After the deleterious effects of the COVID-19 pandemic on healthcare worker mental health, we tested the effectiveness of an interactive chatbot, Vitalk, for improving wellbeing and resilience among healthcare workers in Malawi, a country with few mental health professionals. We conducted a randomized, controlled trial (RCT) to investigate our hypothesis that Vitalk is more effective in improving mental health and resilience outcomes than passive Internet resources. For our 2-arm, 8-week, parallel RCT (ISRCTN Registry: trial IDISRCTN16378480), we recruited participants from 8 professional cadres from public and private healthcare facilities. The treatment arm used Vitalk; the control arm received links to Internet resources. Of 1,584 participants, 512 completed baseline and endline assessments. Six assessments provided outcome measures for: anxiety (GAD-7); depression (PHQ-9); burnout (OLBI); loneliness (ULCA); resilience (RS-14); and resilience-building activities. We analyzed effectiveness using mixed-effects linear models, effect size estimates, and reliable change in risk levels. Results from mixed-model analyses support our hypothesis. Difference-in-differences estimators showed that Vitalk reduced: depression (-0.68 [95% CI -1.15 to -0.21]); anxiety (-0.44 [95% CI -0.88 to 0.01]); and burnout (-0.58 [95% CI -1.32 to 0.15]). Changes in resilience (1.47 [95% CI 0.05 to 2.88]) and resilience-building activities (1.22 [95% CI 0.56 to 1.87]) were significantly greater in the treatment group. We observed no treatment effect on loneliness. Our RCT produced a medium effect size. This is the first RCT of a mental health app for healthcare workers during the pandemic in Southern Africa combining multiple mental wellbeing outcomes, and measuring resilience and resilience-building activities. A significant number of participants could have benefited from mental health support (1 in 8 reported anxiety and depression; 3 in 4 suffered burnout; and 1 in 4 had low resilience). Such help is not readily available in Malawi. Vitalk has the potential to fill this gap.

No abstract