Effect of Tai Chi on the pain intensity or disability of patients with chronic low back pain: a systematic review and meta-analysis

Objectives Chronic low back pain (CLBP) is a leading cause of disability worldwide and remains challenging to manage despite numerous treatment options. Tai Chi (TC), a traditional mind–body exercise, has been increasingly used as a non-pharmacological approach for CLBP, but previous trials and reviews have reported inconsistent results, particularly regarding disability outcomes and the influence of different TC parameters. This systematic review and meta-analysis aimed to synthesize current evidence on the effects of TC on pain intensity, disability, and other health-related outcomes in adults with CLBP, and to explore the potential impact of TC styles and training characteristics. Methods Randomized controlled trials (RCTs) comparing TC with control interventions in adults with CLBP were searched in six English and Chinese databases from inception to May 2025. Primary outcomes were pain intensity and disability; secondary outcomes included physical function, general health, muscle function, and proprioception. Risk of bias was assessed using the Cochrane Risk of Bias tool, and the certainty of evidence was rated with the Grading of Recommendations Assessment, Development, and Evaluation (GRADE) approach. Where appropriate, meta-analyses and subgroup and sensitivity analyses were conducted. Publication bias was examined using funnel plots and Egger's regression test. Results Fourteen RCTs ( n = 960) met the inclusion criteria. Compared with control conditions, TC was associated with reduced pain intensity (SMD = −2.14; 95% CI: −2.84 to −1.44; P 0.00001; GRADE: moderate) and improvements in disability measures (SMD = −1.45; 95% CI: −2.49 to −0.40; P = 0.007; GRADE: very low) and nine disability-related subdomains (pain intensity, personal care, lifting, walking, standing, sitting, sleeping, social life, and traveling). Benefits were also observed for physical function, muscle function, and quality of life, although evidence for proprioception was inconsistent. Subgroup analyses suggested that Chen-style TC and higher training frequency (3 times/week) might be associated with larger pain reductions, but heterogeneity across studies remained substantial and the certainty of evidence ranged from moderate for pain intensity to very low for most disability outcomes. Conclusion TC may be a safe and potentially effective adjunctive intervention for individuals with CLBP, with clinically relevant reductions in pain and improvements in physical function for some patients. However, substantial heterogeneity, small sample sizes in several trials, and low certainty of evidence for disability outcomes warrant cautious interpretation. Further high-quality, adequately powered RCTs with standardized TC protocols are needed to confirm these findings and to clarify the optimal style and training dose of TC for CLBP management. Systematic Review Registration PROSPERO CRD420251072382.