Objective: To identify interventions using wearable accelerometers to measure physical activity and/or sedentary behaviour in adults during hospitalization for an acute medical/surgical condition. Data sources: Four databases were searched in August 2019 (MEDLINE, CINAHL, Scopus, EMBASE). Review methods: Studies were selected if they described an intervention in adults with a medical/surgical condition, and concurrently reported an accelerometer-derived measure of physical activity and/or sedentary behaviour while participants were admitted. Items were screened for eligibility in duplicate. Included studies were synthesized to describe intervention types, feasibility and potential effectiveness. Results: Twenty-two studies were included, reporting on 3357 participants (2040 with accelerometer data). Identified types of interventions were: pre-habilitation ( n = 2) exercise ( n = 3), patient behaviour change with self-monitoring ( n = 6), models of care ( n = 5), implementing system change ( n = 2), surgical technique ( n = 2) patients wearing day clothes ( n = 1) and education about activity in hospital ( n = 1). Of 16 studies that reported intervention effects on physical activity, 11 reported a favourable impact including studies of: pre-habilitation, self-monitoring (accelerometry or an activity whiteboard), physiotherapy, an early mobility bundle, minimally invasive surgery, an education booklet and by implementing system change. Of the six studies that reported intervention effects on sedentary behaviour, there was a favourable impact with an activity whiteboard, models of care and an education booklet. Conclusion: Accelerometer-derived measures of physical activity and/or sedentary behaviour have been used to describe sample characteristics and intervention effects in studies of hospitalized adults. Interventions may involve a range of health professionals, but less is known about sedentary behaviour in this setting.