ObjectivesThe aim of this study is to evaluate existing evidence on the effectiveness of amoxicillin and amoxicillin-clavulanate for community-acquired pneumonia in children and adults.DesignSystematic review and meta-analysis.Data sourcesPubMed, Cochrane Library, Web of Science and Ovid-MEDLINER were searched with no language restrictions through 16 July 2024.Eligibility criteriaWe included studies comparing the effectiveness of amoxicillin or amoxicillin-clavulanate versus other antibiotics or placebo.Data extraction and synthesisOnly randomised controlled trials comparing amoxicillin or amoxicillin-clavulanate with another antibiotic or placebo with a primary outcome of clinical resolution or clinical failure were eligible for our review. We used random-effects and fixed-effects logistic regression models to estimate the pooled treatment effect size. Heterogeneity of the studies was evaluated using the τ statistic. We performed an unplanned frequentist random-effects network meta-analysis for the indirect comparison between amoxicillin and amoxicillin-clavulanate. The revised Cochrane risk of bias tool for randomised trials was used to assess and categorise studies into low risk of bias, some concerns or high risk of bias.ResultsWe extracted data from 44 studies including 45 400 patients. We found no evidence of a differential effect on clinical resolution when comparing amoxicillin with other antibiotics (n=15 trials; pooled OR 0.88; 95% CI 0.56 to 1.38, where >1 favours amoxicillin) or amoxicillin-clavulanate with other antibiotics (n=17; OR 0.89; 95% CI 0.76 to 1.04). Similarly, evidence of difference in clinical failure between amoxicillin and other antibiotics was unclear and unable to rule out clinically important benefits or harms (n=8; OR 0.76; 95% CI 0.55 to 1.06, where <1 favours amoxicillin). We found no evidence of a differential effect on clinical resolution between adults treated with amoxicillin and amoxicillin-clavulanate (n=28; OR 1.04; 95% CI 0.64 to 1.70, where >1 favours amoxicillin-clavulanate). Sixty-three per cent and 29% of amoxicillin and amoxicillin-clavulanate studies, respectively, had low risk of bias according to the Cochrane risk of bias tool for randomised trials.ConclusionsCurrent evidence is unclear as to whether amoxicillin or amoxicillin-clavulanate differs from other antibiotics, or from each other, in the treatment of community-acquired pneumonia, owing to the small number of trials and substantial heterogeneity in comparators used across study settings.Prospero registration numberCRD42024568554.