Background: Acute hospitalisation can have adverse effects in older adults, notably functional decline. We aimed to summarize evidence on the effects of exercise interventions in acutely hospitalised older adults. Methods: Relevant articles were systematically searched (PubMed, Web of Science, Rehabilitation & Sports Medicine Source, and EMBASE) until 19th March 2020. Randomized controlled trials (RCTs) of in-hospital exercise interventions versus usual care conducted in older adults (> 60yrs) hospitalised for an acute medical condition were included. Methodological quality of the studies was assessed with the PEDro scale. Primary outcomes included functional independence and physical performance. Intervention effects were also assessed for other major outcomes (length of hospital stay, incidence of readmission, and mortality). A meta-analysis was conducted when >= 3 studies analysed the same outcome. Results: Fifteen studies from 12 RCTs (n = 1748) were included. Methodological quality of the studies was overall high. None of the studies reported any adverse event related to the intervention. Exercise interventions improved functional independence at discharge (standardized mean difference [SMD] = 0.64, 95% confidence interval = 0.19-1.08) and 1-3 months post-discharge (SMD = 0.29, 95%CI = 0.13-0.43), as well as physical performance (SMD = 0.57, 95%CI = 0.18-0.95). No between-group differences were found for length of hospital stay or risk of readmission or mortality (all p > 0.05). Conclusions: In-hospital supervised exercise interventions seem overall safe and effective for improving - or attenuating the decline of - functional independence and physical performance in acutely hospitalised older adults. The clinical relevance of these findings remains to be confirmed in future research.