In China (and increasingly elsewhere too), the gentle, meditative exercise of tai chi is being promoted and used for disease prevention, particularly for the prevention of cardiovascular disease (CVD). But are these exercises effective? We carried out a Cochrane review to find out.
We searched both English language and Asian electronic databases as well as trial registers and reference lists for relevant studies. No language restrictions were applied. We considered randomised clinical trials (RCTs) of tai chi lasting at least three months and involving healthy adults or adults at high risk of CVD. The comparison groups received no or only minimal interventions. Our outcome measures were CVD clinical events and CVD risk factors. We excluded trials involving multifactorial lifestyle interventions or focusing on weight loss. Two reviewers independently selected trials for inclusion, abstracted the data and assessed the risk of bias of each included study.
We identified 13 trials with a total of 1520 participants and three on-going studies. All of them had at least one domain with unclear risk of bias, and some were at high risk of bias. Duration and style of tai chi differed between trials. Seven studies recruited 903 healthy participants, the other studies recruited people with hypertension, elderly people at high risk of falling, and people with ‘liver or kidney yin deficiency syndromes’.
No studies reported on cardiovascular mortality, all-cause mortality or non-fatal events as most studies were short-term. There was also considerable heterogeneity between studies, which meant that it was not possible to combine studies statistically for cardiovascular risk. Nine trials measured systolic blood pressure (SBP), and 6 of them found reductions in SBP. Two trials found no clear evidence of a difference, and one trial found an increase in SBP with tai chi. A similar pattern was seen for diastolic blood pressure (DBP): three trials found a reduction in DBP, while three found no clear evidence of a difference.
Three trials reported lipid levels and two found reductions in total cholesterol, LDL-C and triglycerides, while the third study found no clear evidence of a difference between groups on lipid levels. Quality of life was measured in only one trial: tai chi improved quality of life at three months. None of the included trials reported on adverse events, costs or occurrence of type 2 diabetes.
From these findings, we drew the following conclusions: “There are currently no long-term trials examining tai chi for the primary prevention of CVD. Due to the limited evidence available currently no conclusions can be drawn as to the effectiveness of tai chi on CVD risk factors. There was some suggestion of beneficial effects of tai chi on CVD risk factors but this was not consistent across all studies. There was considerable heterogeneity between the studies included in this review and studies were small and at some risk of bias. Results of the ongoing trials will add to the evidence base but additional longer-term, high-quality trials are needed.”
These findings are somewhat disappointing. Tai chi might convey many health benefits, but whether a reduction of cardiovascular risk is amongst them seems doubtful. Even if a risk reduction were established beyond doubt, one would need to ask whether its effect size is larger than that achievable through regular conventional exercise. In my view, this is unlikely.
Just a quick question – how is ‘risk of bias’ normally assessed? It seems many trials are thrown out from meta analyses because of this, but it’s not clear to me how these decisions are made. Thank you.
the Cochrane Collaboration has a tool for this based on aspects of the methodology as described in the article; normally trials are not excluded for high risk of bias. however, if possible, one does sub-analyses of trials with high vs low risk of bias to see how robust a result might be.
Thank you. Sorry if this isn’t the right place for this discussion, but am I right to assume a trial might be judged to have a high risk of bias if the practitioners and/or study designers involved make their living from the procedure in question? If that is the case, is this not a serious ‘Catch 22’ situation? Are surgical procedures judged in this way? These aren’t facetious questions, I’m just hoping to learn more about this area which is puzzling to me.
NONE OF THE TOOLS I KNOW ACCOUNT FOR WHO THE AUTHOR IS. I THINK IT WOULD BE IMPORTANT [IN ALT MED] BUT IMPOSSIBLE TO IMPLEMENT.
But they do for the practitioners?