Last-Resort Antibiotic Resistance Surges Across East Africa in Humans and Environment

Carbapenems are supposed to be the final line of defense, the antibiotics clinicians reach for when nearly everything else has failed against a dangerous bacterial infection. A new systematic review and meta-analysis spanning a decade of research across East Africa now shows that Escherichia coli, one of the world’s most common and versatile bacterial pathogens, is increasingly shrugging off these last-resort drugs. Pooling data from twenty-nine studies conducted between September 2015 and September 2025, researchers calculated that 12.7 percent of E. coli isolates across the region are resistant to carbapenems, a figure that places East Africa among the areas of greatest concern in the global fight against antimicrobial resistance. With resistance genes detected in hospitals, sewage, food, livestock, and the wider environment, the study delivers a stark warning that the machinery of resistance is already embedded across the region’s human, animal, and ecological systems.

The research, led by Tizazu Zelelie, Biniam Mekonnen, and Demiss Nigussie and published in Public Health in Practice, was designed around the One Health framework, the internationally endorsed recognition that human, animal, and environmental health are inseparable. First formalized in 2015 by the tripartite coalition of the Food and Agriculture Organization, the World Health Organization, and the World Organization for Animal Health, and later expanded to include the United Nations Environment Programme, the approach acknowledges that resistant bacteria and their genes do not respect the boundaries between hospital wards, farms, and waterways. Resistant organisms shed into the environment can transfer their resistance genes horizontally to other bacteria infecting humans and animals, creating reservoirs of resistance that no single-sector intervention can drain. East Africa presents particularly fertile ground for this dynamic, combining agricultural and pastoral livelihoods with patterns of antimicrobial misuse, limited access to clean water, poor sanitation, and inadequate infection prevention infrastructure.

To capture the full scope of the problem, the authors searched PubMed, ScienceDirect, and Google Scholar for studies reporting carbapenem-resistant E. coli isolated from humans, animals, food, or the environment in thirteen East African countries. Following the PRISMA reporting guidelines and registered prospectively on PROSPERO, the review screened more than 4,300 records, ultimately retaining twenty-nine studies that met rigorous eligibility criteria. Each study was assessed for bias by two independent reviewers using the Newcastle Ottawa Scale, and all were judged to carry low risk of bias, scoring at or above the six-star threshold. The researchers then performed a random-effects meta-analysis using restricted maximum likelihood estimation in the R statistical environment, quantifying heterogeneity with the I-squared statistic and probing publication bias with funnel plots and Egger’s test.

The numbers underlying the headline finding are substantial. Across the included studies, the combined sample size reached 11,581 participants and specimens, from which 3,908 E. coli isolates were recovered. Of those, 442 strains proved resistant to carbapenem antibiotics. The studies were heavily weighted toward clinical settings, with seventy-six percent hospital-based and human samples accounting for nearly eighty percent of all sources, but the remainder drew from environmental reservoirs, food, and animals, giving the analysis its distinctive cross-sector character. Geographically, Ethiopia dominated the evidence base with fourteen studies, followed by Uganda with seven, Sudan with four, Tanzania with two, and single studies from Djibouti and Kenya. No eligible studies emerged from the other seven countries in the region, itself a telling indicator of surveillance gaps.

Beneath the pooled estimate of 12.7 percent lies a sobering spread. The lowest reported prevalence was 0.5 percent in an Ethiopian study, while the highest, drawn from sewage at a Ugandan referral hospital, reached a remarkable 50 percent. In subgroup analyses, Sudan recorded the highest country-level pooled prevalence at 24.7 percent, followed by Uganda at 17.4 percent and Ethiopia at 9.9 percent. More striking still was the finding that the environment emerged as the most contaminated interface of all, with a pooled prevalence of 24.2 percent among environmental isolates, compared with 11.9 percent in humans, 7.7 percent in food, and 4.2 percent in animals. Clinical samples of multiple types yielded a pooled prevalence of 15.8 percent. These figures suggest that hospitals, farms, and communities are continuously seeding the environment with resistant organisms, which in turn serve as a reservoir capable of reinfecting the human population.

