Ethno-Epidemiology Of Febrile Illness At The Human Livestock Interface In Uganda

Abstract

Undifferentiated febrile illness is a leading cause of morbidity and mortality in many resource-limited countries. Effective management of febrile illness in areas of limited resources remains a challenge due to delayed recognition and reporting in addition to lack of adequate diagnostic infrastructure. There is increasing evidence linking Non-Malarial Febrile Illness (NMFI) to contact with animals. However, the syndemic human-livestock interactions that drive these illnesses are complex and poorly understood. The overall goal of this dissertation is to improve our understanding of the role that the hu-man-livestock relationship plays in the occurrence of NMFIs in humans. Understanding the cultural context of the human-livestock relationship in Western Uganda, and under-standing its relationship to NMFI occurrence, could help guide development of novel infectious disease prevention and management strategies consequently reducing infectious disease spillover events in areas that lack diagnostic infrastructure. Toward this end, we employed a sequential mixed method ethno-epidemiological approach, comparing and contrasting two culturally different livestock management systems, agro pastoral and pastoral, present in a cultural mosaic in Western Uganda (Hoima and Kasese Districts). Specifically we hypothesize that there are culturally relevant NMFI “syndromic” case definitions that correlate with specific socio-cultural factors associated with livestock ownership and management, which drive NMFI prevalence, and disproportionately contribute to NMFI clustering along the human-livestock inter-face. The study involved an ethnographic study, two cross-sectional surveys and a Brucella sero-survey. It revealed that perception of illness and associated risk factors was heavily influenced by the predominant livelihood activity of the community. The ordinal odds of febrile illness was positively associated with (a) milking, (b) use of bulls for breeding and (c) participating in fishing/hunting livelihood activities; while control of vectors was negatively associated with the ordinal odds of febrile illness (P<0.05). We also noted that among the Brucella positive households, nearly half reported subjective fever in the preceding year. Among the seropositive individuals, 44% reported diagnosis of malaria to explain their fevers and 31% reported no confirmatory diagnosis. Apparent Brucella herd prevalence in cattle was 8.6% (3.7-18.6; 95% CI), while animal level sero-prevalence was 7.1% (3.3-14.6; 95% CI). None of the goats tested positive.