a epidemiological data. Economic impact assess- major study of animal health research priorities ments gained increasing momentum as ILRI’s commissioned by UK Department for International mandate broadened. The epidemiology group Development (DFID), the team made a substantial concluded that no longer should studies of the contribution to the design of ILRI’s new strategy, economics of diseases of production animals be which emerged in 2002, and to the poverty- limited to animal scientists seeking the collabor- focused agendas of other organizations such as ation of agricultural economists to affix prices the Food and Agriculture Organization of the to estimated productivity losses, and the new United Nations (FAO) and the Wellcome Trust. discipline of animal health economics emerged in which the quality of economic evaluations de- pended on integrating the products of good epi- Impact of ILRI’s research demiological studies into economic frameworks. In 1998, the World Organisation for Ani- Veterinary epidemiological and economic impact mal Health (Office International des Epizooties, sciences at ILRAD and ILRI have left a valuable or OIE) approached ILRI to compile and edit a legacy of publications in peer-reviewed journals, special edition of the OIE Scientific and Technical strategic reports and policy documents, as well Review on Animal Health Economics, which com- as methodologies and approaches that have prised nine chapters on various topics such as the been applied in virtually all corners of the world, demands of economic impact knowledge, and and many trained epidemiologists now serving seven case study chapters addressing specific different institutional needs in Africa, Asia, Latin diseases and different scenarios affecting the America, Europe and Australia. validity of the economic studies. The range of diseases subject to epidemio- Developmental impacts logical and economics research has been wide, and included the viral infections rinderpest, foot- While we cannot ascribe higher productivity to and-mouth disease (FMD), Rift Valley fever (RVF) ILRI’s veterinary epidemiology research per se, and highly pathogenic avian influenza (HPAI), the research has increased our understanding and the context of impact assessments has been of infection dynamics, which has influenced applied to trade, livelihoods and poverty reduction. disease control and the role of interventions in ILRI has worked with a multitude of partners different settings. Veterinary Epidemiology at ILRAD and ILRI, 1987–2018 211 Capacity development the preparedness and responses to RVF in east- ern Africa, to a greater understanding of the Over the years, epidemiology has gone through economic impact of rinderpest in Africa and of both administrative and locality changes, seen FMD in Africa, Asia and Latin America, and to as a unified entity for 15 years, and later under regional understanding of the drivers of rabies the new ILRI strategy it was fragmented and control. More recently, epidemiology research at diminished. However, during the 15-year period ILRI has contributed substantially to our under- from 1987 to 2002, the institution had sup- standing of food safety risks in formal and ported approximately 15 MSc students and 37 informal markets and to the dynamics and risks PhD students with approximately 50 scientists of zoonotic diseases. The research has also con- predominantly from African countries. In add- tributed to the global understanding of the im- ition, there have been several postdoctoral fel- portance of these and other diseases to African lows who have been trained. As most of the epi- livestock systems and to the particular animal demiology is conducted under a specific disease, health constraints facing the poorer sectors of it is difficult to determine how many more have Africa’s livestock-engaged communities. been trained to date. Partnerships Impacts on ILRI’s research and strategy During the days of ILRAD, the epidemiology ILRI has worked with a multitude of partners and socio-economic programme had little or around the world and contributed substan- no impact on ILRAD’s research and strategy; tially to research and development networking rather, it was seen as providing evidence justi- in many areas of animal health, including fying the existence of the laboratory-based tick-borne diseases, rabies and the fundamen- vaccine research for the two target haemopara- tal tools of epidemiology and socio-economic sitic diseases. Nevertheless, after ILRI’s birth in impacts. In 2007, the Participatory Epidemi- 1995, the programme played an important role ology Network for Animal and Public Health in providing impact assessment services, which (PENAPH) was developed to connect groups progressively enhanced the engagement of the and individuals who apply participatory epi- institution with different national, regional and demiology in controlling emerging and exist- donor clients. This situation changed dramat- ing diseases. ically in 2002 following the publication of the Impacts on human resources capacity in DFID-commissioned study on animal research veterinary epidemiology priorities for poverty reduction. The matrix of three ‘pathways out of poverty’ provided the During the 15-year period from 1987 to 2002 a framework of the new institutional thematic substantial number of MSc and PhD students structure, not just for animal health research were trained through incorporation into ILRI’s but also for ILRI’s entire programme. research activities; these were predominantly from African countries. Introduction Impacts on national animal health departments and services Veterinary epidemiology is the systematic char- The epidemiology group provided a role model acterization and explanation of patterns of animal of investigative problem solving, which was diseases and, importantly, the use of this infor- picked up, copied and adopted by some institu- mation in the resolution of animal and human tions in African countries. health problems. It is a subject that exploits an increasing inventory of tools for effective data Impacts on animal health constraints in gathering, assembly and analysis, targeted at de- developing countries cision making in the field of animal disease con- trol and sustainable livestock enterprise develop- ILRI’s epidemiology research has made sub- ment. The integration of epidemiology with stantial contributions to our understanding and agricultural economics and other social sciences control of ECF and trypanosomiasis in Africa, to provides a uniquely effective tool for evaluating 212 B. Perry et al. disease as a constraint to broader development quently extended for a further 2 years) supported agendas, for assessing the absolute and relative what became the Epidemiology and Socioeco- economic importance of diseases, and for evalu- nomics Programme. The group was joined in 1992 ating the costs and benefits of alternative inter- by an ecologist to explore the environmental vention options, at different levels ranging from impacts of trypanosomiasis control (Reid et al., farm to national to global. ILRI and its predecessor, 1995). The team expanded in the late 1990s to ILRAD, have played an important international include another staff epidemiologist and two role in exploiting epidemiological tools for the postdoctoral epidemiologists. investigation and resolution of animal health The group rapidly laid out a work plan, constraints to livestock production and poverty beginning with the establishment of databases reduction in many regions of the developing on African production systems at risk from the world. Furthermore, ILRI has been a leader in two diseases and on methodologies for determining exploring new epidemiological approaches, their impact. This challenge led to the realization and in widening the disciplinary spectrum of that disease incidence and prevalence data in Af- epidemiological investigations. However, argu- rica were scarce and unreliable, notably on the ably most important of all, ILRI has played a fa- structure and ownership of the livestock popula- cilitating role in collaborating with countries, tions at risk. The need for structured quantita- institutions and organizations in Africa, Asia tive epidemiology capacity emerged, which led and Latin America in response to requests for both to a sustained programme of data assembly, digi- short- and long-term partnership and support at tal data documentation and assembly, the devel- international, regional, national and local levels, opment of modelling techniques and, of course, and in extensive capacity building in epidemio- the gathering of field data. logical tools, techniques and approaches. Field studies in Kenya The introduction of veterinary epidemiology and economics at ILRAD The first field site of diverse ecosystems and dis- ease impacts was in Kilifi District on the Kenyan The fundamental belief at the creation of ILRAD coast. A collaborative programme between the in 1974 was that vaccines against ECF and tryp- International Livestock Centre for Africa (ILCA) anosomiasis were the mandate of ILRAD. The and the Kenya Agricultural Research Institute evidence available to answer these questions at (KARI) was established in 1988 at KARI’s Mtwapa the time was derived almost entirely from African Regional Research Centre, near Mombasa. veterinary services and diagnostic laboratories, The smallholder dairy group was setting up which had, for the previous 60 or so years, been a broad study on the constraints to smallholder servicing livestock production enterprises of the milk production in the coastal lowlands of colonial powers. There were little if any economic Kenya and how extension services covering the data to quantify the impacts of these diseases, areas of feed and health could be improved. Spe- even in commercial systems, and the potential cifically, the study estimated the demand for milk returns from vaccines. and dairy products, identified technical and pol- ILRAD alone among the international agri- icy constraints on production in mixed small- cultural research centres of CGIAR had a unique holder farming systems, evaluated dairy cattle mandate to carry out basic research and as such breed resources, estimated disease risk to dairy undertook very little of the technology transfer cattle and tested disease control methods, and functions. Pressure progressively increased from developed feeding systems appropriate to small- partners to quantify the impacts of these two holder dairy production systems. diseases on African agriculture in order to better The ILRAD epidemiology team provided justify the scientific investment in the study of support to the studies on the epidemiology and these two diseases. impact of ECF in the form of design and analysis Thus, it was some 13 years after the estab- of studies led by KARI staff. The challenge was lishment of ILRAD that veterinary epidemiology a total lack of data on ECF occurrence, and so a was introduced into the institute. The Rockefeller series of cross-sectional studies was set up. Key was Foundation initially for 3  years (and subse- understanding the link between infection preva- Veterinary Epidemiology at ILRAD and ILRI, 1987–2018 213 lence, as measured by antibodies to Theileria par- first to Kiambu District, with a 1-year study of va in an indirect fluorescent antibody test, and the dynamics of theileriosis (O’Callaghan et al., disease incidence. ILRAD’s entry into this re- 1998). This was followed in 1994 by investiga- search area was at a time when infection preva- tions in the neighbouring Muranga District, lence, as measured by antibody prevalence, had which hosted a range of livestock production not been correlated with disease incidence, and, systems in diverse AEZs. There were five distinct where prevalence studies had been undertaken, AEZs within Muranga District, giving the oppor- they had often been reported on the basis of ad- tunity to investigate the influence of a range of ministrative boundaries such as the FAO’s 1975 climate, livestock breeds and farming practice study in Kenya (Kariuki, 1988). variables on ECF dynamics. The work started In 1983, the Farm Management Handbook of with a cross-sectional serological study on 750 Kenya was published (Jaetzold and Schmidt, smallholder dairy farms in Muranga District, se- 1983), which provided a unique landscape syn- lected in a stratified random sample (Gitau et al., thesis of Kenya’s agricultural environment, ag- 1997), which showed the markedly different gregating a number of variables into a kaleido- prevalences of T. parva infection across AEZs. scope of colours representing the suitability for The area was typical of the highland areas of different crops and agricultural enterprises. With eastern Africa in which the process of smallholder the knowledge that the epidemiology and impacts dairy intensification was gaining momentum. of ECF were highly dependent on environmental The investigation continued with a study that suitability for the main vector tick, Rhipicephalus related prevalence with incidence, case morbid- appendiculatus (the brown ear tick), the zone ity and case mortality (Gitau et al., 1999), and boundaries provided a new and useful sampling with a study of how these infections affected frame that had previously been unexploited. weight gain in calves (Gitau et al., 2001). The A series of studies was set up in coastal Kenya work concluded with a synthesis of the implica- to determine the prevalence and incidence of tions of the research on disease risk and on the ECF and the other tick-borne infections such as potential role of vaccination against ECF (Gitau anaplasmosis and babesiosis, and to evaluate et al., 2000). the role of immunization against ECF using the The synthesis concluded that ECF risk is infection-and-treatment method (ITM). The low in predominately zero-grazing areas. Thus, studies provided an initial quantitative assess- tick control or future vaccination programmes ment of antibody prevalence to the spectrum of will probably only be used by very risk-averse tick-borne disease parasites in order to assess the farmers who wish to protect their highly valu- epidemiological status of these infections in both able cows from the low risk of ECF mortality. In indigenous Zebu cattle kept, and in improved dairy contrast, for open grazing systems, particularly cattle in three different agroecological zones (AEZs) in the lower- elevation upper-midland 4 (UM4) (Deem et al., 1993; Maloo et al., 2001a,b,c). zone, the risk of ECF is much greater and prob- The coast work provided an opportunity to ably much more variable. In this system, there engage at the front line with national partners will be much more substantial direct impact of and to explore impact study design; it also illus- ECF control programmes. In areas where ECF trated the need to disaggregate factors affecting control will be through vaccination, irrespect- ECF epidemiology and impact. However, as the ive of the grazing management system, there coastal systems were not fully representative of will be a greater likelihood of the development the intensifying livestock systems in the temper- of endemic stability. Increased vaccination ate highland areas of eastern Africa, additional coverage to enhance the development of herd studies were set up. The first was in Uasin Gishu, immunity, combined with modification of where larger-scale dairy and beef production acaricide control strategies to allow sufficient was rapidly being replaced by small-scale com- challenge, seemed to offer the best prospect for mercial dairy enterprises (Mukhebi et al., 1992a). establishing endemic stability. It was quite clear This work coincided with the introduction of the from these studies that attention needs to be discipline of human nutrition into the impact paid to the variation in ECF risk, both spatially equation (Curry et al., 1996). (as ECF risk changes over relatively short In 1992, the focus of the ECF epidemiology