Rabies, a disease caused by a negative sense single-stranded RNA virus in the family Rhabdoviridae, is an acute viral infection of the brain that results in high mortality for humans and other warm-blooded animals (1, 2). The disease occurs when saliva or brain tissue from an infected animal comes into direct contact with mucus membranes or broken skin (generally via a bite). Despite vaccination campaigns and post-exposure prophylaxis, rabies continues to kill approximately 50,000-61,000 people globally each year (1, 3). Over 95% of these fatal rabies cases occur in Asia and Africa (4). For example, rabies claimed approximately 21,000-25,000 lives in Africa during 2015 (5). In Uganda, a total of 14,865 annual dog bites were reported in 2021 (6) and rabies claimed approximately 36 lives per year (7).

The global estimates for rabies incidence and prevalence in Africa are not well documented. A meta-analysis by Wobessi et al. reports misrepresented incidence of rabies in Africa for exposed humans, dogs and other species as 83.4%, 44.1% and 33.8%, respectively, and the prevalence as 33.8%, 19.8% and 3.6% (8). In Uganda, the prevalence has been over-estimated to be 78.9%, 66.7% and 84.6% by Omodo et al (9) in dogs, jackals and livestock (goats, sheep, and cows), respectively (9). The exact statistics on rabies prevalence among humans in Uganda is lacking.

Rabies prevention and control programs, such as mass dog vaccination, sterilization, and post-exposure prophylaxis (PEP), have been performed seasonally in Uganda. However, due to limited funding, various endemic parts of Uganda do not receive these programs. Consequently, this leads to sub-optimal coverage according to the GARC, OIE, FAO recommendations and ensures that transmission of the virus continues in rural parts of Uganda. Limited PEP availability, awareness, and access makes it even more difficult to provide timely treatment for those that are infected. Only two out of ten people bitten by a suspected rabid animal have access to PEP in Uganda. This is due to many factors, including cost and long distances to hospitals. For example, PEP is approximately 40 USD in Uganda, yet citizens live on less than 1 USD per day. Also, it takes an average of three hours to reach the nearest hospital (Atutur) from Pian sub-counties (i.e. Kocheka).

People living in Pian Upe are largely nomadic pastoralists with poor access to health facilities. Practices and beliefs of these pastoralists, such as grazing cows with dogs, fetching firewood in the game reserve (GR), and frequent hunting (generally twice a week) predispose them and their animals (dogs and livestock) to close encounters with wildlife.

Approximately 99% of human rabies cases are transmitted by rabid domestic dogs (2), while a relatively small proportion of human cases are caused by wildlife. For example, a study in South Africa revealed that black-backed jackals (Canis mesomelas) contributed to 0.14% of rabies cases in the country during 2016 (10). Anecdotal findings in Bukedea region show frequent jackal and dog bites in humans, especially for communities near Pian Upe GR. This may be a potential risk for Rabies spillover, leading to epidemics in communities surrounding GRs and national parks. The aim of this study was to determine community reported incidence and mortality estimates of rabies at the wildlife-dog-livestock-human interface of Pian Upe GR in Bukedea, Eastern Uganda. We hypothesized that households near the GR experience more rabies cases than their counterparts situated more than 14km away.

In this study, we hypothesized that rabies risk would be higher in communities near wildlife protected areas (GR). Our findings support this hypothesis, as households closer to Pian Upe GR had higher rabies incidence, possibly due to more frequent interactions between livestock, dogs and wildlife. Rabies in wildlife can be a threat to humans and domestic animals due to human demography, animal translocations, and environmental alterations; however, efforts to control wildlife associated rabies is largely lacking in Uganda (12). The communally owned grazing lands in Bukedea (Aaro) act as contact points between livestock, dogs, humans, and wildlife. Most of these Aaro´s are close to the GR, giving wildlife (especially jackals) a greater chance of preying on livestock, facilitating close contact between dogs, jackals, and livestock. Anecdotal reports from community members and our own observations during data collection confirmed that jackals were frequently seen near livestock in the Aaro and near livestock tethered at home (at least 3-4 times a week), especially along the border between the GR and human settlements. Other studies show similar observations. For example, in the Mymensingh district of Bangladesh, jackals and dogs have been implicated in the transmission of rabies to domestic ruminants (13). However, Bangladesh has a far higher burden of rabies than Uganda. This is likely due to the larger population, with approximately 165.7 million compared to Uganada’s 49.2 million people (14), and the location, as it is situated between China and India, and is ranked first and second globally in rabies burden (15). The similarity concerning jackal and dog distribution in the two countries lies in somewhat similar climatic conditions and forest cover that acts as habitat for these canids (16).

In our study, we found a mean incidence of 8.9% for rabies cases among humans and dogs and 6.6% in livestock. This is much higher than the reported national rabies incidence in Uganda (0.04% (6). This discrepancy could be due to the risk-based sampling that we carried out as opposed to population wide sampling done from national epidemiological data (6). The large discrepancy between the two numbers implies a neglected rabies risk at the interface between wildlife, dogs, livestock, and humans. The incidence is less than 10%, which may be regarded as a low risk. However, as untreated rabies cases are fatal, every single case is of high public and economic importance for the household.

A study by Millán et al. 2013 (17) reported a similar sero-prevalence of rabies in dogs (12.7-28.6 %) in Queen Elizabeth national park (17). The two protected areas (Pian Upe GR and Queen Elizabeth national park) host the largest numbers of canines in Uganda, increasing the possibility for rabies spillover through bites between the wild canines and dogs, along with canine-human interactions (18). Furthermore, a study in Zimbabwe reported rabies transmission between jackals (Canis adustus and Canis mesomelas) and dogs at the wildlife-human interface (19). Jackals were the most implicated wildlife species, accounting for 91% of wildlife rabies cases (20). The unmet need for rabies control in wildlife is likely to make the rabies elimination agenda unattainable by 2030 in some parts of Africa, especially those that surround wildlife protected areas.

