Skip to content

Heated Yoga Linked to Reduction of Moderate to Severe Depressive Symptoms in Randomized Controlled Trial

Key findings

  • The Depression Clinical and Research Program at Massachusetts General Hospital shows that heated Bikram yoga can effectively reduce depressive symptoms
  • Participants experienced a more rapid and significant decline in depressive symptoms, as measured by the Inventory of Depressive Symptomatology scale (IDS-CR)
  • Response and remission rates in the yoga group, 59% of participants experienced a ≥50% decrease in IDS-CR scores, and 44% achieved remission (IDS-CR score <14)
  • Despite the intensive nature of Bikram yoga, involving 90-minute sessions in a 105°F room, there is high acceptability among participants

In 2021, a randomized controlled trial published in the Journal of Affective Disorders showed heated yoga outperformed a waitlist control in a group of women with mild-to-moderate depression. Now, researchers at Massachusetts General Hospital have extended the findings to a larger, mixed-gender and more severely depressed population.

Maren B. Nyer, PhD, director of yoga studies at the Depression Clinical and Research Program at Mass General and assistant professor of Psychiatry at Harvard Medical School, David Mischoulon, MD, PhD, director of the program, and the Joyce R. Tedlow professor of Psychiatry at Harvard Medical School, and colleagues published the details in The Journal of Clinical Psychiatry.


Recruitment for the newer randomized controlled trial occurred in the Boston community between March 2, 2017, and August 5, 2019. Key inclusion criteria were age between 18 and 60 and score ≥23 on the clinician-rated Inventory of Depressive Symptomatology scale (IDS-CR). Instead of an upper limit on symptoms, patients were excluded if they reported having had an active suicidal plan or intent within the past year.

80 individuals were randomly assigned to:

  • Bikram yoga—At least two weekly classes for eight weeks at either of two community-based studios. Bikram yoga is a standardized form of hatha yoga practiced in a 105°F room. The 90-minute sessions involved 26 strenuous yoga postures, preceded and followed by five minutes of breathing exercises
  • A waitlist—Participants were asked not to add additional treatments, yoga practice, or other exercise during their eight-week waitlist period

Participants in either group who were receiving psychotherapy or antidepressants were asked to maintain stable treatment throughout the study. Before starting yoga, participants in the intervention arm completed a 50-minute preparation session with the principal investigator or designee, and during the intervention, they received biweekly telephone calls for support.

A clinician blinded to the randomization arm re-administered the IDS-CR at weeks 1, 3, 5, and 8.

Primary Outcome

65 participants (82% female) were included in analyses (yoga, n=33; waitlist, n=32):

  • IDS-CR scores declined in general but at a faster rate in the yoga group than the waitlist group (time x arm effect: β, −3.09; P<0.001)
  • Antidepressant use status did not influence the effect of randomization status on depressive symptoms over time
  • At week 8, the yoga group had significantly lower scores on the IDS-CR than the waitlist group (Cohen d, 1.04; P<0.001)

Response and Remission

Response to treatment was defined as ≥50% decrease in IDS-CR score and remission was defined as IDS-CR score <14:

  • Response—16 participants (59%) in the yoga group and 2 (6%) in the waitlist group
  • Remission—12 (44%) and 2 (6%), respectively

Adverse Events

In the yoga arm the only adverse event judged to be definitely related to the intervention was gagging, classified as mild. Three mild events (headache, back pain and knee pain) and four moderate events (chills/nausea, dizziness/vertigo, back pain) were considered probably related to yoga.

Feasibility and Acceptability

The yoga participants attended an average of 10 classes (range, 1–30) over the study period. 12 (36%) attended 12 classes of more, which was 75% of the prescribed "dose" of at least 16 classes.

On a 1–10 Likert scale for acceptability, average ratings were:

  • "How much did you enjoy the yoga itself?"—7.17
  • "How much did you enjoy the aftereffects of the yoga?"—8.33
  • "How reasonable did you find the time commitment to be?"—6.90

Fewer Sessions May Suffice

Previous research into yoga for depression has found one to two classes per week to be effective. In this study, heated yoga was associated with reductions in depressive symptoms when attendance was about 1.25 classes per week. A frequency closer to once per week could be a feasible dose for a larger follow-up study.

Learn more about the Depression Clinical and Research Program

Explore research in the Department of Psychiatry

Related topics


Kathleen L. Benson, PhD, John W. Winkelman, MD, PhD, and Atilla Gönenç, PhD, detected similar impairments in four white matter tracts on brain diffusion tensor imaging in individuals with major depressive disorder or primary insomnia compared with healthy controls, suggesting shared neurobiology of the two disorders.


Maurizio Fava, MD, and colleagues created an artificial intelligence–assisted method for predicting each individual's propensity to respond to placebo in randomized, controlled trials of interventions for major depressive disorder, which can make treatment arms more comparable and boost the efficacy signal.