geographical distances) and temporally (sea- studies moved to the central highlands of Kenya, sonally), to develop optimal combinations of 214 B. Perry et al. control measures for ECF under different eco- which all exert their influence as a gradient (or logical and grazing situations. cline) of effects: It was in 1997 that the ECF epidemiology work expanded into other parts of Kenya, follow- • The ecological cline, in which the climatic ing the submission of a research proposal to the suitability for the tick vector varies with International Fund for Agricultural Develop- rainfall and altitude; the ecological cline ment (IFAD). The revised proposal strengthened gradient can be affected by differences in the epidemiology and impact assessment com- vegetation cover. ponents and placed them in an ex ante context. • The host genetic cline, in which purebred Vaccine efficacy trials were limited to two sites in taurine cattle bred under tick-free condi- Kenya, while the impact assessment broadened tions are highly susceptible to disease, and into new areas. The impact assessment studies taurine cattle bred in tick-borne infection included an evaluation of mechanisms for optimal endemic areas and some Zebu breeds (such delivery, adoption and impact of the p67 vaccine as Boran) bred in tick-free conditions are (p67 is the major surface protein of T. parva moderately susceptible, but Zebu cattle bred sporozoites), determining the impact of a recom- in tick-borne infection endemic areas are of binant vaccine on a series of productivity and low susceptibility to disease. economic indicators in smallholder dairy systems. • The feeding management cline, which con- The project also included key laboratory studies to trols the exposure of hosts to the ecological support the field studies and disease-modelling conditions; this can range from no influence, work (Ochanda et al., 1998). where cattle are herded on natural pasture, Table 5.1 outlines the various studies under- to complete influence, where cattle are kept taken in Kenya, illustrating the differences in on concrete and fed on cultivated forage impacts by region, AEZ and grazing management. grasses, as in the smallholder zero-grazing There were a wide variety of products units of eastern Africa. emerging from the IFAD study, ranging from • The tick control cline, where tick control Technical Advisory Notes such as ‘Assessing ranges from highly effective, regular appli- farmer preferences for the provision of livestock cation through to no tick control at all. health services’ to international presentations (Leneman et al., 2000; Kiara et al., 2000, 2003; Ndung’u et al., 2000, 2003; Wanyangu et al., 2000; Di Giulio et al., 2003, Karimi et al., 2003; Tick-borne disease dynamics in eastern O’Callaghan et al., 2003; Diaz et al., 2003) to and southern Africa peer-reviewed papers in scientific journals. In a review article of the ECF work (Perry and At the start of these intense epidemiological Young, 1995), it was postulated that the degree investigations in Kenya, a regional meeting of mortality and production losses from T. parva on tick-borne diseases was held in Lilongwe, infections were dependent on four key factors, Malawi, in 1988 (Dolan, 1989), which provided Table 5.1. ECF risks by region, AEZ and grazing management. Study O’Callaghan (1998) Gitau (1998) Maloo (1993) Calves (<1 year of age); Calves (≤6 months of age); Calves ( <1 year of age); Incidence rate type incidence density cumulative incidence incidence density Grazing management Zero grazing Pasture Zero grazing Pasture Zero grazing Pasture Number of animals (n) 93 108 134 91 38 50 ECF morbidity rate (%) 5.5 10.9 11.8 49 36.4 68.8 ECF mortality rate (%) 0 2.2 1.7 20.6 20.8 49.7 Case-fatality proportion (%) 0 25 28.6 38.1 57.1 72.2 Seroconversion rate (%) 41.4 56.5 58.4 74 0 0 Morbidity proportion (%) 13.3 7.7 12.2 36.1 0 0 Veterinary Epidemiology at ILRAD and ILRI, 1987–2018 215 an opportunity for sharing of information and Development Community (SADC), the Univer- understanding on theileriosis throughout the sities of Florida (USA) and Warwick (UK), and eastern and southern African regions, and where ILRI determined and quantified the infection dy- it became apparent that there were significant namics of heartwater in the major production differences in the disease epidemiology between systems and AEZs of Zimbabwe, the economic the eastern and southern regions. The Lilongwe impact of the disease, and the technical and eco- meeting led to a textbook on theileriosis, The Epi- nomic viability of different control interven- demiology of Theileriosis in Africa (Norval et al., tions, with particular emphasis on the role of 1992b). The book was published by Academic inactivated vaccines. The project outputs were Press, with ILRAD supporting the time contribu- as follows: tions of the three authors and the preparation of camera-ready copy, and the book became the • Distributions of tick vectors in Zimbabwe reference point for all those working on control were defined and documented. A national of the disease. This book remains the only text- survey of 3000 collections determined that book on theileriosis, a product of ILRAD’s epi- Amblyomma hebraeum is the dominant tick demiology team of substantial impact. in the south and that it had spread into cen- The disease called ECF was said to have tral and eastern areas of the high veld. The been eradicated in southern Africa as a result survey also found that Amblyomma variega- of an intensive dipping programme, with the tum is present mainly in the north-west but last case occurring in Swaziland in 1960. that it is also found in central and eastern Nevertheless, T. parva infections persist, but with parts of the high veld, with some overlap the official eradication declared, the names of of the two species (Norval et al., 1994; Peter theileriosis and corridor disease have been used et al., 1998, 1999). to describe disease outbreaks. This curious con- • Spread of heartwater was documented and frontation of science and officialdom of disease quantified, and factors affecting the spread and parasite nomenclature was reviewed in an were determined. A. hebraeum had spread article entitled ‘The naming game: the changing far north due largely to movement of cattle fortunes of East Coast fever and Theileria parva’ (and some wildlife) to the high veld. A grad- (Perry and Young, 1993). It is encouraging ual reduction in acaricide use, particularly how these studies have led to our current in the communal lands, contributed to the understanding of theileriosis epidemiology, as expanding distribution of this tick (Norval illustrated by a recent review by Gachohi et al. et al., 1992a). (2012). • Infection dynamics in the tick vector and mammalian hosts were determined and quan- tified. Endemic stability, in which population The heartwater studies in Zimbabwe immunity develops, was found to be wide- spread but not present where acaricides were Through funding from the US Agency for used intensively to interrupt natural infec- International Development (USAID), a 5-year tion. These results suggest that use of inacti- project on the epidemiology and impact of an- vated vaccines in many circumstances will other important tick-borne disease of livestock allow a reduction in acaricide use with a in Africa, heartwater (also known as ehrlichio- transition to endemic stability and subse- sis; caused by Ehrlichia ruminantium infection, quent natural infection boosting the vac- formerly Cowdria ruminantium) was initiated in cinal immunity. 19941. This work subsequently received ILRI’s • The impact of endemic stability, and carrier award (and ILRI’s nomination for the CGIAR infections, on sheep productivity was deter- Chairman’s Award) for Scientific Partner- mined. Studies in sheep revealed that creating ship, 2000. endemic stability artificially with vaccines The collaborative research in epidemiology does not harm the health and reproductive and economics between the Veterinary Research performance of breeding ewes or the growth Laboratory (VRL) in Harare, the Heartwater and milk consumption of their lambs (Mar- Research Project of the Southern African tinez et al., 1999a,b). 216 B. Perry et al. • Infection dynamics models were devel- The successes of this collaborative project oped using data generated by the research. were considerable, with all objectives met and all A mathematical model of the infection dynam- findings published within a period of 5  years. ics of the heartwater pathogen, E. ruminan- The results of the project have given scientists a tium, showed that endemic stability is due sound understanding of the factors influencing principally to the protection of calves and the distribution of tick vectors, and the infection lambs against disease by innate or maternally dynamics and impacts of the disease in different derived factors (O’Callaghan et al., 1998). AEZs and production systems, and predictions • Economics of livestock production in heart- of the technical and economic impacts of control water areas was determined. Both large- with a new generation of inactivated vaccines now and small-scale livestock production could emerging. The results generated have strong im- be increased significantly with more, and plications for heartwater control in other coun- more cost-effective, heartwater control tries of Africa. Design and modelling features of methods (Perry et al., 1998; Chamboko the study have been used in studies of other tick- et al., 1999). borne livestock diseases. • The economic impact of heartwater, and of Apart from its technical achievements, the future vaccine use, was determined. The project boosted scientific capacity, particularly annual total direct losses in Zimbabwe (acari- in Zimbabwe, through postgraduate training for cide costs, milk losses, treatment costs) from national scientists. Project members produced heartwater were estimated to be US$5.6 mil- 23 papers in peer-reviewed journals, 15 of which lion. A new inactivated vaccine was predicted were authored by VRL scientists. Project staff to have a benefit:cost ratio of 2.4:1 in the produced another 36 publications, including 22 communal sectors and 7.6:1 in the com- presentations at scientific meetings and ten mercial sectors (Mukhebi et al., 1999). articles in the project newsletter. • The efficacy of future vaccine use was evaluated in epidemiological models. The timing of vaccination and frequency of Economic impact assessments revaccination were shown to have greater of tick-borne diseases effects on population protection than vac- cine efficacy. In the face of an epidemic, the The first opportunity to undertake an economic frequency of administration is critical to a impact assessment came in the late 1980s with vaccine’s success. Vaccines of relatively the ILRAD/KARI partnership on the Kenya low efficacy (about 50%) can significantly coast, where an immunization trial was being reduce livestock morbidity and mortality if carried out using ITM to control ECF. Adrian administered with appropriate frequency Mukhebi showed greater profitability in immun- (O’Callaghan et al., 1999). ized cattle compared with unimmunized, through • The economic impact of the disease and of lower mortality and higher weight gains (Muk- its control through vaccines was evaluated hebi et al., 1989). This was a solid first piece of and quantified in the countries of the SADC evidence, although it was compiled on a state-run region. In total, 31 million cattle and 28 Agricultural Development Corporation beef ranch. million small ruminants were found to be at The authors commented that ‘these results and risk of heartwater in the nine SADC coun- the recommendation apply to one immunization tries affected: Angola, Botswana, Malawi, trial on one farm which was under an atypical Mozambique, South Africa, Swaziland, Tan- management system for the region’. zania, Zambia and Zimbabwe. The total annual Building on the potential for ITM, Mukhebi losses were estimated at US$47.6 million, of et al. (1990) then dissected the complicated vac- which 61% were production losses and 39% cine preparation process and calculated the costs control costs. New inactivated heartwater of establishing a vaccine production facility (at a vaccines could yield benefit:cost ratios of up hypothetical site in Kenya but using a method- to 4.4:1, particularly in commercial and ology generic to other countries). emerging market-orientated systems of the These initial forays into the economics region (Minjauw et al., 1998, 2000). of ECF and its control led to an Africa-wide Veterinary Epidemiology at ILRAD and ILRI, 1987–2018 217 assessment, presenting the estimated total costs assessments, running the software ARC/INFO of ECF in affected countries in 1989 as US$168 from the Environmental Systems Research Insti- million (Mukhebi et al., 1992b). The authors tute (ESRI). A memorandum of understanding calculated benefit:cost ratios for the control of was set up between ILRAD and UNEP, and two ECF through vaccination of between 8.9 and UNEP scientists contributed data assembly and 16.8, depending on the intensity of post- analysis time over the following 18  months, vaccination acaricide use. The authors warned before the epidemiology and socio-economics that ‘the input values and hence results pre- group at ILRAD obtained core funds to establish sented in this paper are dependent upon sparse its own GIS capacity, led by Russ Kruska. Research and inadequate data and are largely illustra- in the late 1980s and early 1990s centred on tive of the methodology and data needs. Never- vector-borne disease distribution and impact theless, they provide an estimated magnitude studies; the group also contributed to the estab- of the economic losses attributed to theilerio- lishment in 1992 of the UNEP/CGIAR partner- sis, and the economics of its control by the ship on the development of digital datasets for infection and treatment method in the infected research on a wide range of topics including region’. natural resources, ecology, environment and socio- A series of further economic studies of economic factors. ECF was undertaken by PhD student Hezron Efforts to model the potential distribution of Nyangito, who, in a partnership with the R. appendiculatus were assisted by two key inputs. D epartment of Agricultural Economics at The first was the database on tick field samplings Texas A&M University, used a whole-farm assembled by Jane Walker. The second was a cli- simulation model to estimate the financial mate matching model, Climex, developed by and economic pay-offs from the use of ITM Robert Sutherst, with parameters for the condi- vaccination, drawing on data collected from tions favoured by the brown ear tick (among Uasin Gishu, Kenya (Nyangito et al., 1994, others). Instead of the model being run on cli- 1995, 1996). mate data for any given location, it was run for The study by Mukhebi et al. (1999) pro- the whole of Africa, in each of the 25 km2 pixel vided one of the first attempts to truly inte- cells of an interpolated climate surface for the grate epidemiology and economics models to continent. The results plotted the potential dis- predict future economic impacts of different tribution of R. appendiculatus (Lessard et al., control scenarios under a set of epidemiological 1990; Perry et al., 1990, 1991), but when scenarios. compared with the database of Jane Walker and others of where the tick had been recorded, the potential distribution exceeded the historical records, suggesting that Climex did not tell the Tick and tick-borne disease distribution whole story. This finding stimulated interest modelling in other predictive modelling approaches to estimate tick distribution (Randolph, 1993), in The need to understand the geographical scale which much closer attention was paid to the of impact of both ECF and trypanosomiasis very climatic requirements of all three instars of the quickly led into the area of modelling, primarily