Yoga and tai chi

Tai chi, also written tai chi chuan or taiji, developed as a Chinese martial art and is now practised mainly as exercise. It consists of slow, continuous sequences in which body weight shifts from one leg to the other, with attention on posture and breathing. Qigong is a related group of Chinese movement and breathing practices. Yoga covers many styles; the trials discussed here mostly used hatha, Iyengar or viniyoga styles, which combine held postures (asanas), breathing work and relaxation [4].

The physical demands differ. Tai chi places continuous load on balance and leg control, while yoga emphasises flexibility and holding positions that require muscular strength. Both are low-impact and can be adapted for older adults and people who cannot run or lift weights. The underlying physical qualities are covered in balance training and flexibility and mobility.

The term mind-body refers to the deliberate focus on breathing and attention. It also makes these practices hard to study: there is no convincing placebo version, so participants know which group they are in, and outcomes such as pain, stress and quality of life are self-reported.

The Shanghai Men's Health Study enrolled men aged 40 to 74 from urban Shanghai and recorded their exercise habits with a validated questionnaire [1]. Among 61,477 men followed for a mean of 5.48 years, 2,421 died. After adjustment for other risk factors, men who exercised regularly had 20% lower all-cause mortality than men who did not exercise (hazard ratio 0.80, 95% CI 0.74 to 0.87). The hazard ratio is the death rate in one group relative to another. For specific activities, the figures were 0.80 (0.72 to 0.89) for tai chi, 0.77 (0.69 to 0.86) for walking and 0.73 (0.59 to 0.90) for jogging, with similar patterns for cancer and cardiovascular deaths. The authors reported that the largest benefit was seen at about 5 to 6 hours of exercise a week.

This is a single study of men in one city, where tai chi is a common morning activity among older and retired people. Men who were already unwell were probably less likely to practise, so part of the difference may reflect illness keeping people from exercise rather than exercise protecting health. The results place tai chi alongside walking and jogging as moderate exercise; they do not indicate a separate effect specific to tai chi.

Falls are a leading cause of injury and loss of independence in older people, and at least one-third of people over 65 living in the community fall each year [2]. A Cochrane review pooled 108 randomised trials of exercise for preventing falls, with 23,407 participants (average age 76, 77% women) in 25 countries. Across all types of exercise, programmes reduced the rate of falls by 23% (rate ratio 0.77, 95% CI 0.71 to 0.83; high-certainty evidence) and the number of people who fell by 15%.

Tai chi reduced the number of people who fell by 20% (risk ratio 0.80, 95% CI 0.70 to 0.91; 8 trials, 2,677 participants; high-certainty evidence) and may have reduced the rate of falls by 19% (rate ratio 0.81, 0.67 to 0.99; 7 trials; low-certainty evidence) [2]. Balance and functional exercise programmes had a similar effect, and programmes combining several types of exercise, usually balance and functional exercise with resistance training, probably reduced falls by 34%. The review was uncertain about the effect of dance or walking programmes.

Tai chi consists of slow, controlled weight transfers from one leg to the other, which train the balance and leg control that fail during a fall. Other measures, including home hazard assessment and medication review, are covered in fall prevention.

For tai chi, a Cochrane review of trials in healthy adults and adults at high cardiovascular risk found 13 small trials with 1,520 participants, all with follow-up of one year or less [5]. None reported deaths, heart attacks or strokes. Results varied so much that they could not be combined. Of nine trials that measured blood pressure, six found a reduction in systolic pressure, two found no clear difference and one found an increase of 5.2 mmHg. Of three trials that measured blood lipids, two found reductions in total and LDL cholesterol and triglycerides. The authors concluded that no conclusions could yet be drawn about tai chi and cardiovascular risk factors.

Yoga has a larger body of trials. A pooled analysis of 44 randomised trials with 3,168 participants, including healthy adults, adults at high cardiovascular risk and people with type 2 diabetes, compared yoga with usual care or no intervention [3]. Yoga lowered systolic blood pressure by 5.85 mmHg (95% CI 2.89 to 8.81) and diastolic by 4.12 mmHg (1.69 to 6.55), and improved heart rate, waist circumference, total cholesterol, HDL cholesterol, triglycerides, HbA1c (a measure of average blood sugar) and insulin resistance. Most trials had a high or unclear risk of bias. Compared with other exercise, yoga differed only in a larger rise in HDL cholesterol (3.70 mg/dL), consistent with yoga acting much like other moderate exercise.

These are changes in risk markers over weeks or months. No trial has shown whether the reductions persist or translate into fewer heart attacks, strokes or deaths.

The most thorough review of yoga for a clinical condition covers chronic non-specific low back pain. A Cochrane review included 21 randomised trials with 2,223 adults from the USA, India, the UK, Croatia, Germany, Sweden and Turkey, most of them women in their 40s or 50s [4]. Compared with no exercise, yoga improved pain by 4.53 points on a 0 to 100 scale at three months (moderate-certainty evidence) and back-related function by 1.69 points on the 24-point Roland-Morris questionnaire (low-certainty evidence). Both changes were smaller than the minimal clinically important differences used in the review, 15 points for pain and 5 points for function.

Compared with back-focused exercise such as physical therapy, yoga showed little or no difference in function at three months [4]. Low-certainty evidence from eight trials indicated more adverse events with yoga than with no exercise, mainly increased back pain (43 per 1,000 compared with 9 per 1,000), but a similar rate to other exercise. All trials were unblinded with self-assessed outcomes, and the authors judged it unlikely that blinded trials would find a clinically important benefit.

The evidence for yoga consists of trials measuring blood pressure, blood lipids, pain and function over weeks or months [3, 4]. None of these trials measured deaths, and the only mortality study of a mind-body practice concerns tai chi in Chinese men [1]. Claims that yoga extends lifespan or reverses biological ageing therefore go beyond the available measurements.

For tai chi, the fall-prevention evidence is the strongest, based on randomised trials rated high-certainty for the number of people who fell [2]. The cardiovascular trials are small, short and inconsistent [5]. For both practices, lack of blinding and self-reported outcomes limit what trials can show about pain, stress and wellbeing. Where the evidence exists, it places yoga and tai chi alongside other moderate exercise, as described in exercise and longevity, rather than indicating effects unique to the practices.