The burden of zoonoses in Nepal

Chapter 1 provided the background for this thesis. The main goal of public health policy is to promote, enhance and protect population health. This requires information on the health status of the population, often referred to as the “burden of disease”. Population health is a multifactorial phenomenon with many facets. As a result, the disease burden of a population can be described by a variety of indicators. As current health policy requires a global overview of public health, combining morbidity and mortality and taking into account health-related quality of life, so-called summary measures of population health (SMPH) are gaining wider importance. Driven by the influential Global Burden of Disease projects initiated in the early 1990s, the Disability-Adjusted Life Year (DALY) has become the dominant SMPH. The DALY is a health gap measure, reflecting the number of healthy life years lost due to disease and death. In the DALY philosophy, every person is born with a certain number of life years potentially lived in optimal health. People may lose these healthy life years through living with illness and/or through dying before a reference life expectancy. The importance of burden of disease estimates for health policy in Nepal becomes evident from recent policy documents and recommendations. The Second Long-Term Health Plan 1997–2017 was the first document to recognize the importance of prioritizing health sector needs, motivated by the scarce human, financial and physical resources available. Notwithstanding the importance of disease burden estimates, Nepalese DALYs are scarce and not rooted in local data. Especially for zoonotic and other neglected diseases, this may lead to a vicious cycle of indifference, under-recognition and under-funding. Chapter 2 introduced the rationale and objectives of this thesis. The main objective of this thesis was to unravel the burden of zoonoses in Nepal and to quantify this burden using the DALY metric. To achieve this goal, we contributed to a further standardization of the DALY metric. Although the philosophical and methodological aspects of the DALY calculation have been described (and debated) in great detail, the steps preceding the actual calculation remained less well documented. In Chapter 3, we therefore proposed a stepwise approach for conducting a DALY-based disease burden study, consisting of the following five consecutive steps: Study population definition; Disease mo del definition; Data collection; Data adjustment; and DALY calculation. Nearly every DALY estimation is subject to data uncertainty and modelling choices. The resulting DALY estimate is therefore hardly ever a single, fixed value, defined with perfect accuracy and precision. In Chapter 4, we studied sources of uncertainty inherent to DALY calculations through a systematic review of DALY-based disease burden studies. Of the 228 studies published between 1994 and 2013, only 105 (46%) had performed some sort of uncertainty quantification. Identifying, quantifying and analysing uncertainties should become a standard part of DALY calculations. We recommend probabilistic sensitivity analysis for quantifying parameter uncertainty and scenario analyses for quantifying model and methodological uncertainties. To our knowledge, there are no standardized tools available for stochastic DALY calculation. We therefore designed a free and open-source tool for calculating DALYs that allows for the incorporation of input uncertainty and the computation of DALY credibility intervals through Monte Carlo simulations. Chapter 5 provides an overview of the functionalities of this tool, called the DALY Calculator. Further work is needed to increase the flexibility and user-friendliness of the tool. Based on the guidelines and tools developed in these first chapters, we then went on to accomplish the main objective of the thesis. In Chapter 6, we reviewed the occurrence and burden of parasitic zoonoses in Nepal. Between 2000 and 2012, the highest annual burden was imposed by neurocysticercosis and congenital toxoplasmosis, followed by cystic echinococcosis. Nepal is endemic for several other parasitic zoonoses, but these probably have a much lower population burden. We identified several critical data gaps and highlighted the need for enhanced surveillance of the identified endemic parasitic zoonoses. In addition, we reviewed the epidemiology, burden and control of rabies in Nepal in Chapter 7. Limited data indicate that rabies still is a major zoonosis in Nepal. However, more and better data are needed, especially from rural areas, to estimate the true burden of animal and human rabies. The current control of rabies is hampered by insufficient vaccine availability across the country and limited collaboration, both within the country and within the region. To overcome these hurdles, high-level political commitment is required. Making rabies the model zoonosis for successful control may be a powerful step towards achieving this. In Chapter 8, finally, we placed our findings in a broader perspective, discussed their main limitations and presented avenues for future research. Our work contributed to identifying sources of methodological variability in DALY estimations, but did not manage to resolve the lack of comparability across studies. The development of a checklist of assumptions, based on our work, could increase methodological transparency in DALY-based disease burden studies. Further methodological research needs to focus on the problem of multimorbidity and the extrapolation of health statistics. Our work also contributed to an improved understanding of the epidemiology and burden of zoonoses in Nepal and highlighted the importance of rabies, both at the population-level and the patient-level. However, our work has been limited by the fact that only the health burden was quantified, and not the economic burden. Furthermore, by focusing on parasitic zoonoses and rabies, we covered only one part of the large spectrum of zoonotic diseases. Further evidence on the burden of foodborne zoonoses in Nepal is available from the WHO Foodborne Disease Burden Epidemiology Reference Group. Combining all available evidence shows that the zoonoses with the highest disease burden in Nepal are non-typhoidal salmonellosis, campylobacteriosis, toxoplasmosis, cysticercosis and rabies. The ultimate aim of disease burden estimates is to inform decision makers on setting the right research and control priorities. However, we have no evidence that our work managed to inform decision makers. Future, similar projects should therefore emphasize capacity building and knowledge transfer. Notwithstanding the possible disconnect between burden estimates and policy, we do believe that it is of paramount importance to continue generating disease burden estimates. Only then can we gain the knowledge that is required to take the right actions.