tick, which eventually led to a more biologically of the vectors but also to a degree of the disease sound spatial prediction platform (Randolph itself. In the absence of high-quality field data on and Rogers, 1997). R. appendiculatus distribution, the group first In a follow-up case study of predicting sought data on the key drivers of climate and outbreaks of theileriosis in Zimbabwe using mul- vegetation, and in late 1987 and early 1988, a tiple climatic variables (Duchateau et al., 1997), contract was drawn up between ILRAD and the the methodology for assessing distribution Global Resource Information Database (GRID) drivers using climate databases was addressed. group led by Harvey Croze at the United Nations The database was considered to suffer from col- Environment Programme (UNEP) in Nairobi. linearity, because most climatic variables share This group was using geographical information qualities with (or are influenced by) other vari- systems (GIS) for various African continent-wide ables in the database. Fitting logistic regression 218 B. Perry et al. models to disease occurrence with highly correl- exploration of modelling in impact assessment, ated independent variables can lead to mislead- and ILRAD, in collaboration with FAO, organ- ing conclusions if the true biological meaning is ized a modelling workshop in Nairobi in No- not clearly understood. This case study used the vember 1992 to explore the approaches being analysis of principal components to reduce large made by different research groups (Perry and numbers of variables to smaller sets of variables Hansen, 1994). that more efficiently describe the important fea- The collaboration with Imperial College tures of the database. London, and subsequently the University of An important early stage in the impact Warwick (where Graham Medley moved), led to assessment process had been to determine more the first attempt to develop a quantitative frame- accurately the distribution of diseases and their work of the infection dynamics of theileriosis vectors. This had four major impacts: (Medley et al., 1993). It was able to demonstrate 1. It enhanced the understanding of disease how infection was maintained in cattle populations vector distributions, and factors affecting these, and quantified the important role of carrier ani- such as climate and vegetation. mals. With the progressive understanding emer- 2. It enhanced the understanding of the role of ging from field studies in different regions, the GIS in predicting disease and vector distributions, model was updated and reported by O’Callaghan and the need for appropriate high-resolution geo- et al. (2003). referenced databases, including the use of satellite- The principles behind this first model were derived imagery. then applied to heartwater, and a quantitative 3. It allowed exploration of new methods for framework was produced that demonstrated improving the predictive capacity of distribu- for the first time the concept of endemic stability tion models (illustrated by Duchateau et al., (O’Callaghan et al., 1998). The approach went 1997). on to explore the effects of vaccination against 4. It alerted Ethiopia, where R. appendiculatus heartwater (O’Callaghan et al., 1999). had never been recorded, to the climatic suit- The various observational field studies and ability of the survival of this vector in certain the supportive modelling initiatives raised many parts of the country (Norval et al., 1991). This issues, particularly the mechanism for establish- stimulated the Ethiopian government to revise ment of endemic stability to both theileriosis and its policy on importation of live cattle from heartwater, and the implication of different inter- Kenya. A new study confirming the susceptibil- ventions, particularly tick control, on the devel- ity of Ethiopia to ECF has since been published opment and maintenance of endemic stability. (Leta et al., 2013), repeating the warnings made This led to an extrapolation of the findings to by ILRAD in 1991. other diseases, including malaria, warning that certain interventions might interrupt endemic stability and lead to outbreaks of disease (Cole- man et al., 2001). This Lancet publication won Modelling the infection dynamics the ILRI’s award (and nomination for the CGIAR of vector-borne diseases Chairman’s Award) for the Outstanding Scientific Article of 2002. Soon after the observational studies on ECF on the Kenya coast had started, it was felt that there was a need to develop a mathematical model (Anderson and May, 1992) of T. parva in- fection dynamics that could contribute to our Impacts of Trypanosomiasis understanding of disease impacts and offer a and its Control framework to test the effect of interventions, such as vaccination. This started a 15-year col- Economic impact of trypanosomiasis laboration on infection dynamics of tick-borne infections with Graham Medley, which eventu- The focus of impact assessment remained largely ally moved from theileriosis to E. ruminantium on tick-borne pathogens until 1997. In the early infection (heartwater). This launched a wider 1990s, the epidemiology group adapted a model Veterinary Epidemiology at ILRAD and ILRI, 1987–2018 219 that had been used for theileriosis impact assess- The epidemiology of resistance ment to quantify the impacts of trypanosomia- to trypanocides sis and its control in Cameroon, Gambia, Côte d’Ivoire and Zimbabwe. The model estimated In the late 1990s, ILRI began research on the the proportion of the national herd at risk of the epidemiology and impact of trypanosomiasis in disease and the annual economic cost of the dis- several African countries. Field work was under- ease and produced a breakdown of the costs into taken in Uganda, Kenya, Tanzania and Zambia production losses and input costs. Work based (in collaboration with the national agricultural on this model was reported in Shaw (1992), in research system and the University of Glasgow) Mukhebi et al. (1993) and in the later review of and Burkina Faso (in collaboration with the Shaw (2009). Centre International de Recherche-Développement Subsequently an ex ante model of a poten- sur l’Elevage en zone Subhumide (CIRDES) and tial trypanosomiasis vaccine (Kristjanson et al., the Free University of Berlin (FUB). Highlights 1999) indicated that the potential benefits of this work included the following: of improved trypanosomiasis control, in terms of meat and milk productivity alone, were • Support from the Bundesministerium für US$700 million per year in Africa. The disease wirtschaftliche Zusammenarbeit (BMZ) for cost to livestock producers and consumers was several phases of an ILRI/CIRDES/FUB an estimated US$1340 million annually, with- project on the epidemiology of trypanocide out including indirect livestock benefits such as resistance in West Africa (see Chapter 3, manure and traction. Given an adoption period this volume); of 12 years, a maximum adoption rate of 30%, • Collaboration on trypanocide resistance a discount rate of 5%, and a 30% probability of in  eastern Africa (Kenya, Tanzania and the research being successful within 10 years, Zambia); the net present value of the vaccine research is • Epidemiology studies of drug resistance in estimated to be at least US$288 million, with Mukono County, Uganda. an internal rate of return of 33%, and a bene- • Support for drug resistance studies under- fit:cost ratio of 34:1. taken by the Kenya Trypanosomiasis Research While praising the estimated returns to tryp- Institute (KETRI) in collaboration with anosomiasis control, the critics were uncomfort- Glasgow University. able with these returns being attributed exclusively These studies resulted in a series of multi- to the effects of a vaccine. The predicted reduc- author and multi-institutional publications (e.g. tion of trypanosomiasis could in fact be achieved Gall et al., 2004; Knoppe et al., 2006). by several different interventions, including some for which technologies were already avail- able. It could also be an evaluation of more The development of a modelling effective deployment of tsetse traps, or of genetic- technique for evaluating control options ally engineered livestock resistance to the effects of trypanosomiasis or of effective chemother- An area of emphasis was the development of apy; the productivity impacts may be similar. models to understand factors influencing the What will differ will be the probability of suc- transmission dynamics of trypanosomiasis and cess, the cost of the research/implementation, assessing and predicting the impact of control the time to achieving that success, and the adop- strategies. The objective of this research was to tion rates. This is important, because there are determine whether transition models, as pro- those who believe that a vaccine will be out of posed by Diggle et al. (2002), could be applied. our reach for a long time to come, and while the This modelling approach offered two major ad- evaluation demonstrated probable benefits from vantages over standard methods. The first is that trypanosomiasis control, it was not specific to a risk factor associations can be assessed simul- vaccine as the way to achieve this. Notably, taneously for both new (incident) infections and 15 years on from this study, there is still no vac- recurrent infections after chemotherapy. The cine on the horizon, whereas the results were second is that such statistical methods can allow based on one being available 9 years ago. monthly rather than weekly or fortnightly 220 B. Perry et al. sampling intervals in the field. This latter feature Sustainable trypanosomiasis control in is crucial logistically and would allow the ana- the Ghibe Valley of Ethiopia lysis of data from a much wider variety of sam- pling sites, as monthly sampling is commonly The initial modelling work of ILCA’s research in employed. The use of transition models allowed the Ghibe Valley of Ethiopia moved on into an the distinction to be made between key factors environmental impact study, supported by the influencing both the incidence and persistence International Atomic Energy Agency (IAEA), of trypanosome infections in cattle in the Ghibe who were at the time exploring the potential role Valley, Ethiopia, over a 12-year period from 1986 of the sterile insect technique to eradicate tryp- to 1998 (Schukken et al., 2004). With an ob- anosomiasis. While there was general scepticism served average prevalence, based on microscopic over the widespread use of this technique, there examination, of approximately 50%, Ghibe ranked was at the time substantial political support for as an area of severe trypanosomiasis impact rela- wider tsetse eradication under the Programme tive to other tsetse-infested areas in Africa (Snow against African Trypanosomiasis (PAAT) pro- and Rawlings, 1999). The real benefit of using a gramme. This work also built on previous collab- transition model to investigate infection dynam- orative ILCA/ILRAD studies of environmental ics of trypanosomes in cattle is its ability to assess impact of long-term trypanosomiasis control in both the incidence and persistence of infections. the Ghibe Valley, including impacts on bird spe- This was particularly useful because the main cies richness (Wilson et al., 1997). factor influencing changes in incidence, namely tsetse control, differed from the factor most likely to Spatial modelling of tsetse distributions be responsible for increased duration of infection, namely resistance to commonly used trypano- Underlying several studies on trypanosomiasis cidal drugs. Age and the number of previous in- were studies on predicting the distribution of fections also influenced incidence and duration tsetse species. The GIS capacity set up in the late of infection, raising interesting hypotheses for fur- 1980s was subsequently applied to support ther investigation (e.g. potential of selection of studies on the impact of trypanosomiasis con- trypanotolerance in local Ethiopian breeds). trol (Perry et al., 1994) and later exploited by Robin Reid, who went on to explore various as- Sustainable trypanosomiasis pects of the environmental impacts of tsetse control in Uganda control (Reid et al., 2000) before moving into broader ecosystems research at ILRI. The ILRI epidemiology group began work The leaders in the use of statistical methods e xploring the historical resurgence of human and spatial climate and vegetation databases African trypanosomiasis (HAT; also known as were David Rogers and colleagues at the Univer- sleeping sickness) in Uganda (Fèvre et al., 2001; sity of Oxford (e.g. Rogers et al., 1996; Wint and Welburn et al., 2001) and went on to explore Rogers, 2000). However, with the greater en- different potential control options using modelling gagement of ILRI in predicting the effects of cli- techniques (McDermott and Coleman, 2001). mate change on the length of the growing period HAT remains an important disease in and the implications this had on livestock pro- Uganda, and cattle are its main reservoir. This duction systems, an assessment of the potential project assessed the role of cattle in human dis- for changing tsetse distributions was considered ease and how control of cattle trypanosomiasis (McDermott et al., 2001). Subsequent research can be used to reduce the public health burden has included work on the economic impact of of T. brucei rhodesiense HAT. Activities in- trypanosomiasis (Robinson et al., 2014a). cluded: (i) development of tests to differentiate human-infective and -non-infective T. brucei Preventing and containing trypanocide spp.; (ii) studies into cattle movement in new resistance in the cotton zone of West Africa outbreaks of HAT; and (iii) studies to evaluate factors that influence HAT risk, burden and In April 2012, a final report on ‘Preventing and under-reporting. containing trypanocide resistance in the cotton Veterinary Epidemiology at ILRAD and ILRI, 1987–2018 221 zone of West Africa’ was issued, which was one required to implement tsetse control interven- of a series of projects exploring drug resistance tions and strategic treatment of cattle to suppress in trypanosomiasis control. Previous work had trypanosome populations. Actions improving focused on methods to evaluate resistance to cattle health status, such as helminth control, may trypanocides in north-east Guinea, southern also help further suppress surviving trypano- Mali (Affognon et al., 2009; Talaki et al., 2009) somes. Longer-term research is needed to confirm and south-west Burkina Faso (Der et al., 2011), the subsequent dynamics of resistant trypano- and on testing integrated control strategies to some populations if trypanosomes re-e stablish. reduce the risk of new drug resistance. In the Finally, a preliminary assessment of the final phase of the project, the project continued potential impact of the investments made to date to evaluate resistance and raise awareness and in research on trypanocide resistance indicates capacity to address the problem across much of that adequate returns will be achieved to justify the rest of the zone and to scale up the prevention the investment (Affognon et al., 2010). strategies developed earlier. Appropriate strat- Research findings and outputs from the egies were also being developed for containing – project have been taken up by continuing efforts and, if possible, reversing – resistance in the in the region to improve control of trypanosom- pockets previously characterized, and a specific iasis, notably by a network to monitor drug resist- study was undertaken to assess the impact of the ance (RESCAO: Réseau d’épidémiosurveillance trypanocide resistance research efforts to date. de la résistance aux trypanocides et aux acari- This series of projects has generated an cides en Afrique de l’Ouest), the FAO PAAT and a important body of evidence for improving the 5-year, €3.1 million project involving the princi- sustainability of trypanosomiasis control in West pal German partners