Background Bedside rounds (BR) have been proposed as an ideal method to promote patient-centred hospital care, but there is substantial variation in their implementation and effects. Our objectives were to describe the implementation of BR in hospital settings and determine their effect on patient-centred outcomes.

Methods Data sources included Ovid MEDLINE, Ovid Embase, Scopus and Ovid Cochrane Central Registry of Clinical Trials from database inception through 28 July 2017. We included experimental studies comparing BR to another form of rounds in a hospital-based setting (ie, medical/surgical unit, intensive care unit (ICU)) and reporting a quantitative patient-reported or objectively measured clinical outcome. We used random effects models to calculate pooled Cohen's d effect size estimates for the patient knowledge and patient experience outcome domains.

Results Twenty-nine studies met inclusion criteria, including 20 from adult care (17 non-ICU, 3 ICU), and nine from paediatrics (5 non-ICU, 4 ICU), the majority of which (n=23) were conducted in the USA. Thirteen studies implemented BR with cointerventions as part of a ‘bundle’. Studies most commonly reported outcomes in the domains of patient experience (n=24) and patient knowledge (n=10). We found a small, statistically significant improvement in patient experience with BR (summary Cohen’s d=0.09, 95% CI 0.04 to 0.14, p<0.001, I²=56%), but no significant association between BR and patient knowledge (Cohen’s d=0.21, 95% CI −0.004 to –0.43, p=0.054, I²=92%). Risk of bias was moderate to high, with methodological limitations most often relating to selective reporting, low adherence rates and missing data.

Conclusions BR have been implemented in a variety of hospital settings, often ‘bundled’ with cointerventions. However, BR have demonstrated limited effect on patient-centred outcomes.