Child and Adolescent Mental Health Repeat Visits to the Emergency Department: A Systematic Review

2017 
OBJECTIVES: Repeat visits represent up to 45% of mental health (MH) presentations to emergency departments (EDs) and are associated with higher health care costs. We aimed to synthesize available literature on predictors of pediatric MH repeat ED visits and differences between repeat visitors and nonrepeat visitors. METHODS: A systematic review was performed using PsycINFO, PubMed, and CINAHL databases. Reporting followed the Preferred Reporting Items for Systematic Reviews and Meta-Analyses statement checklist. Methodologic quality was assessed using the following 8 criteria: design, generalizability, breadth of predictors, reporting of effect sizes, additional outcomes, interaction terms, confounding variables, and clear definition of repeat visits. RESULTS: A total of 178 articles were retrieved; 11 articles met inclusion criteria. Quality assessment revealed that all studies used chart review or administrative data. Predictors were grouped into 3 categories: demographic, clinical, and MH care access and utilization factors. Common predictors associated with repeat ED MH visits included socioeconomic status, involvement with child protective services, and previous and current MH service use. For studies using a 6-month repeat window, the most common factors were previous psychiatric hospitalization and currently receiving MH services. Heterogeneity in statistical analyses and determinants explored precluded the use of meta-analysis. CONCLUSIONS: Findings revealed that repeat visits to the ED for MH concerns is a complex phenomenon that can be attributed to various demographic, clinical, and MH care access and utilization factors. To further elucidate the strongest predictors, future prospective research should consider prospective designs and include family factors. Investigating recency and frequency outcomes can also inform clinical practice.
    • Correction
    • Source
    • Cite
    • Save
    • Machine Reading By IdeaReader
    35
    References
    17
    Citations
    NaN
    KQI
    []