and CIRDES to extend the Africa and elsewhere in sub-Saharan Africa project approach and findings to new countries (McDermott et al., 2003; Clausen et al., 2010). (Togo, Ethiopia and Mozambique). The final project focused on four main outputs, with capacity strengthening in the region as a cross-cutting objective. First, national research Rabies Research: A Networking and teams generated evidence that trypanocide re- Capacity-building Role in Africa sistance occurs in several locations across the cotton zone of West Africa, and the partnership ILRAD had no mandate in rabies research, but, has provided national services improved tools for beginning in the early 1990s, joined the South- detecting and monitoring it. Through collabor- ern and Eastern Rabies Group (SEARG). Its first ation with the Institute of Tropical Medicine contribution was an overview paper on the (Antwerp), progress was made in developing epidemiology of rabies in Africa (Perry, 1992). markers for identifying resistance in trypanosomes As part of the capacity-building function of the (Delespaux et al., 2010); this is expected to pro- epidemiology and socio-economics programme, vide even more rapid and increasingly accurate partnership with the rabies control groups in diagnostics for detecting and monitoring drug Kenya and neighbouring countries was estab- resistance. lished, including the supervision of Philip Kitala Second, informational aids, decision tools in his PhD research on rabies in the Machakos and media messages targeting farmers and ani- District of Kenya, and a series of papers emerged mal health service providers to reduce the risk of (Kitala et al., 2000, 2001, 2002). Other stra- resistance were further developed (Grace et al., tegic contributions emerged from ILRAD and 2008, 2009). There is a better understanding of ILRI (Bingham et al., 1993, 1995; Perry, 1993, how farmers access information about animal 1995; Perry and Wandeler, 1993). In addition, health care and of the most important actors in the epidemiology team was called in to evaluate national-level information networks that com- the controversy surrounding the role of rabies and municate such information. A third set of activ- rabies vaccination in the demise of the African ities demonstrated the effectiveness of integrated wild dog packs in the Aitong region of Kenya’s control strategies for containing trypanocide Maasai Mara (Macdonald et al., 1992). resistance in a location once it has established. The impacts of the engagement with rabies In such situations, the public sector is probably were substantial, and mostly centred on the 222 B. Perry et al. building of a rabies epidemiology and control ILRAD in 1995, the EU, which at the time was network through SEARG, in capacity building making large investments in rinderpest eradica- on rabies epidemiology, diagnosis and control tion through the AU-IBAR, approached both throughout the eastern and southern African institutions. The PARC programme had been region, and in highlighting priority research planning for some time to undertake an economic needs. Furthermore, ILRI’s work contributed to evaluation of the rinderpest control programme. some of the principles of dog rabies control, such However, the funds made available by the EU were as the need to understand the vaccination cover- not considered sufficient to employ an independ- age required to prevent rabies (Coleman and Dye, ent economist. As a result, the task was offered 1996), the need for a sound understanding of in 1994 to both ILRAD and ILCA, on the suppos- population ecology (in this case, dog ecology) in ition that they could supplement the limited order to target vaccination initiatives (Perry, 1993) funds with their institutional capacities in agri- and the need to exploit community engagement in cultural economics. ILRAD, with its historical rabies vaccination campaigns (Perry et al., 1995). focus on vector-borne haemoparasitic diseases, turned the offer down, while ILCA accepted it. The two institutions were then amalgamated, and ILRI was born, and the newly created Systems The Economic Impacts Analysis and Impact Assessment Group (SA/IA, of Rinderpest Control the successor to the Epidemiology and Socioeco- nomics Programme) ‘inherited’ the project. This Rinderpest has had a devastating effect on the provided a new opportunity for ILRI to work livestock industries of Africa since its introduc- with AU-IBAR, and two agricultural economists tion to the continent in the late 19th century. In were recruited under the leadership and super- its classical form, it was responsible for high vision of the SA/IA group. levels of mortality and its mere presence con- The study indicated that, for a sample of ten strained trade in livestock. During the 1960s, sub-Saharan African countries, the rinderpest the first coordinated international control pro- campaign was implemented in a cost-effective gramme was put in place, known as the Joint manner, with average per livestock unit costs Project (JP) 15. Although largely successful, rin- appearing within the narrow range of US$0.30– derpest returned in a major epidemic through- 0.66 (Tambi et al., 1999). Benefit–cost analysis out much of the continent after JP15 concluded revealed that the benefits of the campaign in in the late 1970s. As a result, the Pan-African each of the ten countries covered the value of Rinderpest Control (PARC) programme was ini- the investment. The estimated average return tiated under the auspices of the African Union– over the ten countries of US$1.98 for each US Interafrican Bureau for Animal Resources dollar invested in the campaign indicated that (AU-IBAR) funded by the European Union (EU) rinderpest control in Africa has been economic- and national governments to control and ultim- ally profitable. The net present value of US$32 ately eradicate rinderpest from Africa. A decade million indicates that the rinderpest campaign after the campaign started in 1986, increasing has been a wise public investment decision. donor concern about its impact, coupled with an The work by ILRI scientists demonstrated increasing public and private demand for infor- further that rinderpest control has also im- mation on the benefits and costs of rinderpest con- proved the well-being of livestock farmers in trol, prompted the call for an economic impact sub- Saharan Africa, as well as that of consumers assessment of the campaign. of livestock products. Analysis of the distribu- Despite Africa being the last bastion of rin- tion of welfare gains from rinderpest control be- derpest before its global eradication in 2011, tween producers and consumers revealed that ILRAD did not become involved in research into producers derived the greater share (80%) of the its control, although following its eradication, US$64 million in net value of production losses ILRI did exploit the participatory epidemiology avoided due to rinderpest control in the ten and surveillance tools used in the final phases of countries, while consumers derived approximately the campaign during its research into HPAI in 20% in net benefits from increased supplies lead- Indonesia. Just before the merger of ILCA and ing to lower prices (Roeder and Rich, 2009). Veterinary Epidemiology at ILRAD and ILRI, 1987–2018 223 Applying Economic Impact A regional FMD coordination unit was set up, Assessment Tools to Foot-and-mouth based in Bangkok, Thailand, led by Laurie Gleason, Disease Control and an advisory committee was established, on which Brian Perry of ILRI was invited to sit. The With the formation of ILRI, the new institution first meeting was held in Bangkok on 1 March embarked on the development of an appropriate 1998. This set the scene for the first of the FMD research role in what was a new sphere of influ- economic impact studies, which focused on Thai- ence for the organization, and an Asia Action land within a South-east Asia regional context. Group was established to plan strategic engage- Thai epidemiologist Wantanee Kalpravidh was ment. For the epidemiology group, the first assigned to the study, and agricultural economics contact with the new region was a strategic support was provided by Suzan Horst of Wagen- attendance at the Federation of Asian Veterin- ingen University, the Netherlands. The World ary Associations (FAVA) Conference in Cairns, Reference Laboratory for FMD (WRL-FMD) in Australia, 24–28 August 1997, where a series Pirbright, UK, provided the FMD-specific tech- of meetings was held with those attending. The nical support. Later in 1998, after he was recruited representatives of the Australian Centre for to ILRI’s epidemiology and impact assessment International Agricultural Research (ACIAR; group, Tom Randolph took over the economic John Copland), and the OIE (Yoshihiro Ozawa) impact analysis components of the initiative. facilitated a discussion with multiple partners The first product of this work was presented on how ILRI could provide added value to on- at the annual meeting of the South-east Asia going initiatives within the Asian region. Foot-and-Mouth Disease (SEAFMD) group, under For ILRI to make an effective move into Asia the auspices of OIE, in Phnom Penh, Cambodia, in in animal health research, it was considered that February 1999, and emerged as a peer-reviewed it should exploit the generic research capacity it publication later the same year (Perry and Ran- had developed in epidemiology and economic dolph, 1999). This group provided a cost–benefit impact, rather than its traditional disease focus analysis of different FMD control scenarios and of vector-borne haemoparasites of ruminants, different emerging trading opportunities that as these were not considered to be a high priority would result from greater FMD control. in the region. In addition, it was considered that ILRI work on FMD in the Philippines had a the first phase of ILRI’s involvement in the re- significant impact. Randolph et al. (2002) devel- gion should be to better define constraints to oped scenarios, based on the plans and timetable livestock production and trade, and ways of alle- of the Government of the Philippines, and more viating these. To this end, it was suggested that optimistic and pessimistic assumptions, each dis- enhancing the regional animal disease surveil- cussed in detail with national stakeholders. It also lance and monitoring programme, the Animal had an additional component, which was an ana- and Plant Health Information System for Asia lysis of the distribution of the benefits. It illustrated (APHISA), in the field of epidemiological and that, while the FMD control programme was funded economic impact assessment would be the most entirely from public sector government coffers, appropriate entry point. ILRI was invited to in eradication scenarios the major beneficiaries undertake an initial case study on the impact of would be the private sector pig producers, traders FMD in the region, and the impact of alternative and marketers; the commercial swine sector was FMD control strategies. This provided an oppor- estimated to capture 84% of the benefits gener- tunity to enhance impact assessment capacity in ated by the public investment in eradication, ver- the region, initiate longer-term disease control sus 4% by backyard swine producers. priority evaluations and provide immediate sup- ILRI later undertook a collaborative study port to the newly created OIE-coordinated FMD on FMD on smallholder agricultural enter- control and eradication programme that had prises in southern Laos (Perry et al., 2002a). been set up. The OIE operation was funded by This demonstrated the widespread impacts the the Australian, Swiss and Japanese governments, disease had on multiple species and enterprises with the Swiss Government offering to provide in the smallholder systems of southern Laos and the ILRI portion of the funding. has been often cited as evidence of the disrup- tion to the livelihoods of smallholders globally. 224 B. Perry et al. The southern Africa FMD economic 16% of the increased value of economic activity impact study resulting from trade is eventually transferred as income to low-income households in both rural The results of a benefit–cost analysis showed and urban areas. that FMD control would benefit the economy of The direct impacts on the poor of FMD, and Zimbabwe (Perry et al., 2003). First, in a com- of measures established to control it, are very parison between the Baseline Scenario and the limited. FMD has not been a problem in commu- pessimistic FMD Control Scenario 3 (in which nal areas where the majority of the poor live, disinvestments in FMD control by 50% and re- and its effects on indigenous cattle are consider- sultant loss of beef export markets was pre- ably less than on commercially orientated herds. dicted), it was shown that for every US$1 that Furthermore, despite the fact that about 75% of Zimbabwe disinvests in the FMD control pro- poor households own or have access to cattle, gramme, a further US$5 would be lost by the over 60% of these households own fewer than country. No transboundary effects were taken five animals. Most of these households use cattle into account, and the losses calculated were in- for wealth storing and other livelihood functions curred by Zimbabwe alone. However, the associ- such as traction, and do not have the herd size ation of the outbreak of FMD in south-eastern capacity to engage actively in commercial cattle Botswana in March 2002 (after over 30 years of marketing. As such, only about 2% of house- freedom from the disease) with the outbreaks in holds are engaged in regular marketing of cattle. western Zimbabwe suggested that the costs to This study provided one of the most extensive the region as a whole of Zimbabwe’s disinvest- analyses of FMD impacts carried out, applying ments could be much greater. new methods such as the SAM/CGE modelling to Second, the results showed that if Zim- a most complex subject. Randolph et al. (2005) babwe were to invest further in the fences and explored further the highly skewed equity impacts the veterinary service infrastructures required emerging from this study. to create a much larger and much more secure export zone that was internationally recognized as FMD free by the OIE, there would be returns of Economic impacts of FMD in Peru, approximately US$1.5 for every US$1 invested. Colombia and India As in the disinvestment scenario, this does not incorporate benefits to the region as a whole In 1995, the Joint FAO/IAEA Division of the through greater disease security for FMD control, IAEA requested ILRI support for an economic nor does it include the other benefits that would assessment of FMD control. An economic impact result from an enhanced national veterinary assessment plan emerged (Romero et al., 2001). service. This analysis did not consider whether Unlike the South-east Asia partnerships, these Zimbabwe would be able to maintain the cap- Andean initiatives did not result in completed acity, in terms of quantity and quality of beef, to benefit–cost analyses; rather, the impacts of supply the export market on a sustainable basis. these studies in the region were in the field of Importantly, the distributions of the costs networking, training, capacity building, aware- and benefits turned out to be highly skewed. Ex- ness raising and methodology development. penditures from FMD control are borne almost entirely by the public sector, but when losses from trade bans resulting from FMD outbreaks are included, private sector costs are dominant. Economic impacts of FMD control in The majority of impacts of FMD and the benefits endemic settings in low- and from its control are related to the ability to trade middle-income countries internationally, and so most of the benefits ac- crue to the commercial sector, comprising cattle In April 2006, Brian Perry approached the Well- production, beef processing, and related input come Trust and the EU for support for an inter- industries and services. The Social Accounting national workshop on the research needs for Matrix/Computable General Equilibrium (SAM/ better FMD control in endemic settings of many CGE) modelling indicates that approximately low- and middle-income countries. This was Veterinary Epidemiology at ILRAD and ILRI, 1987–2018 225 approved, and the Global Roadmap for Improv- FMD-free countries, and their trading opportun- ing the Tools to Control Foot-and-Mouth Disease ities, while Programme 2 focused on the needs in Endemic Settings was duly held in Agra, of endemic settings, principally in developing India, in late November 2006 (Perry and Sones, countries. At this point, ILRI assumed the lead- 2007b) and the Global Roadmap was launched ership of Programme 2, and initiated contact with in April 2007. potential research sponsors and development agencies. Programme 1 was assigned to the four participating research laboratories in the USA, the UK, Canada and Australia, funded through The Global Foot-and-Mouth Disease national bodies supporting each laboratory. The Research Alliance (GFRA) outcome would be a better set of tools to manage the risk to the four countries currently posed by ILRI developed a proposal for a GFRA with five FMD in endemic areas. This would focus on im- research pillars: proved vaccines and diagnostics and the further development of antivirals. 