Evaluation of an App-Delivered Psychological Flexibility Skill Training Intervention for Medical Student Burnout and Well-being: Randomized Controlled Trial
Elizabeth Ditton,Brendon Knott,Nicolette Hodyl,Graeme Horton,Christopher Oldmeadow,Frederick Rohan Walker,Michael Nilsson
DOI: https://doi.org/10.2196/42566
2023-02-06
Abstract:Background: Physician burnout is a common problem, with onset frequently occurring during undergraduate education. Early intervention strategies that train medical students in psychological flexibility skills could support well-being and mitigate burnout risks associated with unmodifiable career stressors. There is a need for randomized controlled trials to assess effectiveness. As psychological flexibility varies contextually and among individuals, tailoring interventions may improve outcomes. Smartphone apps can facilitate individualization and accessibility, and the evaluation of this approach is an identified research priority. Objective: This study aimed to evaluate the effectiveness of a stand-alone app-delivered Acceptance and Commitment Training intervention for improving medical students' self-reported burnout, well-being, psychological flexibility, and psychological distress outcomes. We aimed to explore whether an individualized app would demonstrate benefits over a nonindividualized version. Methods: This parallel randomized controlled trial was conducted with a sample of medical students from 2 Australian universities (N=143). Participants were randomly allocated to 1 of 3 intervention arms (individualized, nonindividualized, and waitlist) using a 1:1:1 allocation ratio. Individualized and nonindividualized participants were blinded to group allocation. The 5-week intervention included an introductory module (stage 1) and on-demand access to short skill training activities (stage 2), which students accessed at their own pace. Stage 2 was either nonindividualized or individualized to meet students' identified psychological flexibility training needs. Results: The mean differences in change from baseline between the intervention groups and the waitlist group were not statistically significant for burnout outcomes: exhaustion (primary; individualized: -0.52, 95% CI -3.70 to 2.65, P=.75; nonindividualized: 1.60, 95% CI -1.84 to 5.03, P=.37), cynicism (individualized: -1.26, 95% CI -4.46 to 1.94, P=.44; nonindividualized: 1.00, 95% CI -2.45 to 4.46, P=.57), and academic efficacy (individualized: 0.94, 95% CI -0.90 to 2.79, P=.32; nonindividualized: 2.02, 95% CI 0.02-4.03, P=.05). Following the intervention, the individualized group demonstrated improved psychological flexibility (0.50, 95% CI 0.12-0.89; P=.01), reduced inflexibility (0.48, 95% CI -0.92 to -0.04; P=.04), and reduced stress (-6.89, 95% CI -12.01 to 5.99; P=.01), and the nonindividualized group demonstrated improved well-being (6.46, 95% CI 0.49-12.42; P=.04) and stress (-6.36, 95% CI -11.90 to -0.83; P=.03) compared with waitlist participants. Between-group differences for the individualized and nonindividualized arms were not statistically significant. High attrition (75/143, 52.4%) was observed. Conclusions: This trial provides early support for the potential benefits of Acceptance and Commitment Training for medical student well-being and psychological outcomes and demonstrates that psychological flexibility and inflexibility can be trained using a smartphone app. Although postintervention burnout outcomes were not statistically significant, improvements in secondary outcomes could indicate early risk mitigation. Replication studies with larger samples and longer-term follow-up are required, and future research should focus on improving implementation frameworks to increase engagement and optimize individualization methods. Trial registration: Australian New Zealand Clinical Trials Registry 12621000911897; https://tinyurl.com/2p92cwrw. International registered report identifier (irrid): RR2-10.2196/32992.