The Effects of a Single Session of Mindful Exercise on Anxiety: A Systematic Review and Meta-Analysis

Abstract Objective To evaluate the effects of a single session of mindful exercise on anxiety and examine whether effects varied according to participant characteristics, mindful exercise exposure, or research design features. Methods Consistent with PRISMA guidelines for conduct and reporting, we identified randomized or non-randomized controlled trials (RCTs) that: (1) sampled healthy or unhealthy participants, (2) used a single session of yoga, Tai chi, or Qigong practice, and (3) assessed self-reported anxiety outcomes. Databases searched included Google Scholar, PubMed, PsycINFO, and Web of Science, from which 18 yoga, 5 Tai chi, and 4 Qigong articles were published before August 31, 2020. Studies involved 2341 participants (1904 in yoga studies, 232 in Tai chi studies, 205 in Qigong studies) were selected. Hedges d effect sizes were calculated. Random effect models and multilevel meta-analysis were used to estimate the pooled effects. I2 tests were applied to assess the heterogeneity. In yoga studies, moderating effects were tested by multilevel meta-regression analysis of putative moderators representing participant characteristics, features of mindful exercise exposure, and aspects of research design. Because of the limited number of Tai chi and Qigong studies, sensitivity analyses were conducted by omitting potential outlying effects and evaluating the overall pooled effect. Study quality was judged using the PEDro scale. Results A single session of yoga reduced anxiety by a heterogeneous (I2 = 82.7%) standardized mean effect size 0.32 (95% CI, 0.16 to 0.48, t = 4.108, p = 0.0002). In the multilevel univariate regression model, reductions were larger in participants who had previous yoga experience, or practiced Hatha yoga, or engaged in higher intensity yoga. Larger effects were also found when yoga was compared to no treatment or minimal treatment control than when yoga was compared to non-yoga exercise and when anxiety was measured 0–10 min immediately after yoga. In the multilevel multiple regression model, the effects remained larger for yoga history and yoga versus no treatment or minimal treatment control compared to yoga versus non-yoga exercise. Meta-regression analysis did not suggest other significant (p > 0.05) moderating effects on anxiety outcomes. The low heterogeneous (I2 = 10%) mean effect of a single session of Tai chi on anxiety was −0.06 (95% Cl, −0.38 to 0.24, t = −0.37, p = 0.68). The homogeneous (I2 = 0%) mean effect of a single session of Qigong on anxiety was 0.70 (95% CI, 0.22 to 1.18, t = 4.06, p = 0.02). Conclusion A single session of yoga may have small-to-moderate efficacy for the management of anxiety, but the limited number of studies for Tai chi and Qigong studies makes it premature to conclude their practical effectiveness. In addition, the overall quality of the included studies was relatively low. Cautions should be taken when applying a single session of mindful exercise to reduce anxiety in practice. Further high-quality randomized controlled trials are needed to mitigate methodological bias typically found in the retrieved studies.