1. A detailed understanding of the host immune Program 2 was designed to focus on devel- responses to FMD virus. oping better tools for use in endemic areas with 2. Development of a new generation of inexpen- the overall aim of a gradual reduction of the dis- sive and thermostable vaccines that meet the ease in endemic areas. It was recognized that requirements of both endemic and epidemic FMD targeted research would be needed to develop control and management. specific tools for Programme 2 and that this work 3. A full understanding of the factors that per- could occur both inside and outside of the cur- mit the development of virus carrier animals, rent GFRA partner institutes. It was foreseen the risk that they pose and options for managing that, for Programme 2, much still needed to be them. done in the area of epidemiology and impact as- 4. The identification of antiviral compounds to sessment, and that ILRI would take the lead in inhibit virus replication and rapidly reduce virus this, and that it would be a core component of release. funding under Programme 2. 5. Quantitative predictions of the performance However, ILRI management was not at the of the new technologies developed in different time supportive of ILRI’s engagement with GFRA, settings through the use of epidemiological and in part because it was considered that FMD did economics models. not rank highly enough in the health constraints The leadership of each pillar was assigned to dif- facing smallholder producers. ferent institutions, with ILRI taking on this latter pillar. The US$70 million proposal was launched as a Strategic Global Research Partnership for the Control of FMD in April 2004. Rift Valley Fever The proposal was received with enthusiasm on the scientific side, but participants urged the Research into RVF at ILRI commenced with an development of a business plan. This was duly evaluation of the impacts of the 2006/2007 commissioned and in June 2005 representatives outbreak that occurred in eastern Africa. This of the partnership set out to visit key donors (DFID work was commissioned by USAID and FAO, and and the Department for Environment, Food & focused primarily on the north-eastern region of Rural Affairs (DEFRA) in the UK; the EU; the Kenya, thought to be the epicentre of the epi- Canadian International Development Agency demic in Kenya. The outbreak occurred between (CIDA) in Canada; and various partners in December 2006 and March 2007 and affected the USA). more than 700 people; approximately 150 human Considering the different contexts of FMD fatalities occurred throughout the country. It control in the developed and developing world, was believed that people suffering severe clinical GFRA revised its approach to adopt two comple- disease had close contact with infected livestock. mentary programmes. Programme 1 was targeted The impact assessment was implemented at the FMD vaccine and diagnostic needs of by a team of epidemiologists, economists and 226 B. Perry et al. social scientists. A memorandum of understand- suggest that the 2006/2007 outbreak caused a ing was established with Kenya’s Department of total of 3974 DALYs, or 1.5 DALYs per 1000 Veterinary Services (DVS), enabling the DVS to population. Provisional results further show participate in the project as a key partner. This that strategies to enhance mass vaccination of work was later extended to the Arusha region of cattle and camels over a sustained 2-year period Tanzania with additional support from FAO. would greatly reduce DALYs. It also showed that Surveys conducted in both sites utilized partici- integrating vector control measures, for instance patory epidemiological tools and the data col- through the application of larvicides, would yield lected were synthesized and published (Jost et al., even better results, although the practicability of 2010). The key observations made were that there implementing such interventions through insti- were major weaknesses in preparedness and the tutional collaboration has not been fully resolved. response to the outbreak, and that pastoralists noticed RVF-compatible events long before offi- cial notifications were made by the government. RVF risk maps for eastern Africa At the same time, economic impact assessments were conducted by Karl Rich and Francis ILRI epidemiologists have developed risk maps Wanyoike (Rich and Wanyoike, 2010). This dem- for the eastern Africa region that can be used to- onstrated that the disease induced substantial gether with the decision support tool to enhance production losses, employment losses and reduc- targeting and evaluation of RVF interventions. tions in operating capital among various value This work builds on previous studies done by the chain actors including producers, livestock traders, National Aeronautics and Space Administration animal transporters, and slaughterhouse and (NASA) and other research institutions such as butchery operators. It was estimated that the out- the Centers for Disease Control and Prevention break cost the Kenyan economy US$32 million. (CDC). Two methods that have been applied for These findings fuelled discussions on the this analysis are: (i) ecological niche modelling need for improved warning systems and a based on the Genetic Algorithm for Rule-set Pre- structured contingency plan for managing the diction (GARP); and (ii) a logistic regression model, disease. More importantly, timelines developed followed by mapping predicted probabilities on a with local communities showing events that spatial landscape. Both models use historical preceded the outbreak were transformed into a data on RVF outbreaks from the 2006/2007 decision support tool (Consultative Group for outbreak. Statistical analyses demonstrate that RVF Decision Support, 2010). This is considered RVF risk is significantly associated with excep- a major contribution by ILRI and FAO to RVF tionally high rainfall, low altitude, clay soils and contingency planning given that this tool has high normalized difference vegetation indices. now been incorporated into the Ministry of Live- Such maps could also be used to enhance our stock’s Contingency Plan for RVF. More work still understanding of ecological niches for the virus, needs to be done, however, to develop a harmon- particularly if the existing hotspots can be classi- ized contingency plan that unites the public fied based on their abilities to support disease health and veterinary sectors in line with the persistence. More importantly, these maps are One Health paradigm. being integrated with socio-economic variables to determine areas that are most vulnerable to the Economic impact assessment of disease given their livelihood patterns, capacities to control options and calculation of access public health services and literacy levels. disability-adjusted life years (DALYs) The outputs of the impact study supported the Land-use change and RVF infection formulation of a new study to assess the cost- and disease dynamics effectiveness of RVF control options from a multidisciplinary perspective. This work also With increasing awareness of the impacts of aimed to estimate economic costs of RVF in RVF epidemics, there is a growing interest in de- people using DALYs. The estimates generated termining processes that cause RVF occurrence Veterinary Epidemiology at ILRAD and ILRI, 1987–2018 227 and transmission, as well as those that promote of a process to bring greater coordination among its persistence during inter-epidemic periods. its membership in the funding of livestock re- ILRI is currently leading a project in Kenya that search. As part of this process, it sought to better seeks to understand RVF drivers from a multidis- define the research options and priorities, and ciplinary perspective. The project is founded on DFID proposed that these be placed in the con- the premise that intact ecosystems can regulate text of poverty reduction. This required first disease epidemics, and that factors that disrupt defining poverty and the association with live- ecosystem structure and function, such as cli- stock, and then quantifying the association, a mate, land use and demographic changes, con- process that continues (Robinson et al., 2014b). tribute to disease emergence and spillovers. The There were seven major components to the project involves local partners such as the Kenya study (Perry et al., 2002b). The first was to de- Medical Research Institute (KEMRI), DVS, Uni- scribe and quantify the distribution and extent versity of Nairobi and Ministry of Health. of poverty in South-east Asia, South Asia and Preliminary observations indicate that there sub-Saharan Africa, and to determine the asso- is a great potential for endemic transmission of ciation of poverty with different agricultural RVF in irrigated areas established in arid and production systems that involve livestock. These semi-arid zones, as poorly managed drainage two tasks were accomplished in a companion systems and watersheds provide ideal conditions study commissioned by DFID (Thornton et al., for the development of primary and secondary 2002), which developed maps to quantify popu- vectors of RVF. Observations also show that lations of poor livestock keepers and to predict areas that are RVF endemic tend to be vulnerable how they would change over the next five dec- to other infectious/zoonotic diseases, malnutri- ades. The results provided data on the number of tion or insecurity, presenting multiple chal- poor (people on less than US$1 per day) in each lenges to the implementation of sustainable RVF of the major livestock production systems of the control strategies. world. These figures served as a weighting factor in determining the importance of different live- stock diseases to the poor. Epidemiology of Gastrointestinal The second component was to determine Parasites the priority species to the poor in each region and production system. This was undertaken by a literature review and through stakeholder work- ILRI published a field and laboratory handbook shops in West Africa, eastern, central and south- on the epidemiology and diagnosis of gastro- ern Africa, South Asia and South-east Asia. intestinal parasites entitled The Epidemiology, The third component was to quantify the Diagnosis and Control of Gastro-Intestinal Parasites disease constraints by species. Diseases and of Ruminants in Africa (Hansen and Perry, 1990). syndromes considered to negatively affect the A second edition, The Epidemiology, Diagnosis and livelihoods, productivity outputs and marketing of Control of Helminth Parasites of Ruminants, was livestock products by the poor were identified in published several years later (Hansen and Perry, a set of stakeholder workshops. The socio- 1994). economic (primarily production losses and con- trol costs incurred by the poor), zoonotic (for those diseases transmissible from animals to Priorities in Animal Health Research humans) and national impacts (a combination for Poverty Reduction of marketing impacts on the poor with public sector expenditures on disease control) were In 2000, ILRI began work on animal health and identified and scored. poverty reduction involving scientists and opin- Published literature on the impact of live- ion leaders in Africa, Asia, Europe and North stock diseases and of their control in the target America. This eventually delivered one of the regions was scrutinized and synthesized by com- highest-impact products of ILRI’s epidemiology missioned reviews. Research opportunities to al- group (Perry et al., 2002b). The Inter-Agency leviate these constraints were then identified. Donor Group (IADG) had just been born, as part First, research needs were identified from the end 228 B. Perry et al. users’ perspectives by participants in several was again approached by the Trust, and follow- regional workshops. Second, research oppor- ing discussions in London, the group submitted tunities were identified from the upstream per- a pre-proposal for a Wellcome Centre for Stra- spective by international experts specializing in tegic Veterinary Epidemiology based at ILRI in different diseases. In addition to identifying rele- Nairobi. After deliberations by the Trust, the vant research opportunities, the experts were proposal was not accepted for funding, but it did asked to estimate the cost, time frame, probability reopen the door to dialogue. Strongly influenced of success and available capacity to undertake by the report by Perry et al. (2002b), in July such research. To ensure that issues other than 2002 the Trust announced a new funding pro- technology generation were addressed, additional gramme entitled ‘Animal Health in the Develop- reviews of research opportunities for the better ing World’, under which it set aside £25 million delivery of animal health services were commis- over a period of 5  years to fund researchers to sioned. A specific review of the role of research develop methods of predicting and controlling into the genetics of resistance to disease was also outbreaks of animal diseases. commissioned. A large number of research proposals were The next step was to score the disease im- developed by ILRI in partnership with institutions pacts (Shaw et al., 2003), synthesize the disease in the UK and USA, including gastrointestinal impacts on the poor with the research needed parasitism, FMD epidemiology and dynamics, to reduce them and identify priority research anti-tick vaccines, livestock/disease information opportunities. A conceptual framework matrix platforms for East Africa and African swine fever, was developed to classify different types of disease- among many others. This later led to the Live- specific research: (i) transferring knowledge and stock for Life Programme, launched in December available tools; (ii) developing improved tools 2005. and strategies that were better delivered; and (iii) In January 2007, the Wellcome Trust held developing new tools and approaches by the a meeting to give scientists funded under earlier contribution the research product will make to grant programmes the opportunity to present poverty reduction (by securing the assets of the research findings, and to consider future needs poor; reducing the constraints to intensification in the field (the meeting was entitled ‘Animal or improving marketing opportunities). Health Research: Recent Developments and This study has had a lasting impact on re- Future Directions’). To coincide with the meet- search priorities for development and is still the ing, a policy review paper was commissioned most cited reference in this context. In addition, (Perry and Sones, 2007a), and