Lifestyle and exercise

Yoga 'can help stroke patients improve balance'

“Yoga helps stroke patients recover balance,” The Daily Telegraph headlines, saying that stroke victims who took an eight-week course of yoga had better balance and felt more able in their lives as a whole.

The headline is in fact a deeply unbalanced summary of research that actually found no significant improvement to balance in people who were assigned to yoga therapy.

Of 47 people in the study, 37 were assigned to receive twice-weekly yoga sessions delivered by a trained therapist for eight weeks. The other participants were part of a control group and did not receive yoga therapy. The study used a scale to assess balance and asked participants other questions about fear of falling and quality of life.

Despite the news hype, the study found no difference in the results between the groups. While people in the yoga group did experience modest improvement in balance and a reduced fear of falling, the difference between their scores and the control group scores was small and not statistically significant. So there is no hard evidence that yoga was responsible for the improvement.

This study provides no evidence that yoga is better for balance in people who have had a stroke than their usual care. However, any form of exercise that a person feels comfortable to participate in post-stoke, and feels is helping them, is likely to be a good thing. 

Read more advice about recovering from the effects of a stroke.

Where did the story come from?

The study was carried out by researchers from the Roudebush Veterans Administration Medical Center, Indiana University and other institutions in Indianapolis, US. Funding was provided by the US Department of Veterans Affairs. One of the study authors is a registered yoga therapist and owns a not-for profit yoga therapy organisation. This arguably represents a conflict of interest.

The study was published in the peer-reviewed medical journal Stroke.

The media does not present the main result of this very small trial – that it actually found no difference in balance between those who took part in yoga and those who did not.

What kind of research was this?

This was a randomised controlled trial (RCT) that aimed to test whether yoga-based rehabilitation improved:

  • balance
  • confidence in balance
  • fear of falling
  • quality of life after stroke

The researchers report that as balance impairment is common after stroke, modified yoga may be able to improve this, but so far there is limited literature related to yoga after stroke. This study aimed to develop and test an eight-week yoga-based rehabilitation intervention to improve balance and quality of life after stroke.

A randomised controlled trial is the best way of investigating whether a particular intervention (in this case yoga) affects an outcome (balance and other factors) compared with a comparator (waiting-list control). This is a pilot trial, which means if results are promising larger trials may follow.

What did the research involve?

The study recruited adults (average age 63) who had suffered a stroke at least six months previously. Eligible participants had to have completed all previous stroke rehabilitation, had to be able to stand (supported or unsupported) and had to not be suffering from other significant medical illness, such as heart disease or respiratory problems. Of 222 who were potentially eligible, only 47 were eligible and agreed to participate, 37 of whom were randomised to yoga and 10 to a waiting-list control group, who received usual care.

Yoga involved twice-weekly, hour-long sessions given over eight weeks (16 sessions in total). The intervention was delivered by a trained therapist and included postures, breathing and meditation in sitting, standing and lying down positions. All postures were chosen based on the researchers’ previous experience with stroke, and other evidence suggesting that balance could be improved through a focus on hip and ankle flexibility and strength. Over the eight-week period, yoga sessions increased in intensity to allow for progressive improvement. The yoga group was subdivided (by randomisation) into those who received the group yoga only, and those who also received an audio recording of yoga relaxation to use at home, but they were analysed together as one yoga group.

Assessments were completed by a trained therapist at the start of the study and at eight weeks after the yoga sessions had finished. As the therapist had been involved in delivery of the yoga, they were aware of which group participants had been assigned (i.e. the study was not blinded). A validated scale (the modified Rankin Scale, mRS) was used to assess degree of disability after stroke. Balance was assessed using the Berg Balance Scale (BBS) – a 14-item physical performance measure of balance, which has been validated for assessment after stroke. The BBS has a scoring range from 0 to 56, with higher scores indicating better balance. A score of 46 or less is considered to be an individual at risk of falls after stroke. An additional 16-item Activities-specific Balance Confidence Scale was used to measure how able a person feels to maintain balance control during functional tasks (scoring from 0% “no confidence” to “completely confident” 100%). Participants were also asked “Are you worried or concerned about falling?”, to which they gave a yes or no response. Finally, quality of life was measured on a 49-item Stroke Specific QoL scale.

What were the basic results?

Of the yoga group, only 29 completed all eight weeks of assessments (78% of the yoga group). The number of people in the waiting-list control group who completed the study was unclear.

Overall, there were no significant differences at the final assessment between the yoga and the control groups for any of the outcomes assessed:

  • balance
  • confidence in balance
  • fear of falling
  • disability
  • quality of life

Within the yoga group, however, there was significant improvement from the start of the study to the end of the study in:

  • BBS balance score – 41.3 at study start versus 46.3 at study end
  • proportion who reported having fear of falling – 51% at study start versus 46% at study end
  • those who reported being independent – 57% at study start versus 68% at study end

There were no differences from study start to end within members of the control group.

How did the researchers interpret the results?

The researchers conclude that a yoga-based rehabilitation intervention for people six months or more after a stroke has “potential” for improving balance and function. They say that further testing of group yoga-based rehabilitation interventions is warranted.


This small pilot study was well designed and benefits from the use of validated scales to measure balance and function objectively. However, the important points are:

  • Despite the news hype, the study found no actual difference between the yoga group and waiting-list control for any of the balance outcomes assessed.
  • Within the yoga group, although there was a five-point difference in the 56-point score, this was from 41 at the start of the study to 46 after. A score of 46 or less on the Berg Balance Scale is considered to be someone who is at risk of falling, which means it is not clear whether this score change would have made any difference to functioning and balance.
  • Within the yoga group, although there was a significant 5% decrease in the proportion who reported they had a fear of falling, 46% still reported a fear of falling after taking part in yoga, and this remains a substantial percentage.
  • Both participants and assessors were aware of group assignment. There is the possibility that if people were aware that they had received yoga to try to improve their balance, they could have reported better balance and less fear of falling because they were expecting the intervention to have helped them. This could have explained some of the within-group differences in the yoga group, particularly on subjective questions such as fear of falling.
  • This pilot trial was very small, including only 47 participants. Only 29 of the 37 assigned to yoga completed the assessments, which is 78% of the yoga group. This could limit the reliability of the results, as those who dropped out could have had different results, for example, they may have done so because they didn’t feel they were getting a benefit from yoga.

Overall, this study provides no evidence that yoga is more beneficial to people who have had a stroke than usual care. This may be because the study was too small to detect a difference, but larger studies would be needed to assess whether this is the case.

Any form of exercise that a person feels comfortable to participate in after a stroke, and feels is helping them, is likely to be a good thing.  

NHS Attribution