Seven studies (n=344) were included: 2 RCTs (n=100), 2 quasi-experimental studies (n=58) and 3 cross-sectional studies (n=186). The sample size ranged from 20 to 90.
The quality scores ranged from 22 to 28 (mean 25.1) out of 32. Six studies adequately described the characteristics of the participants and all 7 studies clearly defined aerobic capacity. One study reported blinding of the outcome assessor. In the cross-sectional studies, treatment groups were matched on age, gender and body composition.
The pooled ES for cross-sectional studies was large and statistically significant (ES 1.01, 95% CI: 0.37, 1.66; based on n=186), while the ES for experimental studies was small and not significant (ES 0.33, 95% CI: -0.41, 1.07; based on n=158).
The pooled ES (and hence the aerobic capacity) was larger for women (ES 0.83, 95% CI: -0.43, 2.09; based on n=126) than for men (ES 0.65, 95% CI: -0.04, 1.34; based on n=167), but in neither subgroup was the ES statistically significant.
The pooled ES for the physical activity level of the control group showed that t'ai chi significantly improved aerobic capacity in comparison with sedentary controls (ES 0.80, 95% CI: 0.19, 1.41; based on n=253), but not in comparison with other forms of exercise (ES 0.22, 95% CI: -0.81, 1.24; based on n=91).
The pooled ES for type of t'ai chi showed that the classical Yang style significantly improved aerobic capacity compared with control (ES 1.10, 95% CI: 0.82, 1.38; based on n=224), but no significant difference was observed for modified or unspecified t'ai chi (ES -0.17, 95% CI: -0.54, 0.20; based on n=120).
The pooled ES for duration of t'ai chi showed that interventions lasting 52 weeks significantly improved aerobic capacity compared with control (ES 0.94, 95% CI: 0.06, 1.81; based on n=58), whereas 12 to 16 weeks of t'ai chi was associated with significantly less improvement in aerobic capacity versus the control (ES -0.28, 95% CI: -0.53, -0.02; based on n=100).