Combining seroprevalence and capture-mark-recapture data to estimate the force of infection of brucellosis in a managed population of Alpine ibex

Abstract

In wildlife, epidemiological data are often collected using cross-sectional surveys and antibody tests, and seroprevalence is the most common measure used to monitor the transmission dynamics of infectious diseases. On the contrary, the force of infection, a measure of transmission intensity that can help understand epidemiological dynamics and monitor management interventions, remains rarely used. The force of infection can be derived from age-stratified cross-sectional serological data, or from longitudinal data (although less frequently available in wildlife populations). Here, we combined seroprevalence and capture-mark-recapture data to estimate the force of infection of brucellosis in an Alpine ibex (Capra ibex) population monitored from 2012 to 2018. Because the seroprevalence of brucellosis was 38% in this population in 2012, managers conducted two culling operations in 2013 and 2015, as well as captures every year since 2012, where seronegative individuals were marked and released, and seropositive individuals were removed. We obtained two estimates of the force of infection and its changes across time, by fitting (i) a catalytic model to age-seroprevalence data obtained from unmarked animals (cross-sectional), and (ii) a survival model to event time data obtained from recaptures of marked animals (longitudinal). Using both types of data allowed us to make robust inference about the temporal dynamics of the force of infection: indeed, there was evidence for a decrease in the force of infection between mid-2014 and late 2015 in both datasets. The force of infection was estimated to be reduced from 0.115 year-1 [0.074–0.160] to 0.016 year-1 [0.001–0.057]. These results confirm that transmission intensity decreased during the study period, probably due to management interventions and natural changes in infection dynamics. Estimating the force of infection could therefore be a valuable complement to classical seroprevalence analyses to monitor the dynamics of wildlife diseases, especially in the context of ongoing disease management interventions.

Keywords

• Wildlife disease management
• Force of infection
• Seroprevalence
• Catalytic model
• Survival analysis