ObjectivesDespite documented increases in emergency department (ED) mental health (MH) presentations, there are inconsistent findings on the characteristics of patients with repeat presentations to pediatric EDs (PEDs) for MH concerns. Our study sought to explore the characteristics of MH patients with repeat PED visits and determine predictors of return visits, of earlier repeat visits, and of more frequent repeat visits.MethodsWe examined data collected prospectively in a clinical database looking at MH presentations to a crisis intervention program housed within a PED from October 2006 to December 2011. Predictive models based on demographic and clinical variables were constructed using logistic, Cox, and negative binomial regression.ResultsA total of 4,080 presentations to the PED were made by the 2,900 children and youth. Repeat visits accounted for almost half (45.8%) of all presentations. Multivariable analysis identified five variables that independently predicted greater odds of having repeat presentations, greater risk of earlier repeat presentations, and greater risk of frequent repeat presentations. The five variables were: female, living in the metropolitan community close to the PED, being in the care of child protective services, taking psychotropic medications, and presenting with an actionable need in the area of mood disturbances.ConclusionsRepeat visits account for a large portion of all MH presentations to the PED. Furthermore, several patient characteristics are significant predictors of repeat PED use and of repeating use sooner and more frequently. Further research is needed to examine interventions targeting this patient group to ensure appropriate MH patient management.

Objectives This effectiveness study aimed to evaluate the clinical use of the HEADS-ED tool for patients presenting to a pediatric emergency department (PED) for mental health (MH) care. Methods In this pragmatic trial, PED physicians used the HEADS-ED to guide their assessment and identify areas of MH need in 639 patients (mean [SD], 15.16 [1.40] years; female, 72.6%) who presented to the emergency department with MH concerns between May 2013 and March 2014. Results The HEADS-ED guided consultation to psychiatry/crisis, with 86% receiving a recommended consult. Those with a HEADS-ED score of greater than or equal to 8 and suicidality of 2 (relative risk, 2.64; confidence interval, 2.28–3.06) had a 164% increased risk of physicians requesting a consult compared with those with a score of less than 8 or greater than or equal to 8 with no suicidality of 2. The HEADS-ED mean score was significantly higher for those who received a consult (M = 6.91) than those who did not (M = 4.70; P = 0.000). Similarly, the mean score for those admitted was significantly higher (M = 7.21) than those discharged (M = 5.28; P = 0.000). Agreement on needs requiring action between PED physicians and crisis intervention workers was obtained for a subset of 140 patients and ranged from 62% to 93%. Conclusions Results support the HEADS-ED's use by PED physicians to help guide the assessment and referral process and for discussing the clinical needs of patients among health care providers using a common action-oriented language.