Answers are grounded in this paper and cite their sources.
Systematic review reveals an overreliance on simple statistics in emergency department behavioral health evaluations, suggesting the need for advanced methods to capture real-world complexity.
Background There have been substantial increases in patients presenting for mental and behavioral health (MBH) conditions to Emergency Departments (EDs) in the U.S. In response, health systems have implemented process-based interventions to improve care delivery. While prior systematic literature reviews (SLRs) have cataloged these intervention types, few have critically assessed the methodological approaches used in assessing their impact.Methods We conducted a SLR of U.S.-based studies evaluating ED process-based MBH interventions, focusing on (1) the clinical leadership involved, (2) the outcome measures reported, and (3) the statistical methods used.Results Twenty-seven articles that met inclusion criteria and most interventions were led by ED teams (66.7%) rather than psychiatric teams (33.3%), and most studies (63.0%) reported at least one time-based outcome measure. Just over half (51.9%) used simple hypothesis testing, 18.5% used descriptive statistics alone, and only 25.9% employed advanced statistical methods.Conclusions Reliance on simple statistical methods limits the ability to account for confounders and the complexity of ED systems. Meaningful evaluation of MBH interventions must consider the spatiotemporal and socio-technical nature of care. Aligning evaluation methodologies with real-world complexity can result in more accurate insights and effective MBH patient care.
No takes yet. Share an insight, caveat, or question.
Koscelny et al. (2026) studied this question.
Synapse has enriched 10 related papers. Consider them for comparative context: