In an effort to strengthen global capacity to prevent, detect, and control infectious diseases in animals and people, the United States Agency for International Development’s (USAID) Emerging Pandemic Threats (EPT) PREDICT project funded development of regional, national, and local One Health capacities for early disease detection, rapid response, disease control, and risk reduction. From the outset, the EPT approach was inclusive of social science research methods designed to understand the contexts and behaviors of communities living and working at human-animal-environment interfaces considered high-risk for virus emergence. Using qualitative and quantitative approaches, PREDICT behavioral research aimed to identify and assess a range of socio-cultural behaviors that could be influential in zoonotic disease emergence, amplification, and transmission. This broad approach to behavioral risk characterization enabled us to identify and characterize human activities that could be linked to the transmission dynamics of new and emerging viruses. This paper provides a discussion of implementation of a social science approach within a zoonotic surveillance framework. We conducted in-depth ethnographic interviews and focus groups to better understand the individual- and community-level knowledge, attitudes, and practices that potentially put participants at risk for zoonotic disease transmission from the animals they live and work with, across 6 interface domains. When we asked highly-exposed individuals (ie. bushmeat hunters, wildlife or guano farmers) about the risk they perceived in their occupational activities, most did not perceive it to be risky, whether because it was normalized by years (or generations) of doing such an activity, or due to lack of information about potential risks. Integrating the social sciences allows investigations of the specific human activities that are hypothesized to drive disease emergence, amplification, and transmission, in order to better substantiate behavioral disease drivers, along with the social dimensions of infection and transmission dynamics. Understanding these dynamics is critical to achieving health security--the protection from threats to health-- which requires investments in both collective and individual health security. Involving behavioral sciences into zoonotic disease surveillance allowed us to push toward fuller community integration and engagement and toward dialogue and implementation of recommendations for disease prevention and improved health security.

Background: Many ecologically important plants are pollinated or have their seeds dispersed by fruit bats, including the widely distributed African straw-colored fruit bats (Eidolon helvum). Their ability to fly long distances makes them essential for connecting plant populations across fragmented landscapes. While bats have been implicated as a reservoir of infectious diseases, their role in disease transmission to humans is not well understood. In this pilot study, we tracked E. helvum to shed light on their movement patterns in Tanzania and possible contact with other species. Methods: Tracking devices were deployed on 25 bats captured in the Morogoro Municipal and Kilombero District area near the Udzungwa Mountains of Tanzania. Nightly flight patterns, areas corresponding to foraging bouts and feeding roosts, and new day roosts were determined from bat movement data and characterized according to their proximity to urban built-up and protected areas. Sites for additional environmental surveillance using camera traps were identified via tracking data to determine species coming in contact with fruits discarded by bats. Results: Tracking data revealed variability between individual bat movements and a fidelity to foraging areas. Bats were tracked from one to six nights, with a mean cumulative nightly flight distance of 26.14 km (min: 0.33, max: 97.57) based on data from high-resolution GPS tags. While the majority of their foraging locations were in or near urban areas, bats also foraged in protected areas, of which the Udzungwa Mountains National Park was the most frequented. Camera traps in fruit orchards frequented by tracked bats showed the presence of multiple species of wildlife, with vervet monkeys (Chlorocebus pygerythrus) observed as directly handling and eating fruit discarded by bats. Conclusions: Because we observed multiple interactions of animals with fruits discarded by bats, specifically with vervet monkeys, the possibility of disease spillover risk exists via this indirect pathway. With flight distances of up to 97 km, however, the role of E. helvum in the seed dispersal of plants across both protected and urban built-up areas in Tanzania may be even more important, especially by helping connect increasingly fragmented landscapes during this Anthropocene epoch.

Diarrhoeal diseases can be debilitating, especially for children and young animals. In many rural areas, particularly pastoral communities, livelihoods are characterized by close interaction between household members and their livestock herds, and children often care for young animals, creating opportunities for the transmission of multiple zoonotic pathogens. Using a One Health approach, we first evaluated whether diarrhoeal diseases were a problem for pastoral households in Tanzania and then investigated their calf herds to identify the prevalence and risk factors for diarrhoeal disease and the shedding of the zoonotic pathogens Cryptosporidium and Giardia. Sixty percent of households reporting cases of human diarrhoea also had diarrhoea detected later in their calf herds, and calf herds shedding Cryptosporidium oocysts were six times more likely to be diarrhoeic. Because Cryptosporidium shares a similar transmission mode with a wide range of diarrhoeagenic organisms and calf diarrhoea outbreaks can involve multiple pathogens with mixed infections, it is possible that calf diarrhoea may be indicative of shared risk of zoonotic pathogens from environmental contamination. To mitigate the risk of transmission of faecal-borne zoonotic pathogens from herds to households (boma-livestock pens to banda –household building), we describe a conceptual disease early-warning method proposing diarrhoeic calves as animal sentinels. Such a calf warning system, combined with appropriate interventions designed to minimize exposure, could serve as a practical solution for reducing risks of diarrhoeal diseases among animals and people.

