Seven RCTs were included (n=1,972 participants). The overall methodological quality of the included trials was good: randomisation was adequate in all trials; concealment of allocation was good (well covered) in three RCTs and adequate in one RCT; blinding was good (well covered) in three RCTs and adequate in three RCTs. Drop-out rates ranged from 2 to 29%. Sample sizes ranged from 112 to 702 participants.
Number of people falling with one or more falls (four RCTs): Results were mixed. Two trials reported a reduction in the proportion of patients with one or more falls in tai chi compared to control groups, but only one trial reported the reduction as statistically significant. The other two trials reported no significant difference between tai chi and control. In a subgroup analysis, one of these trials reported a significant reduction in the tai chi group among the pre-frail elderly but a significant increase in the frail elderly.
Adjusted risk of one or more falls (six RCTs): Three RCTs reported a significant reduction in the adjusted risk of one of more falls in tai chi groups compared to control groups. The other three trials reported no significant difference between tai chi and control interventions.