Relax ... with Progressive Muscle Relaxation or Yoga

By Dónal P. O'Mathúna, PhD, Dr. O'Mathúna is a lecturer in Health Care Ethics, School of Nursing, Dublin City University, Ireland; he reports no consultant, stockholder, speaker's bureau, research, or other financial relationships with companies having ties to this field of study.

Source: Smith C, et al. A randomised comparative trial of yoga and relaxation to reduce stress and anxiety. Complement Ther Med 2007;15:77-83. Epub 2006 Jun 21.

Abstract: The purpose of this trial was to compare yoga and relaxation as treatment modalities at 10 and 16 weeks from study baseline to determine if either modality reduces subject stress, anxiety, or blood pressure, or improves quality of life. A randomized comparative trial was undertaken comparing yoga with relaxation in 131 subjects from South Australia who had mild-to-moderate levels of stress. Subjects received 10 weekly one-hour sessions of relaxation or hatha yoga. Outcome measures were changes in the State Trait Personality Inventory sub-scale anxiety, General Health Questionnaire, and the Short Form-36. Following the 10-week intervention, stress, anxiety, and quality-of-life scores improved over time. Yoga was found to be as effective as relaxation in reducing stress and anxiety and improving health status on seven domains of the SF-36. Yoga was more effective than relaxation in improving mental health. At the end of the six-week follow-up period there were no differences between groups with regard to stress, anxiety, and five domains of the SF-36. Vitality, social function, and mental health scores on the SF-36 were higher in the relaxation group during the follow-up period. Yoga appears to provide a comparable improvement in stress, anxiety, and health status compared to relaxation.


Numerous strategies are used to help people reduce stress. In spite of widespread use, little research has examined effectiveness. The authors of this report found only two studies examining yoga as a method of reducing stress. While both found that people assigned to yoga benefited more than those in the control group, the studies were small and of poor quality. With relaxation training, two systematic reviews found a variety of techniques more beneficial than the control. This study compared hatha yoga with relaxation training, but did not include a non-intervention group. The authors mentioned that such a control group would be advantageous to ensure that changes are due to the interventions, not confounding factors.

The relaxation technique used was progressive muscle relaxation (PMR). The participants assumed a relaxed position and listened to instructions on tensing and relaxing various muscles. Ocean music was played in the background. During PMR, people were given 10-15 minutes for complete quiet and possible meditation. The sessions lasted one hour and were to be attended once per week; so too for the hatha yoga sessions. These were described in detail, but it was acknowledged that the procedure varied according to people's abilities and needs. The description added that: ''The aim was to achieve appropriate relaxation to start and finish for the classes for 10 min.'' A number of unclear statements of this type interfered with comprehension of the article.

The stress, anxiety, and health status questionnaires were administered at baseline, and blood pressure was measured. The interventions were administered for 10 weeks and all outcomes measured. Six weeks later, all the outcomes were measured again. At the end of the intervention period, yoga was as effective as PMR on all questionnaire outcomes except for one: Blood pressure did not change. After the follow-up period, the groups did not differ on most outcomes, although the relaxation group had better outcomes in three of the eight domains of the SF-36.

However, during the follow-up period, significantly more people continued to use PMR than continued with yoga. The different outcomes may reflect participants' preference for PMR rather than long-term effects of the intervention. The authors note that this is an important finding in itself that may relate to the ease of learning PMR or of incorporating it into people's routines.

All of the outcomes must be interpreted with caution for at least two reasons. Although the protocol was designed to have people come to weekly sessions for the initial 10-week period, participants in both groups attended an average of five sessions. People gave understandable reasons for not attending, but the implications of deviating from the study design were not discussed. In addition, the researchers calculated that 80% power at the 0.05 level would require 200 participants. Only 131 subjects were randomized and yet no discussion was given of the implications of this for the significance of the results.

The authors conclude that yoga and relaxation produce similar improvements in stress, anxiety, and health status. They repeatedly referred to PMR as ''relaxation'' (even in the title), which fails to make clear that they were comparing yoga to one specific approach to relaxation. Yoga itself is believed to be beneficial because it can induce a relaxation response. While the study was designed to focus on a comparison of the two approaches, some description of the actual outcomes compared to baseline would have been welcome. The data given in the article did not permit an examination of the clinical significance of the changes found.

The researchers are to be commended for taking important steps to improve the quality of the study. They used an intention-to-treat analysis and well-validated outcome measurement instruments. It was not possible to blind the participants to the intervention, but the study analyst was blinded. Randomization was done appropriately and carefully concealed. Drop-out rate overall was relatively low (11% at the end of 16 weeks), although attendance at the relaxation sessions was only 50%. The study design provides a good example of a comparative randomized trial of two complementary therapies. However, in carrying out the study, changes to the study design introduced significant weaknesses that impact the confidence we have in acting upon the stated conclusions.