Genetic analysis revealed the molecular arsenal behind this resistance. Half of the included studies, 51.7 percent, reported the detection of carbapenemase-encoding genes, including KPC and GES from the class A serine beta-lactamases, NDM-1 and NDM-5, VIM, IMP, and SPM among the metallo-beta-lactamases, and OXA-48 and OXA-181 oxacillinases. These genes encode enzymes that destroy carbapenems and are readily transferred between bacterial species on mobile genetic elements, meaning a resistance trait that emerges in one organism can rapidly colonize others. The World Health Organization ranks carbapenem-resistant pathogens first on its global priority list of antibiotic-resistant bacteria precisely because treatment options for them are so desperately limited: essentially reduced to polymyxins, fosfomycin, and tigecycline, older or more toxic drugs that represent imperfect substitutes for an entire antibiotic class.

The stakes of this trend are measured in lives. Global modeling published in the Lancet estimates that bacterial antimicrobial resistance was associated with nearly 5 million deaths in 2021, contributing to roughly one in eight deaths worldwide, and projects that as many as 8.22 million deaths annually could be attributable to AMR by 2050 if current trajectories continue. In the WHO African region alone, more than one million deaths were associated with resistant bacteria in 2019. Among Gram-negative bacteria, resistance to carbapenems has climbed faster than to any other antibiotic class, with associated deaths rising from 619,000 in 1990 to 1.03 million in 2021. Carbapenem-resistant Enterobacteriaceae infections are linked to longer hospital stays, higher healthcare costs, and elevated mortality, burdens that fall hardest on health systems with the least capacity to absorb them.

The review also captured a subtle temporal signal. Prevalence in studies published between 2021 and 2025 was marginally lower, at 12.6 percent, than in the 2015 to 2020 window, which averaged 13.5 percent, though the authors note this difference was not statistically significant and may simply reflect variation in study designs, settings, and target populations. Regional trend data from Ethiopia’s Amhara Public Health Institute, which showed resistance climbing to a peak in 2019 before declining, offers tentative grounds for the hope that sustained interventions can bend the curve. At the same time, the extraordinarily high statistical heterogeneity among the studies, with an I-squared value of 93.3 percent, underscores how differently the epidemic of resistance is unfolding across countries, populations, and sample types, and the authors caution that significant publication bias was detected and statistically adjusted in their estimates.

Perhaps the most consequential implication of the study is what it says about surveillance and regional coordination. The authors attribute the region’s elevated burden to weak laboratory capacity, poor health system governance, limited diagnostic infrastructure, and the destabilizing effects of conflict in countries such as Sudan and Ethiopia, all compounded by prolonged and often inappropriate antimicrobial use in both human and animal populations. Because routine testing for carbapenemase production is rarely performed in African clinical laboratories, the true prevalence may well exceed what published studies capture. The authors call for effective implementation of the One Health approach across sectors, strengthened routine diagnostic capacity, national and regional surveillance programs, rapid detection methods, and integration that extends beyond health policy alone. Without such coordinated action, the environmental reservoirs documented at prevalence rates approaching one in four isolates will continue to replenish the human burden, ensuring that one of medicine’s final antibiotics loses its power across one of the world’s most vulnerable regions.

Subject of Research: Prevalence and molecular epidemiology of carbapenem-resistant E. coli across human, animal, and environmental interfaces in East Africa

Article Title: Carbapenem-resistant E. coli with one-health approach in East Africa: Systematic review and meta-analysis

Article References: Zelelie, T., Mekonnen, B., & Nigussie, D. (2026). Carbapenem-resistant E. coli with one-health approach in East Africa: Systematic review and meta-analysis. Public Health in Practice, 12, Article 100843. https://doi.org/10.1016/j.puhip.2026.100843

Image Credits: AI Generated

DOI: 10.1016/j.puhip.2026.100843

Keywords: carbapenem-resistant E. coli, antimicrobial resistance, One Health, East Africa, systematic review, meta-analysis, carbapenemase genes, NDM, public health surveillance, Ethiopia, Uganda, Sudan