**Background:** Antimicrobial resistance (AMR) is a major global health threat, with E. coli serving as a key sentinel organism due to its widespread occurrence and capacity to acquire and transfer resistance genes. In developing countries like Ethiopia, data on AMR in livestock and environmental samples are sparse, particularly in extensive smallholder production systems. This study aimed to characterize the distribution of AMR in E. coli isolated from livestock feces and soil in low-resource, extensive smallholder livestock production systems in Ethiopia.
**Methods:** A cross-sectional study was conducted across two agroecological zones in Ethiopia: a highland mixed crop–livestock production system (Menz Mama and Menz Gera districts) and a pastoral system (Yabello and Eleweya districts). A total of 539 samples were collected from 77 households, comprising 462 livestock fecal samples (cattle=152, sheep=180, goats=130) and 77 soil samples. E. coli was isolated using standard microbiological methods including MacConkey agar, eosin-methylene-blue agar, and IMViC biochemical tests, with further identification using the Biolog OmniLog ID system, which also identified E. coli O157:H7 serotype. Antimicrobial susceptibility testing was performed using the Kirby-Bauer disk diffusion method on Mueller-Hinton agar against 15 antibiotics. Risk factor analysis was conducted using univariable and multivariable logistic regression, with separate models for livestock and soil samples.
**Key Results:** E. coli was isolated from 86.2% of cattle fecal samples, 77.8% of sheep samples, 68.5% of goat samples, and 39.2% of soil samples. Of 462 isolates, 42 (9.1%) were identified as E. coli O157:H7, with the highest occurrence in goats (20.2%), followed by cattle (11.4%), soil (6.8%), and sheep (1.4%). Resistance to at least one antimicrobial was found in 51.9% of cattle isolates, 33.6% of sheep isolates, 58.4% of goat isolates, and 52.9% of soil isolates, with an overall prevalence of 47.8% (95% CI, 43.3–52.4). The highest resistance was against streptomycin (33.2% overall; 37.3% cattle, 16.8% sheep, 49.4% goat, 36.4% soil), followed by amoxycillin/clavulanate (23.2% overall; 25.8% cattle, 11.2% sheep, 37.0% goat, 24.8% soil) and tetracycline (8% overall; 3.9% cattle, 5.8% sheep, 13.6% goat, 11.3% soil). Resistance to newer drug classes was low: ciprofloxacin (4.2%), cefotaxime (6.1%), and chloramphenicol (2.1%). Of 375 isolates tested against all 15 antibiotics, 170 (45.3%) were resistant to at least one antibiotic, and 100 (26.7%) showed multidrug resistance (resistant to ≥2 antibiotic classes), with one isolate resistant to 8 antibiotic classes. The most common co-resistant phenotype was amoxycillin/clavulanate and streptomycin (18.8%). Among E. coli O157:H7 isolates, 45.2% showed MDR. Multivariable logistic regression revealed that livestock from lowland pastoral production systems had nearly three times higher odds of resistance to ≥2 antibiotics compared to highland mixed crop–livestock systems (OR: 2.98; 95% CI, 1.72–5.17; p=0.000). For soil samples, households that used manure for fuel or sold it had 85% lower odds of having ≥2 resistance phenotypes compared to those leaving manure on the farm or discarding it into the environment (OR: 0.15; 95% CI, 0.03–0.75; p=0.03).
**Clinical Implications:** This study demonstrates that AMR in E. coli is prevalent in both livestock and environmental samples in Ethiopian smallholder systems, with resistance concentrated against older, commonly used antibiotics. The finding that soil contaminated with livestock feces can harbor drug-resistant E. coli, including O157:H7, underscores the environmental dimension of AMR transmission and the potential for human exposure through contaminated water, produce, and direct contact. The strong association between pastoral production systems and higher resistance levels, as well as the link between poor manure management and resistance in soil, identifies actionable targets for intervention. The authors recommend improved animal health and biosecurity practices—particularly manure management—and the establishment of an integrated AMR surveillance system to monitor resistance patterns and trends in these low-resource settings.