This study investigates the effects of a 12-week Tai Chi intervention on balance function in community-dwelling older adults. Participants are randomly assigned to either a Tai Chi exercise group or a health education control group. The primary outcome is balance function measured by the Berg Balance Scale. Secondary outcomes include physical function, psychological well-being, cognitive function, and sleep quality. The study also explores the predictive and modifying roles of baseline circadian age on intervention outcomes.

BACKGROUND AND RATIONALE Falls are a leading cause of injury and disability in older adults, and balance dysfunction is a key modifiable risk factor. Tai Chi, a traditional Chinese mind-body exercise, has been shown to improve balance, reduce fall risk, and benefit cognitive health in older populations. However, individual variability in response to Tai Chi remains poorly understood, and the biological mechanisms underlying its benefits are incompletely characterized.

Circadian rhythm disruption has emerged as a hallmark of biological aging. The investigators have developed and validated a circadian age (CircAge) assessment system based on the UK Biobank cohort (N>90,000), which independently predicts mortality risk. Whether individuals with accelerated circadian aging benefit more from Tai Chi intervention is unknown. This trial addresses this gap by integrating circadian age assessment with multi-modal outcome evaluations.

OBJECTIVES The primary objective is to evaluate the effect of a 12-week Tai Chi intervention on balance function in community-dwelling older adults. Secondary objectives are to assess effects on physical function, psychological health, cognitive function, and sleep quality. Exploratory objectives are to examine the predictive and modifying roles of baseline circadian age acceleration on intervention outcomes.

STUDY DESIGN This is a two-arm parallel, single-blind (assessor-blinded) randomized controlled trial. Ninety community-dwelling older adults are randomly allocated 1:1 to a Tai Chi group or a health education control group. Randomization is computer-generated using variable block sizes, stratified by sex and baseline Berg Balance Scale score (≤45 vs >45). Allocation concealment is maintained via sequentially numbered opaque sealed envelopes. Outcome assessors are blinded to group allocation. After the 12-week intervention, the control group is offered free Tai Chi classes to ensure ethical equipoise.

ASSESSMENT SCHEDULE Wrist-worn accelerometer (ActiGraph) recordings are collected at baseline over 7 consecutive days for circadian age computation. All baseline accelerometer recordings are completed prior to randomization. Full outcome assessments are conducted at baseline (week 0) and at the end of intervention (week 12). A streamlined midpoint assessment (week 6) includes only the Berg Balance Scale and Timed Up and Go test. Fasting blood samples for routine clinical chemistry are collected at week 0 and week 12.

STATISTICAL ANALYSIS A sample size of 90 (45 per group) provides >80% power to detect a standardized effect size of d=0.50 in Berg Balance Scale change at week 12, assuming α=0.05 (two-sided) and 20% attrition. The primary analysis follows the intention-to-treat principle using linear mixed-effects models adjusted for baseline score, age, sex, and circadian age acceleration, with class as a random effect for the Tai Chi group. Exploratory analyses include circadian age × group interaction terms.

LONG-TERM FOLLOW-UP After the 12-week intervention, participants are invited to enter a long-term health database with periodic follow-ups tracking balance function, fall events, and overall health status.

Inclusion Criteria:

• Aged 55 years or older
• Community-dwelling
• Able to walk independently (assistive devices such as canes are permitted)
• Berg Balance Scale (BBS) score ≤52 at screening
• Able to understand and follow verbal instructions
• Willing to wear a wrist accelerometer continuously for 7 days at baseline
• Willing to accept random group assignment
• Willing to undergo blood collection during the study
• Provides written informed consent

Exclusion Criteria:

• Regular Tai Chi or yoga practice (more than once per week) within the past 6 months
• Severe cognitive impairment (Montreal Cognitive Assessment [MoCA] score <20 at screening, or previously diagnosed dementia)
• Unstable cardiovascular disease (myocardial infarction, stroke, or hospitalization within the past 3 months)
• Severe musculoskeletal disease that limits exercise participation
• Neurological diseases affecting balance (e.g., Parkinson's disease, severe peripheral neuropathy)
• Currently participating in another exercise intervention study
• Planned surgery or extended absence during the study period
• Contraindications to functional near-infrared spectroscopy (fNIRS) examination (e.g., scalp injury, metallic implants in the head region)
• Severe psychiatric illness (e.g., schizophrenia, bipolar disorder)
• Considered unable to safely participate in moderate-intensity physical activity as judged by the study physician or attending healthcare provider

Outcome assessors are blinded to group allocation. Assessors are not involved in randomization or intervention delivery. Assessments are conducted in a separate location from the intervention sessions. All participants wear standardized clothing and footwear during assessment to prevent inadvertent unblinding. Participants are reminded not to disclose their group assignment to assessors.