Surveillance of diseases in Kenya and elsewhere in East Africa is currently carried out by both human and animal health sectors. However, a recent evaluation highlighted the lack of integration between these sectors, leading to disease under-reporting and inefficiencies. This project aimed to develop an integrated and cost-effective surveillance and reporting system for 15 zoonotic diseases piloted in the counties of Bungoma, Busia, and Kakamega in western Kenya. Specifically, in this paper we describe the operational aspects of such a surveillance system. Interviews were carried out with key informants, and this was followed by field visits to identify sentinel sites and liaise with relevant stakeholders. Based on this information, a sampling strategy comprising 12 sentinel sites, 4 in each county, was developed. Each sentinel site comprised of a livestock market, 1–2 neighboring slaughter houses/slabs, and a hospital in the vicinity; each of the 12 sites, comprising 12 × 3 = 36 sampling locations, was visited every 4 weeks for 20 cycles. At each site, animal or patient sampling included a clinical examination and collection of blood, feces, and nasal swabs; in slaughtered animals, mesenteric lymph nodes, hydatid cysts, and flukes were also collected. At the end of each field visit, data on staff involved and challenges encountered were recorded, while biological samples were processed and tested for 15 zoonotic diseases in the field laboratory in Busia, Kenya. Public engagement sessions were held at each sentinel site to share preliminary results and provide feedback to both stakeholders and study participants. A livestock market visit lasted just over 3 h, and the most common challenge was the frequent refusals of animal owners to participate in the study. At the slaughterhouses, visits lasted just under 4 h, and challenges included poorly engaged meat inspectors or slaughter processes that were too quick for sampling. Finally, the hospital visits lasted around 4 h, and the most frequent challenges included low patients turn-out, frequent staff turn-over leading to poor institutional memory, and difficulty in obtaining patient stool samples. Our experiences have highlighted the importance of engaging with local stakeholders in the field, while also providing timely feedback through public engagement sessions, to ensure on-going compliance.

Quantitative evidence for the risk of zoonoses and the spread of antimicrobial resistance remains lacking. Here, as part of the UrbanZoo project, we sampled Escherichia coli from humans, livestock and peri-domestic wildlife in 99 households across Nairobi, Kenya, to investigate its distribution among host species in this rapidly developing urban landscape. We performed whole-genome sequencing of 1,338 E. coli isolates and found that the diversity and sharing patterns of E. coli were heavily structured by household and strongly shaped by host type. We also found evidence for inter-household and inter-host sharing and, importantly, between humans and animals, although this occurs much less frequently. Resistome similarity was differently distributed across host and household, consistent with being driven by shared exposure to antimicrobials. Our results indicate that a large, epidemiologically structured sampling framework combined with WGS is needed to uncover strain-sharing events among different host populations in complex environments and the major contributing pathways that could ultimately drive the emergence of zoonoses and the spread of antimicrobial resistance.

Hospitals in Kenya continue to use the Febrile Antigen Brucella Agglutination Test (FBAT) to diagnose brucellosis, despite reports showing its inadequacy. This study generated hospital-based evidence on the performance and cost-effectiveness of the FBAT, compared to the Rose Bengal Test (RBT).Twelve hospitals in western Kenya stored patient serum samples that were tested for brucellosis using the FBAT, and these were later re-tested using the RBT. Data on the running time and cost of the FBAT, and the treatment prescribed for brucellosis, were collected. The cost-effectiveness of the two tests, defined as the cost in US Dollars ($) per Disability Adjusted Life Year (DALY) averted, was determined, and a basic sensitivity analysis was run to identify the most influential parameters. Over a 6-month period, 180 patient serum samples that were tested with FBAT at the hospitals were later re-tested with RBT at the field laboratory. Of these 24 (13.3%) and 3 (1.7%) tested positive with FBAT and RBT, respectively. The agreement between the FBAT and RBT was slight (Kappa = 0.12). Treatment prescribed following FBAT positivity varied between hospitals, and only one hospital prescribed a standardized therapy regimen. The mean $/DALY averted when using the FBAT and RBT were $2,065 (95% CI $481-$6,736) and $304 (95% CI $126-$604), respectively. Brucellosis prevalence was the most influential parameter in the cost-effectiveness of both tests. Extrapolation to the national level suggested that an estimated $338,891 (95% CI $47,000-$1,149,000) per year is currently spent unnecessarily treating those falsely testing positive by FBAT. These findings highlight the potential for misdiagnosis using the FBAT. Furthermore, the RBT is cost-effective, and could be considered as the mainstay screening test for human brucellosis in this setting. Lastly, the treatment regimens must be harmonized to ensure the appropriate use of antibiotics for treatment.

Urbanization is predicted to be a key driver of disease emergence through human exposure to novel, animal-borne pathogens. However, while we suspect that urban landscapes are primed to expose people to novel animal-borne diseases, evidence for the mechanisms by which this occurs is lacking. To address this, we studied how bacterial genes are shared between wild animals, livestock, and humans (n = 1,428) across Nairobi, Kenya—one of the world’s most rapidly developing cities. Applying a multilayer network framework, we show that low biodiversity (of both natural habitat and vertebrate wildlife communities), coupled with livestock management practices and more densely populated urban environments, promotes sharing of Escherichia coli –borne bacterial mobile genetic elements between animals and humans. These results provide empirical support for hypotheses linking resource provision, the biological simplification of urban landscapes, and human and livestock demography to urban dynamics of cross-species pathogen transmission at a landscape scale. Urban areas where high densities of people and livestock live in close association with synanthropes (species such as rodents that are more competent reservoirs for zoonotic pathogens) should be prioritized for disease surveillance and control.

The keeping of livestock has been posited as a risk factor for the emergence of zoonoses and the spread of antimicrobial resistance. However, quantitative evidence regarding the major sources of pathogenic and drug-resistant bacteria and transmission routes between hosts remains lacking. In the largest epidemiological study of this nature to date, we sampled Escherichia coli from humans, livestock, food, wildlife and the environment of 99 households across Nairobi, Kenya to gain a deeper understanding of sharing of bacteria among hosts and potential reservoirs. By analysing whole genome sequencing data from 1,338 E. coli isolates, we reconstruct sharing patterns for the sampled E. coli and its antimicrobial resistance determinants. We find that the diversity and sharing patterns of E. coli is heavily structured by household, which is the primary epidemiological interface for bacterial strain sharing. Strain sharing within households was strongly shaped by host type. We also find evidence for inter-household and inter-host sharing, and importantly, between humans and animals, although this occurs much less frequently. We find similar strain sharing patterns for the E. coli accessory genome, suggesting that it is shaped by recent evolutionary history and is strongly associated with the core genome. Resistome similarity, however, were quite differently distributed across host and household, consistent with their being driven by shared exposure to antimicrobials. Our results indicate that there is potential for the exchange of bacteria between humans, livestock and wildlife in the same household in a tropical urban setting, with wider mixing occurring over a period of months or years, but this does not drive the distribution of antimicrobial resistance.