Tai chi or health education for older adults with hypertension: effects on mental health and psychological resilience to COVID-19

Abstract

Objective: To compare the effectiveness of 12 weeks of community-based, in-person, group Tai Chi (TC) and Health Education (HAP-E) in improving health and wellbeing in older adults with hypertension and in promoting psychological resilience during COVID-19.

Methods: A 12-week randomized controlled trial (RCT) in San Diego County, USA. Self-reported depressive symptoms, anxiety, sleep disturbances, gratitude, resilience, mental and physical health were assessed in-person pre- and post-intervention, and by long-term follow-up surveys during COVID-19. Linear mixed-effects models were used to assess study arm differences over time and logistic regression to identify predictors of positive intervention response.

Results: Of 182 randomized participants (72.6 ± 7.9 yrs; 72% female), 131 completed the intervention. Modest improvements in health and wellbeing occurred post-intervention in both arms (Cohen's d: TC = 0.38, 95% CI: 0.25-0.51; HAP-E = 0.24, 0.11-0.37), though positive intervention responses were more than twice as likely in TC (OR = 2.29, 1.07-4.57). Younger age, higher anxiety, and poorer mental health at baseline predicted greater odds of response. Small declines in health and wellbeing were reported at the first COVID-19 follow-up, with smaller declines in the TC arm (Cohen's d: TC = -0.15, -0.31-0.00; HAP-E = -0.34, -0.49 to -0.19). Health and wellbeing stabilized at the second COVID-19 follow-up. Most participants (>70%) reported that the interventions benefitted their health and wellbeing during COVID-19.

Conclusion: TC and HAP-E improved health and wellbeing, though TC conferred greater odds of an improved mental health response. Declines in health and wellbeing were observed at pandemic follow-up, with smaller declines in the TC arm, suggesting increased resilience.

Trial registration: ClinicalTrials.gov NCT02761603.

Keywords: COVID-19; Tai Chi; hypertension; psychological well-being; resilience.

Figure 1.. CONSORT diagram.

Figure 2.. Effect sizes and meta-analytic summaries for health and wellbeing outcomes across study assessments.

Effect size (Cohen’s d) estimates and confidence intervals (CI) for change differences were derived from linear mixed-effects model contrasts, adjusted for age, sex, race, education, marital status, and intervention attendance. Estimates for BDI-II, PROMIS-A, and PROMIS-SD were reversed such that positive values reflect improvement between study assessments. Pooled effect size estimates (meta-analysis) were computed using a fixed-effects meta-analytic method. T1: pre-intervention baseline assessment. T2: post-intervention assessment. T3: first remotely administered survey assessment. T4 : second remotely administered survey assessment. SF-20-MH: Mental Health subscale of the RAND Short Form SF-20. SF-20-PH: Physical Health subscale of the RAND SF-20. PROMIS-A: Patient Reported Outcomes Measurement Information System Anxiety Short Form 8a. PROMIS-SD: Sleep Disturbances Short Form 8a. CD-RISC: Connor-Davidson Resilience Scale. GQ-6: Gratitude Questionnaire 6-item Form. BDI-II: Beck Depression Inventory II. TC: Tai Chi intervention arm. HAP-E: Health Education intervention arm.