ILRI presented it set the stage for measuring the association be- an invited talk on the challenges of research tween poverty and livestock, and for applying outputs influencing policy (Perry and Hooton, greater emphasis to the impacts that research in 2007). animal health have on the processes of poverty In summary, the epidemiology group at reduction, rather than simply on national agri- ILRI undoubtedly had a substantial impact on cultural development. The methodology has the shaping and development of the Wellcome been further developed (Perry and Grace, 2009). Trust’s programmes of funding for livestock dis- ease control in the developing world, and the impacts were spread to many different research The Wellcome Trust Epidemiology institutions and countries. Initiatives In January 2002, the DFID-commissioned study entitled ‘Investing in Animal Health Research to The Broader Economic Impact Alleviate Poverty’ was published (Perry et al., Contributions 2002b). As described above, this had been com- missioned by IADG on pro-poor livestock research Economic impact assessments gained increasing and development, of which the Wellcome Trust momentum as ILRI’s mandate broadened. In a was a member, and had been represented by paper prepared as an invited plenary presenta- Catherine Davies. The ILRI epidemiology group tion to the 17th International Conference of the Veterinary Epidemiology at ILRAD and ILRI, 1987–2018 229 World Association for the Advancement of Vet- high-value livestock product markets in Europe erinary Parasitology, Perry and Randolph (1999) and elsewhere. The studies in Ethiopia showed wrote: ‘The traditional veterinarian views dis- that investments in the quarantine and testing ease as evil, and often embarks on a career with required to ensure that beef feedlots were free of a “Superman” like determination to destroy it, FMD were not the limiting factors affecting the regardless of how important it is. To the classical economic viability of beef exports to Middle East healer, economic considerations are secondary. markets; rather, it was the high cost of feeding The economist on the other hand sees animal animals to ensure that the product arriving in disease as just one, and often an insignificant the market was competitive with others coming one, of a great spectrum of constraints to human from Australia, Brazil and other sources. and societal wellbeing that needs to be put in ILRI later studied the potential role of com- context.’ They concluded that no longer should modity-based trade on international market studies of the economics of diseases of produc- access by developing countries (Rich and Perry, tion animals be limited to animal scientists seek- 2009, 2011). This work concluded that, on a ing the collaboration of agricultural economists geographical basis, the benefits of commodity- to affix prices to estimated productivity losses. based trade are much more likely to be felt in Rather, a new discipline of animal health eco- countries like Argentina, Brazil and India than in nomics emerged in which the quality of economic African countries. Opportunities exist for south- evaluations depended on integrating the prod- ern Africa but are predicated largely on continued ucts of good epidemiological studies into economic preferential access that may or may not be frameworks. sustainable in the long term. While there are In addition, during 1998, the ILRI Epidemi- numerous opportunities for some African coun- ology and Disease Control group leader was tries in niche markets, it is also important to bal- approached by the director general of the OIE ance this potential with the sound exploitation to coordinate the design, compilation and edit- of livestock resources and a pragmatic under- ing of a special edition of the OIE Scientific and standing of the challenges in marketing and Technical Review on Animal Health Economics. competitiveness. The constraints that compli- This peer-reviewed edition comprised nine chap- cate market access for Africa are much more ters on different demands for economic impact those related to infrastructure, productivity and knowledge, and seven case study chapters ad- efficiency throughout the livestock supply chain, dressing specific diseases and different circum- and it is in these areas that policy attention is stances affecting the validity of economics studies. urgently required. This book, in combination with the study by Perry and Randolph (1999) mentioned above, served as important milestones for the inte- grated science of epidemiology and economics. The Responses to Highly Pathogenic Rushton (2009) described as it as ‘the first book Avian Influenza to bring together a number of important themes in animal health economics: farm-level economic The emergence of HPAI, initially in East and assessments; trade implications of sanitary re- South-east Asia with subsequent spread to Africa, quirements; and veterinary service delivery’. caused disquiet in all animal health research ILRI later did an economic analysis of the communities and institutions, and ILRI, in col- potential costs, benefits and competitiveness of laboration with the International Food Policy trade in meat from Ethiopia to the Middle East Research Institute (IFPRI), initiated a wide- (Rich et al., 2008). This report was further devel- ranging consultation to discuss where research oped for a peer-reviewed publication, presented could contribute (ILRI/IFPRI, 2006). It was not by Karl Rich at the International Food and Agri- a straightforward process, providing the chal- business Management Association (IAMA) in lenge of developing a framework and methods June 2009 (Rich and Perry, 2009), where it won for designing appropriate control strategy the Best Paper Award. At the time, it was widely interventions, and generating evidence of the believed that poor countries with abundant live- potential trade-offs with poverty reduction stock were well placed to develop exports to objectives. 230 B. Perry et al. ILRI was active in providing several back- first for ISVEE, in a special edition of the Kenya ground guidance and methodological frame- Veterinarian. The full meeting proceedings are works for the global response to HPAI. In 2009, now also available online (ISVEE, 1994). ILRI was a partner in the production of a man- The next ISVEE was held in Paris in 1998 ual entitled Introduction to Participatory Epidemi- with a strong representation from ILRI and its ology and its Application to Highly Pathogenic partners. But from here the level of participation Avian Influenza Participatory Disease Surveillance. grew substantially, and the 9th and 10th ISVEEs A Manual for Participatory Disease Surveillance (in Colorado and Viña del Mar, Chile, respectively) Practitioners (Ameri et al., 2009). ILRI also devel- brought the research of ILRI’s epidemiology oped a user guide for initial bird flu risk maps as group to new levels of recognition; in the Chile a contribution to improving the surveillance for meeting of 2003, the group had 29 papers and bird flu (Stevens et al., 2009). These have recently posters accepted. At this meeting, under joint been built on and updated in a new risk mapping sponsorship with the International Association report (Gilbert et al., 2014). of Agricultural Economists, Tom Randolph or- A Nigerian Avian Influenza Control and ganized a mini-symposium on Animal Health Human Pandemic Preparedness and Response Economics, which comprised plenary papers, in- Project (NAICP) began in July 2006 and invited dependent papers and a discussion forum. In ILRI to do an independent impact assessment this, he concluded that animal health economics of the project. The evaluation developed ten has established a solid, although remarkably ‘outcome pillars’ to depict the benchmark ‘gold narrow, foundation in the literature, but that it standard’ of best practices against which to had not exploited its potential (Randolph et al., evaluate NAICP (as discussed in Perry et al., 2003). While ILRI continues to be represented at 2010). The evaluation report was presented subsequent meetings, the commitment to and to the Government of Nigeria (Perry et al., impact of ILRI seen during the period 1994–2006 2011) and there were also two independent peer- has waned. reviewed publications (Henning et al., 2013; Bett et al., 2014). The Role of Epidemiology in ILRAD and ILRI The ISVEE Experience Veterinary epidemiology and socio-economic The International Symposium on Veterinary impact research has gone through both admin- Epidemiology and Economics, known commonly istrative and locality changes during its existence by its acronym ISVEE, has been held every over the last 27 years. It was a unified entity for 3 years since the inaugural meeting in Reading, 15 years, but in 2002, when the new ILRI strat- UK, in 1976. It brings together directors of vet- egy was developed, epidemiology and impact erinary services, disease control planners, quan- assessment became both fragmented and dimin- titative epidemiologists, agricultural economists, ished in human resource capacity. modellers and statisticians to present and dis- From 1987 to 1994 under ILRAD, it was the cuss on a wide range of diseases and issues. Epidemiology and Socioeconomics Programme, Kenya was proposed by ILRAD’s epidemiology also varyingly referred to as the Epidemiology group and later confirmed as the venue for 1994 and Socioeconomics Unit and Socioeconomics at the 6th ISVEE in Ottawa in 1991, with Brian Programme in emerging documentation. In the Perry as secretary of the organization (ISVEE, early days of ILRI from 1995 to 1997, epidemi- 1991). This provided the first opportunity to ology was accommodated under the newly cre- bring ISVEE to Africa, and to engage national ated Systems Analysis and Impact Assessment and regional programmes in presenting their Group, placing it under the Production Systems work and participating in the meeting. Through Programme, but this did not last for long, and the hard work and commitment of John Row- from 1997 to 2002 it became the Epidemi- lands, a statistician at ILCA, the full proceedings ology and Disease Control group under the Ani- were handed to participants as they registered, a mal Health Programme. Veterinary Epidemiology at ILRAD and ILRI, 1987–2018 231 The Impacts of ILRAD and ILRI’s Africa, to a greater understanding of the eco- Epidemiology nomic impact of rinderpest in Africa and of FMD in Africa, Asia and Latin America, and to Capacity development in veterinary regional understanding of the drivers of rabies epidemiology and impact control. More recently, epidemiology research assessment at ILRI has contributed substantially to our understanding of food safety risks in formal and informal markets, and to the dynamics During the 15-year period from 1987 to 2002, and risks of zoonotic diseases. The research has a substantial number of MSc and PhD stu- also contributed to the global understanding of dents were trained through incorporation into the importance of these and other diseases research activities; these were predominantly to African livestock systems, and to the par- from African countries. In addition, as men- ticular animal health constraints facing the tioned in the section on ISVEE, the group poorer sectors of Africa’s livestock-engaged and its associated students presented at many communities. international meetings and in most cases pub- lished their research findings in peer-reviewed journals. Impacts on ILRI’s research and strategy Impacts on national animal health departments and services During the days of ILRAD, the epidemiology and socio-economics programme had little or no im- The epidemiology group provided a role model pact on ILRAD’s research and strategy; rather, it of investigative problem solving, which was was seen as providing evidence justifying the ex- picked up, copied and adopted by other institu- istence of the laboratory-based vaccine research tions. However, this mostly occurred where for the two target haemoparasitic diseases. there was a specific donor-funded project to sup- Nevertheless, after ILRI’s birth in 1995, the port the establishment of an epidemiology group group did play an important role in providing and was more common in academic than in impact assessment services, which progressively public service bodies such as veterinary depart- enhanced the engagement of the institution ments. Many newly trained graduates return to with different national, regional and donor cli- their institutions with a sound training but do ents. This situation changed dramatically in not have the opportunity to build on that, often 2002 following the publication of the DFID- because of institutional weaknesses, with inad- commissioned study on animal research prior- equate financial resources for research and for ities for poverty reduction (Perry et al., 2002b). staff development. Veterinary epidemiologists rely The matrix of three ‘pathways out of poverty’ on collaboration with colleagues at the bench, in (see Perry et al., 2002b, Table ES1: securing the field and in the planning arena, and par- assets, reducing constraints to intensifica- ticularly with agricultural economists and other tion and improving market opportunities) social scientists, so may find substantial diffi- and three research and development opportun- culty functioning in a ‘conservative’ public sector ities (transferring knowledge and available environment. tools; improved tools, better strategies better delivered; and new tools and approaches) pro- vided the framework of the new institutional Impacts on animal health constraints in thematic structure, not just for animal health developing countries research but also for ILRI’s entire programme. Ironically, while a key product of the epidemi- ILRI’s epidemiology research has made substan- ology and disease control group provided the tial contributions to the understanding and con- framework for ILRI’s new strategy, by the trol of ECF and trypanosomiasis in Africa, to the same token it also triggered the decline of epi- preparedness and responses to RVF in eastern demiology as an institutional entity in ILRI. 232 B. Perry et al. Acknowledgements contributions of ILRI’s products in the fields of epidemiology and economics. This chapter is an We acknowledge the support of John McIntire updated version of Perry, B.D. (2015) Towards a and Delia Grace for commissioning this review. Healthier Planet: Veterinary Epidemiology Research We thank Bill Thorpe for his insightful com- at the International Laboratory for Research on Animal ments on an earlier draft. We also thank Carolyn Diseases (ILRAD) and the International Livestock Benigno, Martyn Jeggo and Roger Morris for pro- Research Institute (ILRI), 1987–2014. ILRI viding independent impact statements on the Research Report 38. ILRI, Nairobi. Notes 1 It had been conceived at the University of Florida by former ILRAD tick ecologist Andy Norval. On the tragic death of Norval in April 1994, ILRI assumed project leadership. 