Predicting time to emergency department return for anxiety disorders and acute stress reactions in children and adolescents: a cohort study.

PURPOSE: While studies indicate that few children with anxiety disorders receive specialized mental health care, significant knowledge gaps exist for the mental health care trajectory outside of specialized care. We examined whether physician-based care after an emergency department (ED) visit for mental health care predicted time to ED return.
METHODS: We conducted a retrospective cohort study using administrative data from Alberta, Canada. Included in the cohort were 8075 children <18 years discharged from an ED (104 EDs in total), April 2002-September 2010, following a visit for an anxiety disorder or acute stress reaction. We used a multivariable Cox proportional hazards model to estimate time to ED return for mental health care (censored at 90 days). The variables of interest were: number and type of physician visits 30 days after index visit. We adjusted for sociodemographics, clinical acuity, comorbidity, and recent history of ED mental health care in the model.
RESULTS: Within 30 days of ED discharge, multiple physician follow-up visits were associated with shorter return time (adjusted HR 1.24, 95 % CI 1.08-1.43). A single physician follow-up visit was associated with longer return time (adjusted HR 0.68, 95 % CI 0.58-0.79). Physician follow-up visits for mental health care were associated with shorter return time (adjusted HR 2.5, 95 % CI 2.21-2.83).
CONCLUSIONS: Following an index visit, ED return by children with anxiety disorders or acute stress reactions is associated with specific characteristics of subsequent physician visits. Improving physician use of evidence-based treatment and family access to coordinated services may reduce costly hospital-based care.