Reprinted from Research News and Opportunities in Science and Theology with permission.

Many medical and independent studies have touted the health benefits of prayer, meditation and other religious activities, but recent research suggests these findings may be flawed.

Studies on meditation continually claim that the practice has positive effects on reducing stress, tension, anxiety and panic, as well as lowering high blood pressure and relieving chronic pain. Many people use the calming mental exercises to help with headaches and respiratory problems like asthma. Others use it as an enhancement to cancer treatments. Overall, the practice seeks to alleviate factors that can aggravate a variety of medical conditions. Peter H. Canter, research fellow in complementary medicine at Peninsula Medical School in Exeter, England, claims that research on the direct link of meditation and health is fraught with methodological problems. In many cases, the studies are not randomized; researchers select favorably predisposed subjects whose expectations of benefits exceed those of control subjects. In his article, "The therapeutic effects of meditation," published in the British Medical Journal, Canter said that trials of transcendental meditation, when controlled at all, often compare participants who volunteered to be meditators with non-meditators, or long-term meditators with novices. These trials did not check for differences between people who choose to learn meditation techniques and those who do not, or between people who persist with the practice and those who abandon it. In some cases, the research was conducted by people directly involved in organizations keen on proving the unique value of meditation, particularly transcendental meditation, which involves the mental repetition of a mantra. Subjects often partake in a variety of unrelated practices that create uncontrolled variables, making it difficult to attribute health benefits specifically to meditation. The use of multiple cointerventions, high attrition and inadequate statistical analysis has weakened the findings, said Canter. For example, if an experimental group receives meditation training and an altered diet while the control group receives nothing or sham meditation, and a positive change is observed in the experimental group, it is impossible to attribute the change to meditation or to diet or to the combination - the separate effects of the two treatments are obscured. "Additionally, when a large number of subjects drop out of a trial prior to completion, " said Canter, "the generalizability of the results is questionable. Are those who remain different in some way from the general population? If so many people give up meditating, is it a practical and effective treatment that can be recommended with any confidence?" Canter's article also indicates that results can be made to appear more favorable by selectively using statistics. "For example, researchers may only report statistically significant changes in the experimental group without comparing this change to that in the control group," he said. "In order to conclude that the treatment may be effective, the change needs to be significantly greater in the experimental group than in the control group." Jonathan C. Smith, founder and director of the Stress Institute and Distinguished Professor of Psychology at Roosevelt University in Chicago, said Canter is both right and wrong. "Meditation alone is only moderately effective," he said. "But it's not about whether it does or doesn't work; it's part of a larger picture." Smith listed six techniques that work together to provide the complete relaxation experience and are scientifically substantiated: progressive muscle relaxation, autogenic training, breathing exercises, yoga stretching, imagery and visualization and meditation. "Like food groups, you can't rely on one system alone," he said. He agrees that some studies are flawed. "Research that uses self-selected meditators, or studies Catholic or Buddhist monasteries, is not a good model," he said. "A lot of studies don't use that design, though, and don't suffer from those biases." He also said, "The question of meditation is more complex. The validity of these studies is a 20th-century conundrum that still troubles a lot of experts that should know better."

According to Canter, though, the overall evidence for the therapeutic effectiveness of any type of meditation is weak, and evidence for any specific effect above that of credible control interventions even more so. He said he does believe, however, that future trials with improved design may yet provide more concrete results linking health and meditation practices.