Tai chi vs CBT: Which is better for treating insomnia in adults?

A recent study has shown that tai chi is less effective than cognitive behavioural therapy for insomnia (CBT-I) at 3 months, but it is no longer inferior to CBT-I at month 15.

After the 3-month intervention, tai chi and CBT-I improved insomnia symptoms in middle-aged and older adults with chronic insomnia, although the improvements were more marked in the CBT-I group, and tai chi was inferior to CBT-I in treating insomnia,” the investigators said. “At the 12-month follow-up (month 15), tai chi was noninferior to CBT-I in improving insomnia symptoms.”

Furthermore, both tai chi and CBT-I showed comparable benefits on subjective sleep parameters, quality of life, mental health, and physical activity level.

Two hundred Chinese participants aged ≥50 years with chronic insomnia diagnosed according to the Diagnostic and Statistical Manual of Mental Disorders, fifth edition, were randomized to receive tai chi (n=100) or CBT-I (n=100) between 18 May 2020 and 14 July 2022 in Hong Kong.

The two interventions were delivered in group format over 3 months, consisting of 1-h sessions twice weekly for a total of 24 sessions. The change in perceived insomnia severity, measured by the Insomnia Severity Index, after the intervention (month 3) and at 12-month follow-up (month 15) was the primary outcome. A threshold of four points was used as the margin of noninferiority.

At month 3, tai chi resulted in a reduction of 6.67 (95 percent confidence interval [CI], 5.61–7.73) in Insomnia Severity Index scores, while CBT-I showed a much greater decrease at 11.19 (95 percent CI, 10.06–12.32), with a between-group difference of 4.52 (95 percent CI, −∞ to 5.81). Tai chi was inferior to CBT-I since the upper confidence limit surpassed the noninferiority margin. [BMJ 2025;391:e084320]

At month 15, however, the reductions for tai chi and CBT-I were comparable at 9.51 (95 percent CI, 8.47–10.54) and 10.18 (95 percent CI, 8.91–11.40), respectively, with a difference of 0.68 (95 percent CI, −∞ to 2.00) between groups.

“At this point, tai chi was considered noninferior to CBT-I because the upper confidence limit fell within the noninferiority margin,” the investigators said. “Results from the intention-to-treat analysis were consistent with the per-protocol findings.”

In addition, adverse events did not occur during the intervention period.

Pathways

A previous trial comparing the physiological response of tai chi and CBT-I on systemic, cellular, and genomic markers of inflammation in 90 patients with chronic insomnia and a history of breast cancer found that both interventions reduced toll-like receptor 4 stimulated monocyte production of interleukin 6 and tumour necrosis factor α and their coexpression. [Brain Behav Immun 2024;120:159-166]

Furthermore, tai chi and CBT-I led to reductions in antiviral gene expression and inflammatory gene transcripts, as well as increases in antiviral gene transcripts over 15 months (p<0.05 for all).

Notably, tai chi resulted in greater and more sustained reductions in systemic and cellular inflammation at 12-month postintervention follow-up, while CBT-I demonstrated more robust increases in antiviral gene transcripts (p<0.05 for all). 

“These findings suggest distinct physiological pathways by which each intervention modulates inflammatory processes and offers therapeutic benefits and aligns with the results of our study relating to the long-term benefits of tai chi in treating insomnia,” the investigators said.

“Our study supports tai chi as an alternative treatment approach for the long-term management of chronic insomnia in middle-aged and older adults,” they added.