BACKGROUND: Antibiotic overprescribing is a major concern that contributes to the problem of antibiotic resistance. AIM: To assess the effect on antibiotic prescribing in primary care of telehealth (TH) consultations compared with face-to-face (F2F). DESIGN & SETTING: Systematic review and meta-analysis of adult or paediatric patients with a history of a community-acquired acute infection (respiratory, urinary, or skin and soft tissue). Studies were included that compared synchronous TH consultations (phone or video-based) to F2F consultations in primary care. METHOD: PubMed, Embase, Cochrane CENTRAL (inception-2021), clinical trial registries and citing-cited references of included studies were searched. Two review authors independently screened the studies and extracted the data. RESULTS: Thirteen studies were identified. The one small randomised controlled trial (RCT) found a non-significant 25% relative increase in antibiotic prescribing in the TH group. The remaining 10 were observational studies but did not control well for confounding and, therefore, were at high risk of bias. When pooled by specific infections, there was no consistent pattern. The six studies of sinusitis - including one before-after study - showed significantly less prescribing for acute rhinosinusitis in TH consultations, whereas the two studies of acute otitis media showed a significant increase. Pharyngitis, conjunctivitis, and urinary tract infections showed non-significant higher prescribing in the TH group. Bronchitis showed no change in prescribing. CONCLUSION: The impact of TH on prescribing appears to vary between conditions, with more increases than reductions. There is insufficient evidence to draw strong conclusions, however, and higher quality research is urgently needed.