Efficacy of a smartphone app to improve mental health among emergency service workers: a randomised controlled trial

Background Emergency service workers (ESWs) are routinely exposed to highly stressful and potentially traumatic events leading to high rates of psychological distress. Early intervention is vital to prevent chronic impairment and/or psychiatric disorders, with digital health innovations (e.g., smartphone apps) offering a potential means of scaling such intervention while overcoming barriers to help-seeking in this population. This study aims to evaluate the efficacy of an app designed to reduce psychological distress and related outcomes in ESWs. Methods We conducted a randomised controlled trial with ESWs experiencing psychological distress (Kessler Psychological Distress Scale; K10) score >15. Participants were assigned to the intervention group, a full version of the Build Back Better app (including mindfulness, behavioural activation, trauma-focused cognitive therapy skills, activity and mood monitoring, and healthy coping strategies) or a mood/activity tracking-only version of the same app. Assessment occurred via online self-report questionnaires at baseline (T0) and at 1- and 3-month post-baseline (T1, T2) timepoints. The primary outcome was the K10 score at T2. Linear mixed model analyses were conducted based on the intention-to-treat principle. Results N = 880 ESWs were randomized to the full (n = 440) or tracking-only (n = 440) condition. There was no Time x Condition effect for K10 scores, with both conditions showing similar improvements from baseline to 1-month (all ps<.001), and 3-month follow-up (all ps<.001). Exploratory analyses of moderators (engagement; baseline severity) found generally no significant differences in K10 score decreases across different levels of engagement in the full-app condition, however, app use was markedly low overall. At higher baseline psychological distress levels, there was a significant decrease in posttraumatic stress disorder symptoms from baseline to 1-month for the full-app group but not the tracking-only group (p=.002, d=−1.43). Conclusion Despite consistent improvement across both app conditions, the minimal between group differences found here highlight the difficulties in developing effective, scalable resources for ESWs and the limitations of unguided digital programs more broadly.