Randomized Trial of Therapy Dogs Versus Deliberative Coloring (Art Therapy) to Reduce Stress in Emergency Medicine Providers

ABSTRACT Objective Cognitive stress during shift work contributes to burnout in emergency department (ED) workers. We hypothesize that if physicians and nurses interact with a therapy dog for 5 minutes while on ED shift, both their perceived and their manifested stress levels will decrease. Methods In this single‐center, prospective, randomized controlled clinical trial (NCT03628820), we tested the effectiveness of therapy dogs versus coloring a mandala and versus no intervention (control) on provider stress. Consenting emergency medicine physicians and nurses provided three self‐reported assessments of stress and saliva samples at the start (T1), at the middle (T2), and near the end (T3) of shift. Thirty minutes prior to T2, participants were randomized to either interacting with a therapy dog or coloring for 5 minutes; controls had neither. Stress was assessed on visual analog scale (VAS, 0–100 mm) and with salivary cortisol (Salimetrics) and the modified Perceived Stress Scale (mPSS‐10). To assess potential change in participant behavior, patients of providers in either group were asked to complete an internally derived survey of empathic behaviors displayed by providers at T1 and T3. Results We enrolled 122 providers (n = 39 control, n = 40 coloring, n = 43 dog); 48% were residents, and 60% enrolled on an evening shift. At T1, mean (±SD) VAS score was not different between groups (18.2 [±17.8] mm). At T3, VAS tended to increase with coloring (24.5 mm), remain unchanged in controls (20 mm), and decreased slightly with dogs (13.6 mm, p = 0.018 vs. coloring, Tukey’s post hoc). Salivary cortisol levels were consistently highest at the beginning of each providers’ shift and were significantly decreased versus control in both the dog and the coloring groups (p