Many people who meditate believe that the practice makes them healthier and happier, and a growing number of studies suggest the same. Yet some scientists have argued that much of this research has been poorly designed. To address this issue, Johns Hopkins University researchers carefully reviewed published clinical trials and found that although meditation seems to provide modest relief for anxiety, depression and pain, more high-quality work is needed before the effect of meditation on other ailments can be judged.

Madhav Goyal, an assistant professor of medicine at Johns Hopkins, and his colleagues identified 47 clinical trials published through 2012 that evaluated the effects of meditation on individuals with diagnosed health problems. They included only trials in which subjects were randomly assigned to a group that either meditated or participated in a control intervention, such as cognitive-behavior therapy or training to improve attention. More important, to make for a fair comparison, the control condition had to require a similar amount of time and focus as meditation did. Goyal and his colleagues also considered whether the researchers attending to the subjects knew what intervention they had received; ideally they should not, because this knowledge can influence how researchers interact with and assess subjects. Only 3 percent of meditation studies met these stringent criteria.

Describing their results in January in JAMA Internal Medicine, the researchers found moderate evidence that mindfulness meditation alleviates pain, anxiety and depression—the latter two to a similar degree as antidepressant drug therapy. Mindfulness meditation, the most widely researched approach, requires focusing one's attention on experiencing the present moment. The scientists did not have enough data to assess other common claims of its benefits, including that it improves mood or attention, or other forms of meditation, such as mantra-based practices.

Goyal argues the lackluster results simply reflect the fact that there is not enough evidence to reach other conclusions, in part because funds for high-quality meditation research are hard to come by. “That's part of the reason why the trials that we're seeing have relatively small sample sizes, and many of them have problems with their quality,” he says. Plus, meditation may provide broad lifestyle benefits that go beyond treating disease and are thus difficult to measure. Allan Goroll, a professor of medicine at Harvard University, who published a commentary in the journal at the same time, hopes that the results—or lack thereof—“will be a stimulus for scientists to address these questions in a scientific way,” he says. “We need to apply the scientific method to therapies both conventional and unconventional so we can find out what works.