Epidemiology of Brucellosis in Smallholder Farming Systems in Pakistan

Brucellosis is one of the most widespread zoonoses and is endemic in many developing countries. Brucellosis can be considered as an animal disease, where humans are accidental hosts. The disease is neglected in many developing countries including Pakistan and remains a challenge due to lack of public awareness, consistent preventive measures and little effort to control the disease. In Pakistan, brucellosis epidemiology has been investigated minimally and only a few studies are available which report disease prevalence on large and commercial farms. However, these estimates are not generalizable for the predominant system in Pakistan, which is a smallholder production system involving large rural farming communities. The overall goal of this PhD research project was to advance the epidemiological understanding of bovine brucellosis in smallholder farms in Pakistan. The first objective was to identify the best diagnostic test for local field conditions, using this, and estimate disease prevalence and conduct a risk factor investigation in smallholder farms in Pakistan. The second objective was to assess and document the knowledge, attitudes and practices of smallholder farmers, and also to evaluate the drivers for overall perceptions regarding zoonotic diseases in Pakistan.

The diagnostic test evaluation revealed that the test with the highest sensitivity (Se) was the competitive ELISA (C-ELISA), with a range from 76.3% to 81.4%. The Rose-Bengal test (RBT) was found to have the highest specificity (Sp) (99.1–99.4%) of the tests explored. The highest estimated herd-level prevalence, 45%, was found in Jhelum district and the lowest, 1.1%, in Thatta district. The results of this study identified some discrepancies between the published literature on the Se of these tests, especially for RBT. It appears that RBT has lower Se and higher Sp when used in the field conditions of the present study compared with some other published studies. Consequently, it is recommended that none of the evaluated tests should be performed in isolation for the diagnosis of bovine brucellosis in the field conditions of Pakistan; however, the application use of RBT and C-ELISA in parallel, which return optimal Se and Sp, is warranted. The risk factor investigation demonstrated that the herds with a history of last trimester abortion were more likely to be positive than herds without such history (OR = 2.06, 95% CI, 1.09-3.89), providing validation of our findings and identifying that clinical disease is occurring in this region. It was also identified that herds with five to eight buffalo (odds ratio: OR = 3.80, 95% CI, 1.69-8.49), and those with more than eight buffalo (OR = 3.81, 95% CI, 1.51- 9.58) were more likely to be positive for Brucella than those with less (one to two and three to four) buffaloes present in the herd.

The investigation of knowledge, attitudes and practices, demonstrated that almost all farmers (97%) were not aware of the modes of transmission of brucellosis. The majority (66%) of the farmers’ families were reported to consume raw milk and its products, live in shared housing with animals (49%) and not covering hand cuts during contact with animals (74%). All farmers performed at least one risky practice for brucellosis transmission from animal to human on a regular basis. The farmers with no formal education and those who had not heard of the disease displayed greater risky behaviour. The subsequent participatory epidemiological investigation revealed that the farmers were not concerned about zoonotic diseases, with haemorrhagic septicaemia, foot and mouth disease and mastitis being considered the three most important diseases of concern, with economic cost being the main reason explaining this concern. Despite farmers performing practices that pose a risk of zoonotic disease, approximately half were aware that these practices posed a risk of transmission. The drivers for continuation of these risky practices included family tradition, lack of resources and past experience. In the event of an animal health problem, farmers would self-treat or contact senior farmers in the village, with government veterinary assistants only contacted sometimes when the issue cannot be resolved by local options. Similarly, in relation to human health problems, half of the participants reported to use self-treatment, with 58% indicating that they would use the village dispensary. Access to animal and human health stakeholders and to resources was the most important driver of farmer actions.

To conclude, the outcomes of this thesis indicate that none of the three evaluated diagnostic tests are recommended to be used as standalone tests, and based on this research findings RBT and C-ELISA applied in parallel yield best results for Pakistani field conditions. The research also identified regions with high, medium, low disease prevalence that support the need for customised targeted intervention control programs. In addition, it was found that there is poor understanding of the disease, presence of multiple risky practices on farm and at the household, low awareness of zoonotic diseases and a clear disconnection between risk perception and practices conducted, supporting the need of the development of targeted extension programs delivered by trusted stakeholders.