Antimicrobial resistance (AMR) is a serious threat to global public health, but obtaining representative data on AMR for healthy human populations is difficult. Here, we use metagenomic analysis of untreated sewage to characterize the bacterial resistome from 79 sites in 60 countries. We find systematic differences in abundance and diversity of AMR genes between Europe/North-America/Oceania and Africa/Asia/South-America. Antimicrobial use data and bacterial taxonomy only explains a minor part of the AMR variation that we observe. We find no evidence for cross-selection between antimicrobial classes, or for effect of air travel between sites. However, AMR gene abundance strongly correlates with socio-economic, health and environmental factors, which we use to predict AMR gene abundances in all countries in the world. Our findings suggest that global AMR gene diversity and abundance vary by region, and that improving sanitation and health could potentially limit the global burden of AMR. We propose metagenomic analysis of sewage as an ethically acceptable and economically feasible approach for continuous global surveillance and prediction of AMR.

Summary: There are few studies on the presence of extended‐spectrum β‐lactamases and AmpC β‐lactamases ( ESBL / AmpC ) in bacteria that contaminate vegetables. The role of the production environment in ESBL / AmpC gene transmission is poorly understood. The occurrence of ESBL / AmpC in E scherichia coli ( n = 46) from lettuce and irrigation water and the role of irrigation water in the transmission of resistant E . coli were studied. The presence of ESBL / AmpC , genetic similarity and phylogeny were typed using genotypic and phenotypic techniques. The frequency of β‐lactamase gene transfer was studied in vitro. ESBLs / AmpC were detected in 35 isolates (76%). Fourteen isolates (30%) produced both ESBLs / AmpC . Prevalence was highest in E . coli from lettuce (90%). Twenty‐two isolates (48%) were multi‐resistant with between two and five ESBL / AmpC genes. The major ESBL determinant was the CTX ‐ M type (34 isolates). DHA (33% of isolates) were the dominant AmpC β lactamases. There was a high conjugation efficiency among the isolates, ranging from 3.5 × 10 ⁻² to 1 × 10 ⁻² ± 1.4 × 10 ⁻¹ transconjugants per recipient. Water isolates showed a significantly higher conjugation frequency than those from lettuce. A high degree of genetic relatedness between E . coli from irrigation water and lettuce indicated possible common ancestry and pathway of transmission.

Prevention of the emergence and spread of foodborne diseases is an important prerequisite for the improvement of public health. Source attribution models link sporadic human cases of a specific illness to food sources and animal reservoirs. With the next generation sequencing technology, it is possible to develop novel source attribution models. We investigated the potential of machine learning to predict the animal reservoir from which a bacterial strain isolated from a human salmonellosis case originated based on whole‐genome sequencing. Machine learning methods recognize patterns in large and complex data sets and use this knowledge to build models. The model learns patterns associated with genetic variations in bacteria isolated from the different animal reservoirs. We selected different machine learning algorithms to predict sources of human salmonellosis cases and trained the model with Danish Salmonella Typhimurium isolates sampled from broilers (n = 34), cattle (n = 2), ducks (n = 11), layers (n = 4), and pigs (n = 159). Using cgMLST as input features, the model yielded an average accuracy of 0.783 (95% CI: 0.77–0.80) in the source prediction for the random forest and 0.933 (95% CI: 0.92–0.94) for the logit boost algorithm. Logit boost algorithm was most accurate (valid accuracy: 92%, CI: 0.8706–0.9579) and predicted the origin of 81% of the domestic sporadic human salmonellosis cases. The most important source was Danish produced pigs (53%) followed by imported pigs (16%), imported broilers (6%), imported ducks (2%), Danish produced layers (2%), Danish produced cattle and imported cattle (<1%) while 18% was not predicted. Machine learning has potential for improving source attribution modeling based on sequence data. Results of such models can inform risk managers to identify and prioritize food safety interventions.

Bacterial Foodborne Pathogens (FBP) are the commonest cause of foodborne illness or foodborne diseases (FBD) worldwide. They contaminate food at any stages in the entire food chain, from farm to dining-table. Among these, the Diarrheagenic Escherichia coli (DEC), Non typhoidal Salmonella (NTS), Shigella spp. and Campylobacter spp. are responsible for a large proportion of illnesses, deaths; and, particularly, as causes of acute diarrheal diseases. Though existing studies indicate the problem may be severe in developing countries like Ethiopia, the evidence is commonly based on fragmented data from individual studies. A review of published and unpublished manuscripts was conducted to obtain information on major FBP and identify the gaps in tracking their source attributions at the human, animal and environmental interface. A total of 1753 articles were initially retrieved after restricting the study period to between January 2000 and July 2020. After the second screening, only 51 articles on the humans and 43 on the environmental sample based studies were included in this review. In the absence of subgroups, overall as well as human stool and environmental sample based pooled prevalence estimate of FBP were analyzed. Since, substantial heterogeneity is expected, we also performed a subgroup analyses for principal study variables to estimate pooled prevalence of FBP at different epidemiological settings in both sample sources. The overall random pooled prevalence estimate of FBP (Salmonella, pathogenic Escherichia coli (E. coli), Shigella and Campylobacter spp.) was 8%; 95% CI: 6.5–8.7, with statistically higher (P < 0.01) estimates in environmental samples (11%) than in human stool (6%). The subgroup analysis depicted that Salmonella and pathogenic E. coli contributed to 5.7% (95% CI: 4.7–6.8) and 11.6% (95% CI: 8.8–15.1) respectively, of the overall pooled prevalence estimates of FBD in Ethiopia. The result of meta-regression showed, administrative regional state, geographic area of the study, source of sample and categorized sample size all significantly contributed to the heterogeneity of Salmonella and pathogenic E. coli estimates. Besides, the multivariate meta- regression indicated the actual study year between 2011 and 2015 was significantly associated with the environmental sample-based prevalence estimates of these FBP. This systematic review and meta-analysis depicted FBP are important in Ethiopia though majority of the studies were conducted separately either in human, animal or environmental samples employing routine culture based diagnostic method. Thus, further FBD study at the human, animal and environmental interface employing advanced diagnostic methods is needed to investigate source attributions of FBD in one health approach.