Traditional media and the internet are crucial sources of health information. Media can significantly shape public opinion, knowledge and understanding of emerging and endemic health threats. As digital communication rapidly progresses, local access and dissemination of health information contribute significantly to global disease detection and reporting. Health event reports in Nepal (October 2013–December 2014) were used to characterize Nepal’s media environment from a One Health perspective using HealthMap - a global online disease surveillance and mapping tool. Event variables (location, media source type, disease or risk factor of interest, and affected species) were extracted from HealthMap. A total of 179 health reports were captured from various sources including newspapers, inter-government agency bulletins, individual reports, and trade websites, yielding 108 (60%) unique articles. Human health events were reported most often (n = 85; 79%), followed by animal health events (n = 23; 21%), with no reports focused solely on environmental health. By expanding event coverage across all of the health sectors, media in developing countries could play a crucial role in national risk communication efforts and could enhance early warning systems for disasters and disease outbreaks.

Rift Valley fever (RVF) is a re-emerging vector-borne zoonotic disease that causes outbreaks in humans and animals across Africa. To better understand RVF at human–animal interfaces, a prospective longitudinal survey of people, livestock, and mosquitoes was conducted from 2016 to 2018, in two regions of Tanzania, with distinct climatic zones (Iringa and Morogoro). Molecular and serological tools for testing (RT-qPCR and IgM/IgG ELISA) for RVF virus (RVFV) were used to assess infection and exposure in people and animals. Mosquitoes were collected quarterly from 10 sentinel locations. In total, 1385 acutely febrile humans, 4449 livestock, and 3463 mosquito pools were tested. In humans, IgM seroprevalence was 3.75% (n = 52/1385), and overall seroprevalence (IgM and/or IgG positive) was 8.30% (n = 115/1385). People from Iringa had a higher exposure risk than those from Morogoro (aOR 2.63), and livestock owners had an increased risk compared to non-owners (aOR 2.51). In livestock, IgM seroprevalence was 1.09%, while overall seroprevalence was 10.11%. A total of 68.4% of herds had at least one seropositive animal. Sentinel animal follow-up revealed that the probability of seroconversion was significantly higher in Morogoro. Low-level RVFV RNA was detected in 8 human and 22 mosquito pools. These findings indicate active transmission among vectors, livestock, and people during the study period, highlighting the need for One Health surveillance approaches for RVFV and other arboviruses.

Forecasting the risk of pathogen spillover from reservoir populations of wild or domestic animals is essential for the effective deployment of interventions such as wildlife vaccination or culling. Due to the sporadic nature of spillover events and limited availability of data, developing and validating robust, spatially explicit, predictions is challenging. Recent efforts have begun to make progress in this direction by capitalizing on machine learning methodologies. An important weakness of existing approaches, however, is that they generally rely on combining human and reservoir infection data during the training process and thus conflate risk attributable to the prevalence of the pathogen in the reservoir population with the risk attributed to the realized rate of spillover into the human population. Because effective planning of interventions requires that these components of risk be disentangled, we developed a multi-layer machine learning framework that separates these processes. Our approach begins by training models to predict the geographic range of the primary reservoir and the subset of this range in which the pathogen occurs. The spillover risk predicted by the product of these reservoir specific models is then fit to data on realized patterns of historical spillover into the human population. The result is a geographically specific spillover risk forecast that can be easily decomposed and used to guide effective intervention. Applying our method to Lassa virus, a zoonotic pathogen that regularly spills over into the human population across West Africa, results in a model that explains a modest but statistically significant portion of geographic variation in historical patterns of spillover. When combined with a mechanistic mathematical model of infection dynamics, our spillover risk model predicts that 897,700 humans are infected by Lassa virus each year across West Africa, with Nigeria accounting for more than half of these human infections. Author Summary: The 2019 emergence of SARS-CoV-2 is a grim reminder of the threat animal-borne pathogens pose to human health. Even prior to SARS-CoV-2, the spillover of pathogens from animal reservoirs was a persistent problem, with pathogens such as Ebola, Nipah, and Lassa regularly but unpredictably causing outbreaks. Machine-learning models that anticipate when and where pathogen transmission from animals to humans is likely to occur would help guide surveillance efforts and preemptive countermeasures like information campaigns or vaccination programs. We develop a novel machine learning framework that uses datasets describing the distribution of a virus within its host and the range of its animal host, along with data on spatial patterns of human immunity, to infer rates of animal-to-human transmission across a region. By training the model on data from the animal host alone, our framework allows rigorous validation of spillover predictions using human data. We apply our framework to Lassa fever, a viral disease of West Africa that is spread to humans by rodents, and use the predictions to update estimates of Lassa virus infections in humans. Our results suggest that Nigeria is most at risk for the emergence of Lassa virus, and should be prioritized for outbreak-surveillance.