Strengthening the evidential basis of ward-round safety interventions

Correspondence to: Yuquan Chen (email: yuquan.chen@monash.edu)

School of Public Health and Preventive Medicine

Faculty of Medicine

Nursing & Health Sciences

Monash University

553 St Kilda Road

Melbourne VIC 3004

Australia

BJS, https://doi.org/10.1093/bjs/znaf041, published 09 April 2025

We read with great interest the systematic review and meta-analysis by Treloar et al.1 . The authors have collated an impressive body of evidence on a topic of clear clinical relevance; nevertheless, several methodological points merit clarification. First, publication bias was not assessed. We replicated the primary random-effects model using the data provided and constructed funnel plots. Fig. 1a indicated asymmetry suggestive of small-study effects, and a trim-and-fill procedure imputed two potentially missing studies. After adjustment, the pooled odds ratio attenuated to 2.776 (1.161–6.637), Fig. 1b, implying the uncorrected estimate may overstate the benefit. Reporting both original and bias-adjusted results would enhance transparency and align with PRISMA guidance. Second, extreme statistical heterogeneity persists (I² = 96.6%) yet no meta-regression or predefined subgroup analyses (e.g. by intervention type, ward setting, or implementation fidelity) were undertaken; exploring these modifiers could identify contexts where checklists are most effective. Third, 72 uncontrolled before-and-after studies were pooled with 12 randomized trials without GRADE downgrading for design limitations, so causal inference remains uncertain. Fourth, ROB-2 and ROBINS-I assessments were summarized qualitatively and not incorporated into sensitivity analyses, contravening recommendations that synthesis account for study-level bias. Fifth, structural (e.g. electronic prompts) and behavioural (e.g. communication checklists) interventions were combined despite differing mechanisms; separating them would offer clearer implementation guidance. Addressing these issues - systematic appraisal of publication bias with trim-and-fill, investigation of heterogeneity, design-specific synthesis, bias-adjusted sensitivity analyses, and a refined intervention taxonomy - would strengthen the validity and applicability of this valuable work. Nevertheless, we emphasize the study remains a commendable contribution: by consolidating evidence that structured ward-round practices improve documentation and reduce adverse events, it offers an actionable roadmap for clinicians and administrators seeking to enhance bedside safety and care quality.