How to apply appropriate surveillance of apparently healthy biting dogs for human rabies prevention in dog rabies-free countries? A benefit-risk modelling study
Résumé
Surveillance of apparently healthy dogs that bite humans can be recommended to ascertain that they remain healthy indicating that no rabies virus (RABV) transmission occurred through salivary pre-symptomatic excretion. Such surveillance protocol of biting dogs saves many human lives in dog rabies enzootic areas since it allows prompt post-exposure prophylaxis (PEP) administration. Its justification in dog rabies-free areas remains not clearly defined due to its high burden, especially for pet owners and veterinarians, in a context of low rabies risk.
We conducted a benefit-risk assessment of several surveillance protocols of apparently healthy biting dogs, using the French situation as first example. A stochastic scenario tree modeling approach was developed to consider the possible successions of events between a dog bite and a human death attributed either to rabies or to lethal harm associated with the surveillance (lethal traffic accidents when traveling to veterinary clinics or anti-rabies centers). We showed that the current French surveillance protocol, relying on three veterinary visits during a 15-day period post-bite was not beneficial as more deaths were generated (by traffic accidents) than avoided (by prompt PEP administration). Similar results are expected for countries without established rabies reservoir in non-flying mammals (e.g. in Western Europe). In these areas, risk-based surveillance (concerning only dogs at-risk for rabies) could prove more appropriate. In countries with non-flying wildlife rabies reservoir (e.g. in Northern America), non-risk-based surveillance with a reduced number of veterinary visits could also prove beneficial to some extent. These results could thus directly inform policymakers.