Therapy Dogs for Anxiety in Children in the Emergency Department: A Randomized Clinical Trial

Abstract

Importance: Prior evidence suggests that the use of therapy dogs in emergency care reduces anxiety in adults, but no trial has tested the use of therapy dogs in emergency care of children. Objective: To examine whether adjunctive use of therapy dogs in standard child-life therapy reduces child-reported and parent-reported child anxiety in a pediatric emergency department (ED). Design, setting, and participants: This randomized clinical trial was conducted from February 1, 2023, to June 30, 2024, at an academic pediatric ED. Children (aged 5-17 years) with suspected moderate to high anxiety were included. Intervention: All participants received standard child-life therapy, and the intervention group was randomly assigned to have exposure to a therapy dog and handler for approximately 10 minutes. Main outcomes and measures: Anxiety was measured using the 0- to 10-point FACES scale (with 0 indicating no anxiety and 10 indicating very severe anxiety) and salivary cortisol concentrations. Measurements were obtained at baseline (T0), 45 minutes (T1), and 120 minutes (T2) for both child and parents. Results: A total of 80 patients (mean [SD] age, 10.9 [3.8] years; 45 [56%] female) were enrolled (40 in the control group and 40 in the intervention group). At T0, the mean (SD) FACES scores were 5.4 (2.8) for child report and 6.4 (2.4) for parent report; the means were not different between groups. From T0 to T1, child-reported anxiety changed by a mean (SD) of -1.5 (3.4) points in the control group vs -2.7 (2.5) points in the intervention group (P = .02, Mann-Whitney U test); similarly, mean (SD) parent-estimated child anxiety changed by -1.8 (2.7) points in the control group vs -3.2 (2.3) points in the intervention group (P = .008). A total of 9 children (23%) in the control group had a greater than 2.5-point decrease in FACES score vs 18 (46%) in intervention group (P = .04, Fisher test). At T2, mean (SD) child-reported FACES scores decreased to 3.6 (3.4) points in the control group and 3.0 (2.7) points in the intervention group (P = .70). A total of 14 control participants (35%) received ketamine, midazolam, lorazepam, or droperidol vs 7 (18%) in the intervention group (P = .08, Fisher test). Child and parent salivary cortisol decreased from T0 to T1 in both groups but was not different between groups. Parental salivary cortisol was significantly consistently higher than their children's salivary cortisol (P < .001, unpaired t test, for comparisons of child vs parent at T0 and T1 in both groups). Conclusions and relevance: This study of adjunctive use of therapy dogs in standard child-life therapy found a modest but significantly greater reduction in both child-reported and parental-reported child anxiety in the pediatric ED for the intervention vs control group. These findings support the use of therapy dogs to help reduce pain and anxiety without the use of chemical or physical constraint.