The data come from the epidemiological and epizootic surveillance systems established by local stakeholders. In the field of human health, active surveillance was implemented in 2008 following the identification of a risk of introduction of the RVF virus into Mayotte, which was circulating in coastal countries and islands of East Africa. To enable early detection of human cases, any patient presenting with a syndrome suggestive of RVF (dengue-like syndrome) is now systematically tested for the virus using PCR. This virological surveillance, established in collaboration with the Mayotte Hospital Center, proved effective in 2018 with the immediate detection of the first patient. A protocol for surveillance and investigation of cases was then initiated by Santé publique France in collaboration with the ARS, the CH, and the DAAF to assess the situation. Various types of data were collected: sociodemographic (e.g., activities related or unrelated to animal husbandry or milking), clinical and biological, housing, and exposures to the risk of RVF infection (slaughter outside of a slaughterhouse, consumption of raw milk). Similarly, the DAAF monitors RVF in livestock through reports from farmers of spontaneous abortions in females or unusual mortality, leading to biological analyses. The seroprevalence of RVF in livestock has also been continuously monitored since 2008. The population of Mayotte is exposed to numerous infectious risks, and for RVF, all the conditions are present for the emergence of epidemic outbreaks: the presence of small herds often in direct contact with human dwellings, complex veterinary monitoring, the importation of cattle, sheep, and goats (often from countries where RVF is endemic), the presence of disease-carrying mosquitoes (Aedes, Culex), and a lack of public awareness of the disease.

The implementation of these two surveillance systems, along with communication between human and animal health stakeholders, enabled the simultaneous collection of human and animal data during this most recent RVF epidemic. This marked a major first for this disease, and this dataset was analyzed using a mathematical model. The primary purpose of this model was to represent and study viral transmission among animals and from animals to humans, and then to assess the potential impact of livestock vaccination on reducing the number of human cases.

It is important to note that these surveillance systems are driven by a large number of field actors, including veterinary health officials, CoopADEM (the Mahoran Livestock Farmers’ Agricultural Cooperative), CIRAD, and the DAAF, who have been working on this surveillance for over 10 years—without whom this work would not have been possible. This multidisciplinary research is the result of frequent exchanges and relationships of trust that have been established between field workers, health professionals, and researchers over the past few years.

Generally speaking, RVF epidemics and epizootics occur in areas that lack comprehensive surveillance systems. It is rare for an outbreak to be accompanied by in-depth investigations of individual cases that would help elucidate the dynamics of the outbreak and the modes of transmission from animals to humans. Yet this information is essential for health authorities to develop an effective response. In Mayotte, investigations have yielded data pointing to the potential mode of infection for cases (direct contact with animals and/or the presence of mosquito populations in the environment). The model developed demonstrates that vector-borne transmission of the virus to humans may be as significant as transmission via direct contact. This distribution of human cases according to the two modes of transmission presented here is specific to Mayotte during this epidemic and also depends on the assumptions we made for the model. It may vary from one ecosystem to another because it depends on several factors specific to each study area, notably the proportion of the human population regularly exposed to animals, as well as the mosquito species present. Indeed, the RVF virus can be transmitted by a wide variety of vectors, which does not simplify its study. In any case, we present here a framework for estimating these figures and also encourage the collection of detailed entomological data to refine our estimates.

Previous research on RVF in Mayotte, an island territory, had suggested that following the virus’s emergence in 2008, the probability of the disease re-emerging in the subsequent years was low as long as the ecosystem was not exposed to two main factors: the reintroduction of the virus, and a decline in livestock immunity against RVF. Indeed, the further we move away from an epizootic outbreak in time, the more livestock populations are renewed and become susceptible to the virus. The risk of re-emergence then increases, especially if neighboring countries are affected by massive epizootics and animals are imported from those territories. This is what was observed during the 2018 epizootic.

Comprehensive public health, animal health, and environmental surveillance remains the best tool for anticipating and predicting the occurrence of new RVF outbreaks in Mayotte. Human and material resources must be developed to facilitate collaboration among teams, particularly through the development of shared tools for data collection and processing.

Thanks to this model, we are now able to adapt our measures for preventing and controlling the epidemic by proposing strategies tailored to the modes of transmission: vector control around human cases and early vaccination of animals from the onset of the epizootic.

The One Health approach enables a comprehensive examination of diseases, which are addressed from multiple perspectives: public health, animal health, and environmental health. Collaboration among stakeholders in these fields helps to understand and anticipate epidemic trends, thereby enabling a responsive and efficient management approach. For the FVR, animal outbreaks precede human epidemics, so it is essential to develop active and passive surveillance systems in animal and environmental health. In fact, it was thanks to an effective network of partners that the health alert was quickly issued in 2018: a single phone call between Santé publique France and the CoopADEM veterinarian was enough to link the occurrence of the first human case to the increase in seroprevalence among cattle. The alert was thus issued to health authorities to implement a strategy for preventing and controlling the epidemic. It has long been clear that epidemiologists cannot work in isolation to monitor the health of populations. The current challenge is to make this One Health concept operational by providing the tools needed to connect health actors at the local, regional, and international levels.