In the current study, mortality ranged between 44-56 cases per 1000 persons and 50-98 cases per 1000 livestock. These findings, especially those for humans, are higher than estimates across Uganda (16 deaths per 1000 persons per year (21). However, Nyasulu et al. (21) used Uganda rabies weekly surveillance data, which is not representative for some communities that are in rural and hard-to-reach areas. We also think that the elevated figures of mortality are due to risk-based sampling carried out in our study, since we expect more rabies cases at the wildlife-dog-livestock-human interface. Furthermore, our findings for livestock did not agree with the study by Omodo et al. (9), which reported lower incidence of rabies in cattle and goats using laboratory specimens collected between 2015-2022 in Uganda. The discrepancy could have been brought about by differences in geographical areas, given that they picked scattered regions in the country where the risk of rabies varied greatly (9).

In this study, 3% (n=9) of households reported co-occurrence of rabies in humans, livestock, and dogs. This clearly indicates that transmission is linked between all three species in these communities. Culturally, in this pastoralist setting, dogs are often used for herding in Aaro, where interactions with wildlife (especially jackals) occur frequently. This fact is supported by similar observations in selected districts in Tanzania, such as Ulanga and Kilombero, where jackals were rampantly biting livestock and human beings (22). We also report co-occurrence of rabies between dogs and humans, but not between dogs and livestock. This suggests that livestock can become infected from sources other than dogs, which points the suspicion to the black backed jackal, which is rampant in Pian Upe GR (23). This assertion is further supported by the co-occurrence of mortality between wildlife and livestock, and between wildlife and humans. Reporting of rabies cases in livestock is not often emphasized in Uganda, but our study suggests that there is a significant need for this, especially among pastoralist communities. Rabies has been previously reported in cattle in Ethiopia (24). Our study emphasizes that spillover events are most likely to hinder the success of the global dog-mediated rabies elimination target for 2030 (25).

In our study, primary level of education, grazing, and fetching firewood in the GR were associated with rabies cases in households near to the GR. Primary level of education may only be associated with less awareness and knowledge about rabies (26). Grazing and fetching firewood in GR might be associated with rabies cases since these practices increase the possibility for wildlife-dog-human interaction (27). Additionally, those who fetch firewood and allow their animals to graze from the GR are by default staying near the GR. Hence, they find themselves further away from the hospital and have less access to timely post exposure prophylaxis. The low socio-economic status of these households also reduces their ability to afford PEP (28). There are no similar studies in Uganda, with the exception of a study in Wakiso and Kampala districts by Kisaka and coworkers (2020), where they found that secondary level of education was associated with compliance to rabies pre-clinical control guidelines (29). There are volunteer groups, like the Uganda veterinary student outreach program, that go out to communities to implement rabies education efforts in primary and secondary schools (30), but these programs seem inadequate due to poor funding and political priority deficiencies.

One limitation for our study is that it lacks generalizability for households without dogs and livestock, given that these were criteria for household selection. However, having conducted a risk-based sampling, our findings may be applicable to households and communities neighboring wildlife protected areas in Uganda, and Africa in general. We also had to rely on reports given by household heads for rabies prevalence data, rather than using laboratory-confirmed cases. This approach may suffer from recall bias along with differences in knowledge of rabies symptoms between households. Nevertheless, rabies is a very recognizable and devastating disease, which may be well recorded in people’s memory. Thus, our estimates of rabies burden depict a proxy measure that warrants further investigation.

The risk of rabies at the wildlife-dog-livestock-human interface in Uganda is a neglected health threat. Both rabies incidence and rabies mortality among humans, dogs and livestock increased as proximity to the Pian Upe Game Reserve increased.

Those with primary level of education, fetching firewood in the GR and grazing their livestock in the GR had higher risk for rabies. This implies that communities of Bukedea need to exercise more caution on rabies prevention strategies than the rest of neighboring communities. More studies are necessary to carry out rabies surveillance in wildlife cannids of Pian Upe game reserve. Furthermore, increased sensitization for communities neighboring wildlife reserve and national parks is highly urgent to address rabies prevention and control.

The raw data supporting the conclusions of this article will be made available by the authors, without undue reservation.

Ethical clearance was obtained from the Makerere University School of Public Health Research Ethics Committee (SPHREC-2023-509). Permission to collect data was obtained from Bukedea district Chief Administrative officer (CAO) through Bukedea District Veterinary Officer (DVO). A written informed consent for voluntary participation was fully explained in mother language and English to the participant before data collection. Interpretation to a local language (Ateso) was carried out when the respondent in a household was not well conversant with English.

CA: Conceptualization, Data curation, Formal analysis, Methodology, Validation, Writing – original draft, Writing – review & editing. JOA: Conceptualization, Supervision, Visualization, Writing – review & editing, Writing – original draft. MTa: Data curation, Project administration, Validation, Writing – review & editing. PS: Conceptualization, Visualization, Writing – review & editing. MM: Conceptualization, Data curation, Methodology, Visualization, Writing – review & editing. CK: Conceptualization, Methodology, Supervision, Writing – review & editing. OT: Conceptualization, Supervision, Writing – review & editing. SS: Conceptualization, Supervision, Visualization, Writing – review & editing. FM: Conceptualization, Visualization, Writing – review & editing. KB: Visualization-Presentation of published work, Writing – review and editing. MTr: Conceptualization, Writing – original draft, Methodology, thorough Supervision, Visualization, Writing – review & editing.