Guided Baduanjin in long-term care, with the feedback loop closed

Design

Single-blind three-arm randomised controlled trial

Population

117 long-term care residents in Chongqing, China (mean age 75.5, 50% female), 39 per arm

Primary outcome

Fried frailty phenotype at 12 weeks

Effect

Significant group (F=3.46, P=.035) and interaction (F=45.99, P<.01) effects; AR better than control at 12 weeks (P<.01)

One hundred and seventeen residents of long-term care facilities in Chongqing, mean age 75.5, were randomised to augmented-reality-guided Baduanjin, the same exercise delivered by video, or control. The augmented reality system tracked movement and gave feedback in real time.

Fried frailty phenotype, the primary outcome, improved with time, differed between groups and showed a strong interaction, with the augmented reality group ahead of controls at 12 weeks. Physical function, activities of daily living, nutrition, cognition and mood all favoured augmented reality over both video and control. Adherence was 91 per cent with the system and 89 per cent with video.

The interesting comparison is not technology against nothing — it is technology against the same exercise on a screen. Both arms did Baduanjin; only one was told, continuously, whether they were doing it correctly. That is a finding about supervision, and supervision is the expensive ingredient in every exercise programme for frail older people.

• Where group exercise exists in a facility, correcting form matters — the video arm did the same movements and gained less.
• Adherence near 90 per cent in institutionalised residents is achievable; low uptake is not inevitable.
• Baduanjin is low-impact and seated variants exist, which suits residents who cannot stand unsupported.
• Single-blind with self-reported secondary outcomes; expectation effects favour the novel arm.
• The equipment is the barrier to transferring this; the principle — corrective feedback — is not.

Why it matters

It separates the value of the exercise from the value of being watched doing it.

Don't overread it

Single-blind, single-city, 117 residents, with many unadjusted secondary outcomes.

The statistics, in plain English

Repeated-measures ANOVA with a significant group-by-time interaction (F=45.99) means the three arms diverged over the 12 weeks rather than simply differing at the end — the right analysis for this design. With 39 per arm, the secondary outcomes across nutrition, cognition and mood involve many comparisons and no correction is reported, so those should be read as consistent direction rather than as separate findings.

Read the rest in the app

You have read your two free briefings this month. The app carries all 27 specialties, every morning, free — and this finding is waiting in it.

Get Daily Doctor — free

Scan to keep reading on your phone. No account needed to start.