2 Currently, the OIE Sub-Regional Representation for South-East Asia (SRR-SEA) is engaged in FMD control in the region. The SRR-SEA evolved from the South-east Asia Foot and Mouth Disease Regional Coordination Unit (SEAFMD RCU), which was created in 1997 for the control of FMD in South-east Asia, coordinating various pre- vention and control initiatives in countries of the region, in particular Cambodia, Lao People’s Democratic Republic, Malaysia, Myanmar, the Philippines, Thailand and Vietnam. In 2010, the OIE and Association of Southeast Asian Nations (ASEAN) supported the membership of the remaining ASEAN countries (Brunei Darussalam and Singa- pore) and China, which has resulted in a vastly expanded programme, now renamed the South-East Asia and China Foot and Mouth Disease campaign (SEACFMD). References Affognon, H., Coulibaly, M., Diall, O., Grace, D., Randolph, T. and Waibel, H. (2009) Étude des politiques relatives aux stratégies de gestion de la chimiorésistance dans le cadre de la lutte contre la trypano- somose en Afrique de l’Ouest: cas du Mali. ILRI Research Report 17. ILRI, Nairobi. Affognon, H.D., Randolph, T.F. and Waibel, H. (2010) Economic analysis of animal disease control inputs at farm level: the case of trypanocide use in villages under risk of drug resistance in West Africa Live- stock. Research for Rural Development 22, 224. Ameri, A.A., Hendrickx, S., Jones, B., Mariner, J., Mehta, P. and Pissang, C. (2009) Introduction to Partici- patory Epidemiology and its Application to Highly Pathogenic Avian Influenza Participatory Disease Surveillance. A Manual for Participatory Disease Surveillance Practitioners. ILRI, Nairobi. Available at: https://cgspace.cgiar.org/bitstream/handle/10568/367/BirdFlu-Manual_final.pdf?sequence=2 (ac- cessed 13 February 2020). Anderson, R.M. and May, R.M. (1992) Infectious Diseases of Humans: Dynamics and Control. Oxford University Press, New York. Bett, B., Henning, J., Abdu, P., Okike, I., Poole, J., et al. (2014) Transmission rate and reproductive number of the H5N1 highly pathogenic avian influenza virus during the December 2005–July 2008 epidemic in Nigeria. Transboundary and Emerging Diseases 61, 60–68. Bingham, J., Perry, B.D., King, A.A., Schumacher, C.L., Aubert, A., et al. (1993) Oral vaccination of jackals against rabies: recent progress in Zimbabwe. Onderstepoort Journal of Veterinary Research 60, 1156–1157. Bingham, J., Kappeler, A., Hill, F., King, A., Perry, B.D. and Foggin, C. (1995) Efficacy of SAD (Berne) rabies vaccine given by the oral route in two species of jackal (Canis mesomelas and Canis adustus). Journal of Wildlife Diseases 31, 416–419. Chamboko, T., Mukhebi, A.W., Callaghan, C.J.O., Peter, T.F., Kruska, R.L., et al. (1999) The control of heart- water on large scale commercial and smallholder farms in Zimbabwe. Preventive Veterinary Medicine 39, 191–210. Clausen, P.H., Bauer, B., Zessin, K.H., Diall, O., Bocoum, Z., et al. (2010) Preventing and containing trypanocide resistance in the cotton zone of West Africa. Transboundary and Emerging Diseases 57, 28–32. Consultative Group for RVF Decision Support (2010) Decision-support tool for prevention and control of Rift Valley fever epizootics in the Greater Horn of Africa. American Journal of Tropical Medicine and Hy- giene 83, 75–85. Veterinary Epidemiology at ILRAD and ILRI, 1987–2018 233 Coleman, P.G. and Dye, C. (1996) Immunization coverage required to prevent outbreaks of dog rabies. Vac- cine 14, 185–186. Coleman, P.G., Perry, B.D. and Woolhouse, M.E.G. (2001) Endemic stability: a veterinary idea applied to human public health. Lancet 357, 1284–1286. Curry, J.J., Huss-Ashmore, R.A., Perry, B.D. and Mukhebi, A.W. (1996) A framework for the analysis of gen- der, intra-household dynamics and livestock disease control, with examples from Uasin Gishu District, Kenya. Journal of Human Ecology 24, 161–189. Deem, S.L., Perry, B.D., Katende, J.M., McDermott, J.J., Mahan, S.M., et al. (1993) Variations in prevalence rates of tick-borne diseases in Zebu cattle by agro-ecological zone: implications for East Coast fever immunisation. Preventive Veterinary Medicine 16, 171–187. Delespaux, V., Vitouley, H.S., Marcotty, T., Speybroeck, N., Berkvens, D., et al. (2010) Chemosensitization of Trypanosoma congolense strains resistant to isometamidium chloride by tetracyclines and enroflox- acin. PLoS Neglected Tropical Diseases 4, e828. Der, D., Affognon, H. and Sidibe, I. (2011) Réseaux sociaux et échange d’informations vétérinaires: cas de la trypanosomose animale chez les éleveurs de Solenzo dans la province des Banwa, Burkina Faso. Sciences and Medicines in Africa 3, 402–411. Di Giulio, G., Lynen, G., Medley, G.F., O’Callaghan, C.J., Kiara, H.K. and McDermott, J.J. (2003) Character- ization of theileriosis in pastoral production systems in Northern Tanzania. In: ISVEE 10: Proceedings of the 10th International Symposium for Veterinary Epidemiology and Economics, 17–21 November, Vina del Mar, Chile. ISVEE, Fort Collins, Colorado, p. 482. Diaz, C.M., Massawe, S.C., Clemence, A., Gitau, G.K., Kiara, H.K., et al. (2003) Risk mapping of East Coast fever in coastal and highland regions of Kenya based on predicted mortality and morbidity in- cidences. In: ISVEE 10: Proceedings of the 10th International Symposium for Veterinary Epidemiology and Economics, 17–21 November, Vina del Mar, Chile. ISVEE, Fort Collins, Colorado, p. 426. Diggle,P.J., Heagerty, P.J., Liang, K.-Y. and Zeger, S. (2002) Analysis of Longitudinal Data. Oxford Univer- sity Press, Oxford. Dolan, T.T. (1989) Theileriosis in Eastern, Central and Southern Africa: Proceedings of a Workshop on East Coast Fever Immunization Held in Lilongwe, Malawi, 20–22 September 1988. ILRI, Nairobi. Duchateau, L., Kruska, R. and Perry, B.D. (1997) Reducing a spatial database to its effective dimension- ality for logistic-regression analysis of incidence of livestock disease. Preventive Veterinary Medicine 32, 207–218. Fèvre, E.M., Coleman, P.G., Odiit, M., Magona, J.W., Welburn, S.C. and Woolhouse, M.E. (2001) The origins of a new Trypanosoma brucei rhodesiense sleeping sickness outbreak in eastern Uganda. Lancet 358, 625–628. Gachohi, J., Skilton, R., Hansen, F., Ngumi, P. and Kitala, P. (2012) Epidemiology of East Coast fever (Theileria parva infection) in Kenya: past, present and the future. Parasites and Vectors 5, 194–194. Gall, Y., Woitag, T., Bauer, B., Sidibé, I., McDermott, J., et al. (2004) Trypanocidal failure suggested by PCR results in cattle field samples. Acta Tropica 92, 85–95. Gilbert, M., Golding, N., Zhou, H., Wint, G.W., Robinson, T.P., et al. (2014) Predicting the risk of avian influenza A H7N9 infection in live-poultry markets across Asia. Nature Communications 5, 4116. Gitau, G.K., Perry, B.D., Katende, J.M., McDermott, J.J., Morzaria, S.P. and Young, A.S. (1997) The preva- lence of tick-borne infections in smallholder dairy farms in Muranga District, Kenya: a cross-sectional study. Preventive Veterinary Medicine 30, 95–107. Gitau, G.K. (1998) A study of factors influencing endemic stability and instability to Theileriosis and Babesiosis on dairy production in Murang’a District, Kenya. PhD thesis, University of Nairobi, Kenya. Gitau, G.K., Perry, B.D. and McDermott, J.J. (1999) The incidence, calf mortality and calf morbidity due to Theileria parva infections in smallholder dairy farms in Murangéa District, Kenya. Preventive Veterin- ary Medicine 39, 65–79. Gitau, G.K., McDermott, J.J., Katende, J.M., Brown, R.N. and Perry, B.D. (2000) Differences in the epidemi- ology of theileriosis on smallholder dairy farms in contrasting agro-ecological and grazing strata of highland Kenya. Epidemiology and Infection 124, 3–8. Gitau, G., McDermott, J.J., McDermott, B. and Perry, B.D. (2001) The impact of Theileria parva infections and other factors on calf mean daily weight gains in smallholder dairy farms in Murang’a District, Kenya. Preventive Veterinary Medicine 51, 149–160. Grace, D., Randolph, T., Diall, O. and Clausen, P.-H. (2008) Training farmers in rational drug-use improves their management of cattle trypanosomosis: a cluster-randomised trial in south Mali. Preventive Vet- erinary Medicine 83, 83–97. 234 B. Perry et al. Grace, D., Randolph, T., Affognon, H., Dramane, D., Diall, O. and Clausen, P.-H. (2009) Characterisation and validation of farmers’ knowledge and practice of cattle trypanosomosis management in the cotton zone of West Africa. Acta Tropica 111, 137–143. Hansen, J. and Perry, B. (1990) The Epidemiology, Diagnosis and Control of Gastro-Intestinal Parasites of Ruminants in Africa. ILRAD, Nairobi. Hansen, J. and Perry, B. (1994) The Epidemiology, Diagnosis and Control of Helminth Parasites of Rumin- ants. A Handbook. ILRAD, Nairobi. Henning, J., Bett, B., Okike, I., Abdu, P. and Perry, B. (2013) Incidence of highly pathogenic avian influenza H5N1 in Nigeria, 2005–2008. Transboundary and Emerging Diseases 60, 222–230. ILRI/IFPRI (2006) How research can support efforts to control avian flu in developing countries: first steps towards a research action plan. ILRI/-IFPRI Working Group on Avian Influenza. IFPRI, Washington, DC. ISVEE (1991) ISVEE 6: Proceedings of the 6th International Symposium on Veterinary Epidemiology and Economics, 12–16 August, Ottawa, Canada. ISVEE, Guelph, Canada. Available at: www.sciquest. org.nz/elibrary/edition/4891%20 (accessed 13 February 2020). ISVEE (1994) ISVEE 7: Proceedings of the 7th International Symposium on Veterinary Epidemiology and Economics, 15–19 August, Nairobi, Kenya. Published as The Kenya Veterinarian, Vol. 18(2). Available at: www.sciquest.org.nz/elibrary/edition/4896%20 (accessed 13 February 2020). Jaetzold, R. and Schmidt, H. (1983) Farm Management Handbook of Kenya, Vol. II, Part C: Natural Conditions and Farm Management Information. Ministry of Agriculture, Kenya. Jost, C.C., Nzietchueng, S., Kihu, S., Bett, B., Njogu, G., et al. (2010) Epidemiological assessment of the Rift Valley fever outbreak in Kenya and Tanzania in 2006 and 2007. American Journal of Tropical Medicine and Hygiene 83, 65–72. Karimi, S.K., McDermott, J.J., McDermott, B., Gitau, G.K., Gathuma, J.M. and Kinuthia, R.N. (2003). Risk factors for sero-prevalence of tick-borne diseases of calves in Maasai pastoral herds in Kajiado Dis- trict, Kenya. In: ISVEE 10: Proceedings of the 10th Symposium of the International Society for Veterin- ary Epidemiology and Economics, 17–21 November, Vina del Mar, Chile. ISVEE, Fort Collins, Color- ado, p. 507. Kariuki, D.P. (1988) Theileriosis in Kenya. In: Dolan, T.T. (ed) Theileriosis in Eastern, Central and Southern Africa. Proceedings of a Workshop on East Coast Fever Immunization Held in Lilongwe, Malawi, 20–22 September 1988. ILRI, Nairobi. Kiara, H., O’Callaghan, C.J., Randolph, T.F., McDermott, J.J. and Perry, B.D. (2000) Targeting East Coast fever control strategies based on the assessment of biological risk. In: ISVEE 9: Proceedings of the 9th Symposium of the International Society for Veterinary Epidemiology and Economics, 6–11 August, Breckinridge, Colorado. ISVEE, Fort Collins, Colorado, p. 292. Kiara, H.K., McDermott, J.J., O’Callaghan, C.J., Randolph, T.F. and Perry, B.D. (2003) Estimating bio- logical parameters for modelling the transmission dynamics of Theileria parva infection. In ISVEE 10: Proceedings of the 10th International Symposium for Veterinary Epidemiology and Economics, 17–21 November, Vina del Mar, Chile. ISVEE, Fort Collins, Colorado, p. 471. Kitala, P.M., McDermott, J.J., Kyule, M.N. and Gathuma, J.M. (2000) Community-based active surveil- lance for rabies in Machakos District, Kenya. Preventive Veterinary Medicine 44, 73–85. Kitala, P., McDermott, J.J., Kyule, M., Gathuma, J., Perry, B.D. and Wandeler, A. (2001) Dog ecology and demography information to support the planning of rabies control in Machakos District, Kenya. Acta Tropica 78, 217–230. Kitala, P., McDermott, J., Coleman, P. and Dye, C. (2002) Comparison of vaccination strategies for the con- trol of dog rabies in Machakos District, Kenya. Epidemiology and Infection 129, 215–222. Knoppe, T., Bauer, B., McDermott, J.J., Peregrine, A.S., Mehlitz, D. and Clausen, P.H. (2006) Isometamid- ium sensitivity of Trypanosoma congolense stocks from cattle in West Africa tested in mice and the drug incubation infectivity test. Acta Tropica 97, 108–116. Kristjanson, P.M., Swallow, B.M., Rowlands, G.J., Kruska, R.L. and de Leeuw, P.N. (1999) Measuring the costs of African animal trypanosomiasis, the potential benefits of control and returns to research. Agricultural Systems 59, 79–98. Leneman, J.M., McDermott, J.J., Okuthe, O.S. and Randolph, T.F. (2000) Farmers’ perceptions of East Coast fever risk and adoption of control strategies in Kenya. In: ISVEE 9: Proceedings of the 9th Symposium of the International Society for Veterinary Epidemiology and Economics, 6–11 August, Breckinridge, Color- ado. ISVEE, Fort Collins, Colorado, p. 308. Lessard, P., l’Eplattenier, R., Norval, R., Kundert, K., Dolan, T., et al. (1990) Geographical information sys- tems for studying the epidemiology of cattle diseases caused by Theileria parva. Veterinary Record 126, 255–262. Veterinary Epidemiology at ILRAD and ILRI, 1987–2018 235 Leta, S., de Clercq, E. and Madder, M. (2013) High-resolution predictive mapping for Rhipicephalus appendic- ulatus (Acari: Ixodidae) in the Horn of Africa. Experimental and Applied Acarology 60, 531–542. Macdonald, D., Artois, M., Aubert, M., Bishop, D.H.L., Ginsberg, J.R., et al. (1992) Cause of wild dog deaths. Nature 360, 633–634. Maloo, S.H. (1993) Vector-borne diseases and preventive medicine practises in smallholder dairy cattle in coastal Kenya. PhD thesis Veterinary School, University of Glasgow, Glasgow, UK. Maloo, S., Ngumi, P., Mbogo, S., Williamson, S., Thorpe, W., et al. (2001a) Identification of a target population for immunisation against East Coast fever in coastal Kenya. Preventive Veterinary Medicine 52, 31–41. Maloo, S., Rowlands, G., Thorpe, W., Gettinby, G. and Perry, B.D. (2001b) A longitudinal study of disease incidence and case-fatality risks on small-holder dairy farms in coastal Kenya. Preventive Veterinary Medicine 52, 17–29. Maloo, S., Thorpe, W., Kioo, G., Ngumi, P., Rowlands, G. and Perry, B.D. (2001c) Seroprevalences of vector-transmitted infections of small-holder dairy cattle in coastal Kenya. Preventive Veterinary Medicine 52, 1–16. Martinez, T., Meltzer, M., Perry, B.D., Burridge, M. and Mahan, S. (1999a) The effect of subclinical experi- mental Cowdria ruminantium infection in ewes on the growth and milk consumption of pre-weaning lambs. Preventive Veterinary Medicine 41, 105–118. Martinez, T., Meltzer, M., Perry, B.D., Burridge, M. and Mahan, S. (1999b) The effect of subclinical experi- mental Cowdria ruminantium infection on the health and reproductive performance of breeding ewes. Preventive Veterinary Medicine 41, 89–103. McDermott, J. and Coleman, P.G. (2001) Comparing apples and oranges – model-based assessment of different tsetse-transmitted trypanosomosis control strategies. International Journal for Parasitology 31, 603–609. McDermott, J.J., Kristjanson, P.M., Kruska, R.L., Reid, R.S., Robinson, T.P., et al. (2001) Effects of climate, human population and socio-economic changes on tsetse-transmitted trypanosomiasis to 2050. In: Seed, R. and Black, S. (eds) World Class Parasites, Vol. 1. The African Trypanosomes. Kluwer, Boston. McDermott, J., Woitag, T., Sidibé, I., Bauer, B., Diarra, B., et al. (2003) Field studies of drug-resistant cattle trypanosomes in Kenedougou Province, Burkina Faso. Acta Tropica 86, 93–103. Medley, G.F., Perry, B.D. and Young, A.S. (1993) Preliminary analysis of the transmission dynamics of Theileria parva in eastern Africa. Parasitology 106, 251–264. Minjauw, B., Perry, B.D., Kruska, R., Peter, T., Mahan, S., et al. (1998) Economic impact assessment of heartwater in southern Africa. In: Economic Impact Assessment of Heartwater in Southern Africa. Ninth International Conference of Association of Institutions of Tropical Veterinary Medicine (AITVM), 14–18 September, Harare, Zimbabwe. Minjauw, B., Kruska, R., Odero, A., Randolph, T.F., McDermott, J.J., et al. (2000) Economic impact assess- ment of Cowdria ruminantium infection and its control in southern Africa. In: ISVEE 9: Proceedings of the 9th Symposium of the International Society for Veterinary Epidemiology and Economics, 6–11 August, Breckinridge, Colorado. ISVEE, Fort Collins, Colorado, p. 645. Mukhebi, A., Wathanga, J., Perry, B.D., Irvin, A. and Morzaria, S. (1989) Financial analysis of East Coast fever control strategies on beef production under farm conditions. Veterinary Record 125, 456–459. Mukhebi, A.W., Morzaria, S.P., Perry, B.D., Dolan, T.T. and Norval, R.A.I. (1990) Immunization against East Coast fever: production and delivery costs of the infection and treatment method. Preventive Veterinary Medicine 9, 207–219. Mukhebi, A.W., Curry, J.J., Perry, B., Mining, S.K., Delehanty, J.M., et al. (1992a) Targeting smallholder livestock against East Coast fever in Uasin Gishu District, Kenya. Poster presented to the Inter- national Seminar on Livestock Services for Smallholders, Yogyakarta, Indonesia. Mukhebi, A., Perry, B.D. and Kruska, R. (1992b) Estimated economics of theileriosis control in Africa. Preventive Veterinary Medicine 12, 73–85. Mukhebi, A.W., Curry, J., Perry, B.D., Reid, R., Ellis, J., et al. (1993) Comparative case studies of the economic costs of trypanosomiasis. In: ILRAD 1993 Annual Scientific Report. ILRAD, Nairobi, p. 105–106. Mukhebi, A., Chamboko, T., O’Callaghan, C., Peter, T., Kruska, R., et al. (1999) An assessment of the eco- nomic impact of heartwater (Cowdria ruminantium infection) and its control in Zimbabwe. Preventive Veterinary Medicine 39, 173–189. Ndung’u, L.W., Randolph, T.F., Coetzee, G., Krecek, R.C. and Perry, B.D. (2000) An economic assess- ment of current delivery pathways for the control of tick-borne diseases in Kenya. In: ISVEE 9: Pro- ceedings of the 9th Symposium of the International Society for Veterinary Epidemiology and Eco- nomics, 6–11 August, Breckinridge, Colorado. ISVEE, Fort Collins, Colorado, p. 333. 