Psychological problems and burnout among healthcare workers: impact of non-pharmacological lifestyle interventions.

Abstract

Objective: To evaluate role of rajyoga meditation (RYM) versus stress management counselling (SMC) in addressing burnout syndrome and resultant improvement in electrocardiogram (ECG) so as to automate burnout prediction from raw ECG data with machine learning (ML).

Methods: Healthcare providers were assigned to two groups: RYM (n=100) or SMC (n=102). Subjects in RYM received rajyoga for 3 months including one week offline and thereafter, virtual mode. SMC group received counselling for 1 day in offline mode and thereafter, received positive thoughts on a weekly basis. All subjects were assessed for psychological (depression, anxiety, stress scale-21 (DASS-21) and burnout syndrome (Mini Z questionnaire) along with 12-lead ECG at baseline after 4 weeks, and after 12 weeks. Based on response on question 3 of the Mini-Z questionnaire, participants were classified either as burnout or satisfied.

Results: RYM group showed significant reduction in depression, anxiety, and stress in comparison to SMC group. Burnout results display significant reduction in the RYM group in comparison to SMC group. Reduction in burnout and enhancement in satisfaction from visit-1 to visit-3: burnout visit-1 (27.2%), visit-2 (23.8%), visit-3 (19.3%) and, satisfaction visit-1 (72.8%), visit-2 (76.2%), and visit-3 (80.7%). ML algorithms could identify burnout patients using the raw ECG data with time-series features based classifier performing better than Ultra Short HRV features based ML classifier model.

Conclusion: AI based early diagnosis of heart's healthy status using ECG analysis may prevent development of cardiovascular disorder in the long run.