# Post-COVID-19 Rabies Surveillance and Risk Factors in Rural Eastern Cape, South Africa: A One Health Perspective

## TL;DR

This study explores rabies risk and surveillance in rural South Africa, emphasizing the role of education and community behavior in controlling the disease.

## Contribution

The study introduces a Surveillance Gap Index and uses machine learning to identify key rabies risk factors in post-pandemic rural communities.

## Key findings

- 51% of households reported dog-bite exposure, with males at significantly higher risk.
- Educational level was positively linked to pet vaccination rates.
- Machine learning models accurately predicted rabies risk factors with high AUC scores.

## Abstract

Background: Rabies remains a neglected zoonotic disease in South Africa, particularly in rural areas where surveillance weaknesses, behavioral gaps, and limited One Health coordination persist. Objectives: This study assessed rabies surveillance, behavioral risk factors, and system responsiveness in two rural Eastern Cape communities, with a focus on post-pandemic resilience within a One Health framework. Methods: A cross-sectional, community-based pilot study was conducted among 109 residents using structured questionnaires to collect data on demographics, rabies awareness, vaccination practices, and service disruptions. Descriptive, bivariate, and multivariate analyses identified predictors of dog-bite exposure and pet vaccination. Machine learning models (Decision Tree and Random Forest) were applied to explore risk hierarchies. A composite Surveillance Gap Index (SGI) was developed to integrate behavioral and systemic indicators. Results: While 88% of participants were aware of rabies, only 35% attended awareness campaigns. Dog-bite exposure affected 51% of households, with significantly higher risk among males (aOR = 4.33; p = 0.003). Education was positively associated with pet vaccination (aOR = 1.78). Despite 45% reporting COVID-19 disruptions, communities maintained high post-pandemic vaccination coverage (85.7%). Predictive models (AUC = 0.82–0.86) identified education, gender, awareness, and distance as key risk drivers. Conclusions: Integrating behavioral insights and predictive analytics into One Health strategies can strengthen rabies surveillance and support progress toward eliminating human rabies by 2030.