236 B. Perry et al. Ndung’u, L.W., Randolph, T.F., McDermott, J.J., Kiara, H.K. and Perry, B.D. (2003). Best-bet pathways for the delivery of East Coast fever vaccines to smallholder dairy systems in Kenya. In: ISVEE 10: Pro- ceedings of the 10th International Symposium for Veterinary Epidemiology and Economics, 17–21 November, Vina del Mar, Chile. ISVEE, Fort Collins, Colorado. p. 432. Norval, R.A.I., Perry, B.D., Gebreab, F. and Lessard, P. (1991) East Coast fever: a problem of the future for the horn of Africa? Preventive Veterinary Medicine 10, 163–172. Norval, R.A.I., Perry, B.D. and Hargreaves, S.K. (1992a) Tick and tick-borne disease control in Zimbabwe: What might the future hold. Zimbabwe Veterinary Journal 23, 1–15. Norval, R.A.I., Perry, B.D. and Young, A.S. (1992b) The Epidemiology of Theileriosis in Africa. Academic Press, Orlando, Florida. Norval, R.A., Perry, B.D., Meltzer, M.I., Kruska, R.L. and Booth, T.H. (1994) Factors affecting the distribu- tions of the ticks Amblyomma hebraeum and A. variegatum in Zimbabwe: implications of reduced acaricide usage. Experimental and Applied Acarology 18, 383–407. Nyangito, H.O., Richardson, J.W., Mukhebi, A.W., Mundy, D.S., Zimmel, P., et al. (1994) Whole farm eco- nomic analysis of East Coast fever immunisation strategies in Kilifi District, Kenya. Preventive Veter- inary Medicine 21, 215–235. Nyangito, H.O., Richardson, J.W., Mukhebi, A.W., Mundy, D.S., Zimmel, P. and Namken, J. (1995) Whole farm economic evaluation of East Coast fever immunization strategies on farms in Uasin Gishu Dis- trict of Kenya. Computers and Electronics in Agriculture 15, 19–33. Nyangito, H.O., Richardson, J.W., Mukhebi, A.W., Zimmel, P., Namken, J. and Berry, B.P. (1996) Whole farm simulation analysis of economic impacts of East Coast fever immunization strategies on mixed crop-livestock farms in Kenya. Agricultural Systems 51, 1–27. O’Callaghan, C.J. (1998) A study of the epidemiology of theileriosis of smallholder dairy farmers in Kiambu district, Kenya. PhD thesis, Faculty of Graduate Studies, University of Guelph, Guelph, Canada O’Callaghan, C., Medley, G., Peter, T. and Perry, B. (1998) Investigating the epidemiology of heartwater (Cow- dria ruminantium infection) by means of a transmission dynamics model. Parasitology 117, 49–61. O’Callaghan, C., Medley, G., Peter, T., Mahan, S. and Perry, B.D. (1999) Predicting the effect of vaccin- ation on the transmission dynamics of heartwater (Cowdria ruminantium infection). Preventive Veter- inary Medicine 42, 17–38. O’Callaghan, C., Medley, G., Kiara, H.K., McDermott, J.J., Musoke, A., et al. (2003) Quantitative analysis of dose–response effects in the transmission dynamics of Theileria parva and the impact of sporozoite- neutralising vaccines. In: ISVEE 10: Proceedings of the 10th International Symposium for Veterinary Epidemi- ology and Economics, 17–21 November, Vina del Mar, Chile. ISVEE, Fort Collins, Colorado, p. 872. Ochanda, H., Young, A.S., Medley, G.F. and Perry, B.D. (1998) Vector competence of 7 rhipicephalid tick stocks in transmitting 2 Theileria parva parasite stocks from Kenya and Zimbabwe. Parasitology 116, 539–545. Perry, B.D. (1992) The epidemiology of dog rabies and its control in eastern and southern Africa. In: King, A. (ed.) Proceedings of an International Meeting on Rabies in Eastern and Southern Africa, June, Lusaka, Zambia. Editions Fondation Marcel Merieux, Lyon, France, pp. 107–121. Perry, B.D. (1993) Dog ecology in eastern and southern Africa: implications for rabies control. Onderste- poort Journal of Veterinary Research 60, 429–436. Perry, B.D. (1995) Rabies control in the developing world: can further research help? Veterinary Record 137, 521–522. Perry, B. and Grace, D. (2009) The impacts of livestock diseases and their control on growth and development processes that are pro-poor. Philosophical Transactions of the Royal Society B: Biological Sciences 364, 2643–2655. Perry, B.D. and Hansen, J.W. (eds) (1994) Modelling Vector-borne and Other Parasitic Diseases. Proceed- ings of a Workshop organised by ILRAD in collaboration with FAO, 23–27 November 1992, Nairobi, Kenya. ILRAD, Nairobi, p. 369. Perry, B.D. and Hooton, N. (2007) Translating research outputs into policy and intervention outcomes: the art of knowing and influencing one’s ultimate clients. Animal health research: recent developments and future directions. Wellcome Trust Scientific Meeting, Wellcome Trust, London. Perry, B.D. and Randolph, T.F. (1999) Improving the assessment of the economic impact of parasitic diseases and of their control in production animals. Veterinary Parasitology 84, 145–168. Perry, B. and Sones, K. (2007a) Poverty reduction through animal health. Science 315, 333–334. Perry, B.D. and Sones, K.R. (2007b) Global roadmap for improving the tools to control foot-and-mouth dis- ease in endemic settings. Report of a workshop held at Agra, India, 29 November–1 December 2006, and subsequent roadmap outputs. ILRI, Nairobi. Veterinary Epidemiology at ILRAD and ILRI, 1987–2018 237 Perry, B.D. and Wandeler, A.I. (1993) The delivery of oral rabies vaccines to dogs: an African perspective. Onderstepoort Journal of Veterinary Research 60, 451–457. Perry, B.D. and Young, A.S. (1993) The naming game: the changing fortunes of East Coast fever and Theileria parva. Veterinary Record 133, 613–616. Perry, B.D. and Young, A.S. (1995) The past and future roles of epidemiology and economics in the control of tick-borne diseases of livestock in Africa: the case of theileriosis. Preventive Veterinary Medicine 25, 107–120. Perry, B.D., Lessard, P., Norval, R.A.I., Kundert, K. and Kruska, R. (1990) Climate, vegetation and the distribution of Rhipicephalus appendiculatus in Africa. Parasitology Today 6, 100–104. Perry, B.D., Kruska, R., Lessard, P., Norval, R. and Kundert, K. (1991) Estimating the distribution and abun- dance of Rhipicephalus appendiculatus in Africa. Preventive Veterinary Medicine 11, 261–268. Perry, B.D., Kruska, R. and Reid, R.S. (1994) The development and use of geographic information sys- tems to assist trypanosomiasis control. FAO Animal Production and Health Papers. FAO, Rome. Perry, B., Kyendo, T., Mbugua, S., Price, I. and Varma, S. (1995) Increasing rabies vaccination coverage in urban dog populations of high human population density suburbs: a case study in Nairobi, Kenya. Preventive Veterinary Medicine 22, 137–142. Perry, B.D., Chamboko, T., Mahan, S.M., Medley, G.F., Minjauw, B., et al. (1998) The economics of inte- grated tick and tick-borne disease control on commercial farms in Zimbabwe. Zimbabwe Veterinary Journal 29, 21–29. Perry, B.D., Gleeson, L., Khounsey, S., Bounma, P. and Blacksell, S. (2002a) The dynamics and impact of foot and mouth disease in smallholder farming systems in South-East Asia: a case study in Laos. Revue Scientifique et Technique 21, 663–670. Perry, B.D., Randolph, T.F., McDermott, J.J., Sones, K.R. and Thornton, P.K. (2002b) Investing in animal health research to alleviate poverty. ILRI, Nairobi. Perry, B.D., Randolph, T.F., Ashley, S., Chimedza, R., Forman, T., et al. (2003) The impact and pov- erty reduction implications of foot and mouth disease control in southern Africa with special refer- ence to Zimbabwe. ILRI, Nairobi. Perry, B.D., Camus, E., Ellis, T., Mbugua, H., Moore, R., et al. (2010) Second real time global evaluation of FAO’s work on highly pathogenic avian influenza. FAO, Rome. Perry, B.D., Randolph, T.F., Bett, B.K., Grace, D., Globig, A., et al. (2011) Independent impact assessment of the World Bank-funded Nigeria Avian Influenza Control and Human Pandemic Preparedness and Response Project (NAICP). ILRI, Nairobi. Peter, T., Perry, B.D., O’Callaghan, C., Medley, G., Mlambo, G., et al. (1999) Prevalence of Cowdria ruminantium infection in Amblyomma hebraeum ticks from heartwater-endemic areas of Zimbabwe. Epidemiology and Infection 123, 309–316. Peter, T., Perry, B.D., O’Callaghan, C., Medley, G., Shumba, W., et al. (1998) Distributions of the vectors of heartwater, Amblyomma hebraeum and Amblyomma variegatum (Acari: Ixodidae), in Zimbabwe. Experimental and Applied Acarology 22, 725–740. Randolph, S.E. (1993) Climate, satellite imagery and the seasonal abundance of the tick Rhipicepha- lus appendiculatus in southern Africa: a new perspective. Medical and Veterinary Entomology 7, 243–258. Randolph, S.E. and Rogers, D.J. (1997) A generic population model for the African tick Rhipicephalus ap- pendiculatus. Parasitology 115, 265–279. Randolph, T.F., Perry, B.D., Benigno, C.C., Santos, I.J., Agbayani, A.L., et al. (2002) The economic impact of foot and mouth disease control and eradication in the Philippines. Revue Scientifique et Technique 21, 645–659. Randolph, T.F., Saatkamp, H. and Perry, B.D. (2003) A short history of animal health economics. In: ISVEE 10: Proceedings of the 10th International Symposium for Veterinary Epidemiology and Economics, 17–21 November, Vina del Mar, Chile. ISVEE, Fort Collins, Colorado, p. 827. Randolph, T.F., Morrison, J.A. and Poulton, C. (2005) Evaluating equity impacts of animal disease con- trol: the case of foot and mouth disease in Zimbabwe. Review of Agricultural Economics 27, 465–472. Reid, R., Kruska, R., Perry, B., Ellis, J. and Wilson, C. (1995) Environmental impacts of trypanosomiasis control through land-use change: conceptual model, approach and preliminary results. In: Proceed- ings of the 22nd Meeting of the International Scientific Council for Trypanosomiasis Research and Control, 25–29 October 1993, Kampala, Uganda. OAU/STRC, Nairobi, pp. 235–243. Reid, R.S., Kruska, R.L., Deichmann, U., Thornton, P.K. and Leak, S.G.A. (2000) Human population growth and the extinction of the tsetse fly. Agriculture, Ecosystems and Environment 77, 227–236. 238 B. Perry et al. Rich, K. and Perry, B.D. (2009) The impact of changing global animal health trading standards on market access for livestock products by developing countries; assessment of commodity-based trade. Commissioned report to the Department for International Development (DFID) of the Government of the UK. Rich, K.M. and Perry, B.D. (2011) Whither commodity‐based trade? Development Policy Review 29, 331–357. Rich, K.M. and Wanyoike, F. (2010) An assessment of the regional and national socio-economic impacts of the 2007 Rift Valley fever outbreak in Kenya. American Journal of Tropical Medicine and Hygiene 83, 52–57. Rich, K.M., Perry, B.D., Kaitibie, S., Gobana, M. and Tewolde, N. (2008) Enabling livestock product ex- ports from Ethiopia: understanding the costs, sustainability and poverty reduction implications of sani- tary and phytosanitary compliance. Final report for the Texas A&M University Sanitary and Phytosani- tary Livestock and Meat Marketing Program, Texas Agricultural Experiment Station, Texas. Robinson, T.P., Cecchi, G., Wint, G., Mattioli, R. and Shaw, A. (2014a) Benefits of trypanosomosis control in the Horn of Africa. In: Sebastian, K. (ed.) Atlas of African Agriculture Research and Development: Revealing Agriculture’s place in Africa. IFPRI, Washington, DC, pp. 62–63. Robinson, T.P., Wint, W.R.G., Conchedda, G., van Boeckel, T.P., Ercoli, V., et al. (2014b) Mapping the global distribution of livestock. PLoS One 9, 1371–1371. Roeder, P. and Rich, K.M. (2009) Conquering the cattle plague: the global effort to eradicate rinderpest. In: Spielman, D.J. and Pandya-Lorch, R. (eds) Millions Fed: Proven Successes in Agricultural Develop- ment. IFPRI, Washington, DC, pp. 109–116. Rogers, D.J., Hay, S.I. and Packer, M.J. (1996) Predicting the distribution of tsetse flies in West Africa using temporal fourier processed meteorological satellite data. Annals of Tropical Medicine and Para- sitology 90, 225–241. Romero, J.R., Villamil, L.C., Perry, B.D., Randolph, T.F., Vera, V.J. and Ramírez, G.C. (2001) Valoración del impacto económico del programa de control y erradicación de la fiebre aftosa en Colombia. Revista Colombiana de Ciencias Pecuarias Supplement 14. Rushton, J. (2009) The Economics of Animal Health and Production. CAB International, Wallingford, UK. Schukken, Y., van Schaik, G., McDermott, J.J., Rowlands, G., Nagda, S., et al. (2004) Transition models to assess risk factors for new and persistent trypanosome infections in cattle – analysis of longitudinal data from the Ghibe Valley, southwest Ethiopia. Journal of Parasitology 90, 1279–1288. Shaw, A.P. (1992) Development of a methodology for assessing the losses due to trypanosomiasis in Africa. Consultancy report. ILRAD, Andover, UK. Shaw, A.P. (2009) Assessing the economics of animal trypanosomosis in Africa – history and current perspectives. Onderstepoort Journal of Veterinary Research 76, 27–32. Shaw, A.P., Randolph, T.F., Heffernan, C., Rushton, J. and Perry, B.D. (2003) A framework for prioritizing animal health constraints in developing countries according to their impact on the poor. In: ISVEE 10: Proceedings of the 10th International Symposium for Veterinary Epidemiology and Economics, 17–21 November, Vina del Mar, Chile. ISVEE, Fort Collins, Colorado, p. 431. Snow, W. and Rawlings, P. (1999) Methods for the rapid appraisal of African animal trypanosomosis in the Gambia. Preventive Veterinary Medicine 42, 67–86. Stevens, K., de Glanville, W., Costard, S., Métras, R., Theuri, W., et al. (2009) Risk mapping for HPAI H5N1 in Africa-Improving surveillance for bird flu: initial bird flu risk map reports. ILRI, Nairobi. Talaki, E., Sidibé, I., Diall, O., Affognon, H., Grace, D., et al. (2009) Variations saisonnières et facteurs de risques des trypanosomoses animales dans un contexte de chimiorésistance dans la zone de Sikas- so au Mali. Bulletin of Animal Health and Production in Africa 57, 149–160. Tambi, E., Maina, O.W., Mukhebi, A. and Randolph, T.F. (1999) Economic impact assessment of rinder- pest control in Africa. Revue Scientifique et Technique 18, 458–477. Thornton, P.K., Kruska, R.L., Henninger, N., Kristjanson, P.M., Reid, R.S., et al. (2002) Mapping poverty and livestock in the developing world. ILRI, Nairobi. Wanyangu, S.W., Kiara, H., Randolph, T.F., Leneman, J.M., Okuthe, O.S., et al. (2000) The infection and treatment method for East Coast fever immunization: assessing its impact in Kenya. In: ISVEE 9: Proceedings of the 9th Symposium of the International Society for Veterinary Epidemiology and Economics, 6–11 August, Breckinridge, Colorado. ISVEE, Fort Collins, Colorado, p. 339. Welburn, S.C., Fèvre, E.M., Coleman, P.G., Odiit, M. and Maudlin, I. (2001) Sleeping sickness: a tale of two diseases. Trends in Parasitology 17, 19–24. Wilson, C.J., Reid, R.S., Stanton, N.L. and Perry, B.D. (1997) Effects of land use and tsetse fly control on bird species richness in southwestern Ethiopia. Conservation Biology 11, 435–447. Wint, W. and Rogers, D. (2000) Consultants’ report: predicted distribution of tsetse in Africa. FAO, Rome.