Antibiotic use attributable to specific aetiologies of diarrhoea in children under 2 years of age in low-resource settings: a secondary analysis of the MAL-ED birth cohort

OBJECTIVE: To quantify the frequency of antibiotic treatments attributable to specific enteric pathogens due to the treatment of diarrhoea among children in the first 2 years of life in low-resource settings. DESIGN: Secondary analysis of a longitudinal birth cohort study, Etiology, Risk Factors, and Interactions of Enteric Infections and Malnutrition and the Consequences for Child Health and Development (MAL-ED). SETTING: This study was conducted at eight sites in Bangladesh, Brazil, India, Nepal, Peru, Pakistan, South Africa and Tanzania. PARTICIPANTS: We analysed 9392 reported diarrhoea episodes, including 6677 with molecular diagnostic test results, as well as 31 408 non-diarrhoeal stools from 1715 children aged 0–2 years with 2 years of complete follow-up data. PRIMARY AND SECONDARY OUTCOME MEASURES: We estimated incidence rates and the proportions of antibiotic use for diarrhoea and for all indications attributable to the top 10 aetiologies of diarrhoea. We estimated associations between specific aetiologies and antibiotic treatment, and assessed whether clinical characteristics of the diarrhoea episodes mediated these relationships. RESULTS: Shigella and rotavirus were the leading causes of antibiotic treatment, responsible for 11.7% and 8.6% of diarrhoea treatments and 14.8 and 10.9 courses per 100 child-years, respectively. Shigella and rotavirus-attributable diarrhoea episodes were 46% (RR: 1.46; 95% CI: 1.33 to 1.60), and 19% (RR: 1.19; 95% CI: 1.09 to 1.31) more likely to be treated with antibiotics, respectively, compared with other aetiologies. Considering antibiotic uses for all indications, these two pathogens accounted for 5.6% of all antibiotic courses, 19.3% of all fluoroquinolone courses and 9.5% of all macrolide courses. Among indicated treatments for dysentery, Shigella and Campylobacter jenjui/Campylobacter coli were responsible for 27.5% and 8.5% of treated episodes, respectively. CONCLUSIONS: The evidence that Shigella and rotavirus were disproportionately responsible for antibiotic use due to their high burden and severity further strengthens the value of interventions targeted to these pathogens. Interventions against Campylobacter could further prevent a large burden of indicated antibiotic treatment for dysentery, which could not be averted by antibiotic stewardship interventions.