Background: Lack of reliable data on the aetiology of livestock diseases, especially in Africa, is a major factor constraining the design of effective livestock health interventions to improve livelihoods, food security and public health. Rationale: Livestock abortion is an important disease syndrome that affects productivity and livestock economies, and poses risks to public health. Worldwide, several pathogens are associated with livestock abortions but across Africa livestock disease surveillance data rarely include information from abortion events, little is known about the aetiology and impacts of livestock abortions, and data are not available to inform prioritisation of disease interventions. Methods: This paper describes outcomes from a surveillance platform that was set up in 15 wards in northern Tanzania spanning pastoral, agropastoral and smallholder agro-ecological systems to investigate the causes and impacts of livestock abortion. Abortion cases were reported by farmers to Ministry of Livestock and Fisheries livestock field officers and on to investigation teams. The abortion was eligible for inclusion if the research field team or livestock field officer could attend within 72 hours of the event occurring. If so, a field investigation was carried out to collect diagnostic samples from aborted materials and aborting dams and questionnaire data on herd/flock management. Laboratory diagnostic tests were carried out for a range of abortigenic pathogens, and laboratory and questionnaire data analysed to investigate a) determinants of attribution; b) patterns of investigated events, including species and breed affected, history of previous abortion and recent stressful events, and the seasonality of cases; c) determinants of reporting, investigation and attribution; (d) cases in which zoonotic pathogens were detection. Results: Over a two-year period (2017-2019), 215 abortion events in cattle (n=71), sheep (n=44) and goats (n=100) were investigated from 13 of the 15 wards. The number of investigated cases varied widely across wards, with three of the 15 field officers (20%) reporting 70% of investigated cases. Aetiological attribution, achieved for 19.5% of cases through PCR-based diagnostics, was significantly affected by delays in field investigation. Histopathology proved less useful than PCR diagnostics due to rapid deterioration of field samples. Vaginal swabs from aborting dams provided a practical and sensitive source of diagnostic material for pathogen detection. Discussion / Conclusion: Livestock abortion surveillance, even at a small scale and when capturing only a small proportion of events, can generate valuable information on causes of zoonotic disease outbreaks, livestock reproductive losses and can identify important pathogens that are not easily captured through other forms of livestock disease surveillance. The study demonstrated the feasibility of establishing an effective reporting and investigation system that could be implemented across a range of settings, including remote rural areas, with engagement of community-based field officers, and establishment of practical sample collection protocols and molecular diagnostic platforms all being important elements of effectiveness.
