Nudging in MicroBiology laboratory Evaluation (NiMBLE): a systematic review and meta-analysis

Antimicrobial stewardship programmes (ASPs) include collaborating with clinical microbiology laboratories to improve antimicrobial use. Several ASP guidelines recommend selective or cascade reporting of antimicrobial susceptibility results-a behavioural nudge that entails modifying choice architecture to guide prescribing without removing clinician autonomy.

We conducted a systematic review and meta-analysis to evaluate the impact of nudging in microbiology reports on antimicrobial use and clinical outcomes.

Data sources: Following our PROSPERO -registered protocol (CRD42024568670), a search was conducted across four databases.

All studies that evaluated clinical microbiology report nudges and prescribing and/or clinical outcomes were eligible.

Studies in humans conducted in all healthcare settings were included.

Selective, framing, and eye-level nudging strategies in microbiology reports.

Risk of bias was assessed using standard tools for each study design.

Primary outcomes included prescribing appropriateness and antimicrobial utilization. Secondary clinical outcomes included Clostridium difficile infection rates, length of stay and mortality. Reported outcomes were pooled using random-effects meta-analysis.

We analysed 45 studies of the 36 339 records identified. Most studies were conducted among adult inpatient populations, in high-income countries, predominantly in North America. For prescribing appropriateness, nudging interventions were associated with a 143% increase (n= 23 studies; OR, 2.43; 95% CI, 1.91-3.10). Nudging reduced utilization by 1.70 (n= 7 studies; 95% CI, 0.17-3.22) days of therapy. There was no evidence of increased harm; length of stay was 1.1 days longer without nudge (95% CI, -0.4 to 2.5); and no impact to mortality (OR, 1.07; 95%CI, 0.65-1.76) or C. difficile infection rates (OR, 1.08; 95%CI, 0.89-1.30).

Nudging improved antimicrobial appropriateness and utilization, supporting its addition to ASPs. Future research should employ prospective designs to mitigate potential biases, and evaluate clinical outcomes, sustainability and generalizability.