Skip to main content
. 2026 Jul 13;17:1887582. doi: 10.3389/fpsyg.2026.1887582

Figure 7.

Subgroup forest plot with violin-shaped distributions displaying standardized mean differences and 95% confidence intervals across country, exercise frequency, intervention duration, session length, age group, intervention type, comparator type, population type, and outcome scale. Columns report the model, heterogeneity (I²), k (number of effect sizes), N (number of participants), GRADE rating, P value, and P-interaction. Most subgroup estimates favor traditional Chinese mind–body exercise, with low or very low certainty. Only country shows a statistically significant between-subgroup difference.

Subgroup analyses of intervention effects on depressive symptoms in middle-aged and older adults. k, number of effect sizes; N, total number of participants; I2, heterogeneity statistic; GRADE, certainty of evidence; P-int, test for between-subgroup differences. Negative SMD values indicate greater reductions in depressive symptoms, favoring traditional Chinese mind–body exercise. The model column indicates the meta-analytic model used for each subgroup. The p-value represents the statistical significance of the pooled effect within each subgroup, whereas P-int represents the statistical significance of differences between subgroups. Combined intervention refers to interventions in which traditional Chinese mind–body exercises is delivered alongside other active components or co-interventions, such as usual care, psychological support, music therapy, Chinese medicine, rehabilitation programs, or continuing-care management; *p < 0.05, **p < 0.01, ***p < 0.001.