A comparison of risk factor investigation and experts’ opinion elicitation analysis for identifying foot-and-mouth disease (FMD) high-risk areas within the FMD protection zone of South Africa (2007–2016)

Abstract

Foot-and-mouth disease is a controlled disease in accordance with the South African Animal Diseases Act (Act 35 of 1984). The country was classified by the World Organisation for Animal Health (WOAH) as having a FMD free zone without vaccination in 1996. However, this status was suspended in 2019 due to a FMD outbreak outside the controlled zones. FMD control in South Africa includes animal movement restrictions placed on cloven-hoofed species and products, prophylactic vaccination of cattle, clinical surveillance of susceptible species, and disease control fencing to separate livestock from wildlife reservoirs. The objectives of this study were to evaluate differences in identifying high-risk areas for FMD using risk factor and expert opinion elicitation analysis. Differences in risk between FMD introduction and FMD spread within the FMD protection zone with vaccination (PZV) of South Africa (2007–2016) were also investigated.

The study was conducted in the communal farming area of the FMD PZV, which is adjacent to wildlife reserves and characterised by individual faming units. Eleven risk factors that were considered important for FMD occurrence and spread were used to build a weighted linear combination (WLC) score based on risk factor data and expert opinion elicitation. A multivariable conditional logistic regression model was also used to calculate predicted probabilities of a FMD outbreak for all dip-tanks within the study area. Smoothed Bayesian kriged maps were generated for 11 individual risk factors, overall WLC scores for FMD occurrence and spread and for predicted probabilities of a FMD outbreak based on the conditional logistic regression model. Descriptively, vaccine matching was believed to have a great influence on both FMD occurrence and spread. Expert opinion suggested that FMD occurrence was influenced predominantly by proximity to game reserves and cattle density. Cattle populations and vaccination practices were considered most important for FMD spread. Highly effective cattle inspections were observed within areas that previously reported FMD outbreaks, indicating the importance of cattle inspection (surveillance) as a necessary element of FMD outbreak detection.

The multivariable conditional logistic regression analysis, which was consistent with expert opinion elicitation; identified three factors including cattle population density (OR 3.87, 95% CI 1.47–10.21) and proximities to game reserve fences (OR 0.82, 95% CI 0.73–0.92) and rivers (OR 1.04, 95% CI 1.01–1.07) as significant factors for reported FMD outbreaks. Regaining and maintaining an FMD-free status without vaccination requires frequent monitoring of high-risk areas and designing targeted surveillance.