How farm practices and antibiotic use drive disease incidence in smallholder livestock farms: Evidence from a survey in Uganda.

Background: Antimicrobial resistance (AMR) is a growing threat to human and animal health, and the growth in AMR prevalence globally is thought to be partially driven by non-therapeutic antibiotic use in livestock production. However, livestock farms may depend on antibiotics as a prophylactic disease management tool, and reducing antibiotic use in isolation may harm farmers' economic security. In order to help farmers safely reduce their antibiotic use, we must first determine how necessary non-therapeutic antibiotic use is for disease management, and how other farm practices can guard against disease and make antibiotic use reduction safe and feasible.

Methods: Using the (AMUSE) tool, a standardised survey tool for investigating attitudes and practices relating to antibiotic use on farms, we investigated the farming practices and animal disease outcomes of smallholder livestock farms in Uganda. We used logistic regression to investigate the effect of prophylactic antibiotic use; as well as of prophylactic vaccination, non-antimicrobial medicines, and on-farm biosecurity measures; on the likelihood of disease outbreaks.

Findings: We found that prophylactic antibiotic use did indeed seem to guard against disease outbreaks, underlining the rationality of non-therapeutic antibiotic use in smallholder livestock farms and the need to pair antibiotic use reduction with other interventions in order to mitigate risk. The most effective intervention pairing varied by species, with expanded access to animal health services and the use of prophylactic vaccination demonstrating the greatest potential overall.

Implications: These findings echo earlier results generated using the AMUSE survey tool. They should be followed by participatory research in which farmers are consulted to explore intervention options, and subsequently by farm-level intervention trials of combined antimicrobial stewardship interventions to verify their effectiveness.