Conceptualizing clinical decision support as complex interventions: a meta-analysis of comparative effectiveness...

Area of Science:

• Health Informatics
• Clinical Decision Support Systems
• Complex Interventions

Background:

• Complex interventions with multiple components are increasingly used as clinical decision support (CDS).
• Electronic health record (EHR) functionality is often utilized for implementing these CDS interventions.
• Understanding the complexity and reporting of CDS interventions is crucial for effective implementation.

Purpose of the Study:

• To determine the proportion of randomized controlled trials (RCTs) involving complex CDS interventions.
• To identify common gaps in reporting the complexity of CDS research.
• To assess the impact of increased complexity on the effectiveness of CDS.

Main Methods:

• Systematic review using the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) reporting tool for complex interventions.
• Random-effects meta-analysis was employed to evaluate the effect of intervention complexity.

Main Results:

• A significant majority (76%) of included studies evaluated complex CDS interventions.
• Reporting gaps were identified, with limited use of analytical frameworks, causal pathways, or detailed theoretical underpinnings.
• A small but statistically significant positive effect was observed with increasing intervention complexity (SMD = 0.147, P < .01).

Conclusions:

• Most CDS interventions should be classified as complex, yet opportunities exist to improve documentation for replication and adaptation.
• Enhanced reporting of the design, implementation, and evaluation of CDS interventions is necessary to realize greater benefits.
• Framing CDS as complex interventions highlights the need for careful contextual and theoretical consideration.