Determinants of antimicrobial resistance practices in Ethiopia: A One Health mixed-methods study

Abstract

Background: Antimicrobial resistance (AMR) is a growing One Health challenge in low- and middle-income countries, where inappropriate antimicrobial use across human and animal sectors remains common. In Ethiopia, national AMR strategies and One Health frameworks exist, but evidence remains limited on how knowledge, attitudes, and practices shape antimicrobial-use behaviors across community and animal health settings. Methods: We conducted a convergent parallel mixed-methods study from June to November 2024 in Addis Ababa, Oromia, and Afar, Ethiopia. Quantitative data were collected using structured knowledge, attitude, and practice surveys among household respondents (n = 750) and animal health professionals (n = 251). Knowledge was defined as factual understanding of antimicrobials and AMR; attitude as beliefs, perceived risk, and stewardship orientation; and practice as self-reported antimicrobial-use and stewardship-related behaviors. Composite KAP scores were standardized from 0 to 1, with higher scores indicating better outcomes. Fractional logit regression was used to assess factors associated with AMR-related practice scores. Qualitative data were collected through 24 focus group discussions, 58 key informant interviews, and 34 in-depth interviews, and analyzed thematically to contextualize quantitative findings. Results: Households had moderate knowledge (mean = 0.62) and positive attitudes (mean = 0.75) but suboptimal practices (mean = 0.58). Animal health professionals had higher knowledge (mean = 0.88) and attitudes (mean = 0.79), but lower practice scores (mean = 0.49). Attitude was the most consistent factor associated with better AMR-related practices among both households (aOR = 1.26; 95% CI: 1.17-1.36) and animal health professionals (aOR = 1.50; 95% CI: 1.28-1.75). Knowledge was significantly associated with practice only among animal health professionals. Larger household size was negatively associated with household practice, while professional affiliation was associated with animal health professional practice. Qualitative findings showed that financial constraints, non-prescription antimicrobial access, limited diagnostic capacity, weak enforcement, and client pressure constrained the translation of knowledge and positive attitudes into appropriate practice. Conclusions: AMR-related practices in Ethiopia are shaped by attitudes, professional and household contexts, and structural constraints, rather than knowledge alone. Interventions should move beyond information provision to address risk perception, social norms, affordability, diagnostic access, regulatory enforcement, and stewardship support. Strengthened One Health coordination across human and animal sectors is essential to translate Ethiopia's AMR policies into effective practice.