Reactive balance training — deliberately perturbing someone so they practise recovering — has promising trial data but usually needs a treadmill and a harness. This assessor-blinded randomised trial tested a low-cost version, ReacStep, using tether-release reactive stepping and intentional slips once weekly for six weeks in 88 older people, mean age 72. Both intervention and control groups also did eight weeks of home-based strength training. Blinded staff assessed laboratory-induced trips and slips, and weekly SMS surveys tracked real-world falls for 12 months.
Adherence was 90%, and the training was well tolerated. Maximum step length and choice stepping reaction time both improved significantly against controls (P < .05). Neither laboratory-assessed reactive balance falls nor daily-life falls over a year differed between groups (P = .19).
This is a clean, honest negative trial and worth reading as one. The intermediate measures moved — participants really could step further and faster — and it made no difference to falling, either in the laboratory paradigm the training resembled or in life. That gap is the recurring lesson of falls research: improving a component of balance is not the same as preventing a fall, which depends on hazard, medication, vision, blood pressure, footwear and cognition as much as on stepping. With 88 participants the trial cannot exclude a modest benefit, but it should stop anyone presenting reactive stepping drills as an established fall prevention intervention.
- Keep multifactorial falls assessment as the intervention with evidence — medications, vision, postural blood pressure, home hazards.
- Strength and balance exercise remains worthwhile; this trial does not argue against it.
- Do not present reactive stepping drills as proven fall prevention.
- Improved step length is a component measure, not an outcome — say so when reporting progress.
- Weekly SMS fall reporting is a cheap and effective method worth copying for your own audit.
The statistics, in plain English
With 88 participants and both groups also doing strength training, this trial could only detect a large difference in falls, so P = .19 means undetected rather than absent. Surrogate outcomes moving while the real outcome does not is the commonest way promising interventions fail, and is a reason to be sceptical of any programme evaluated only on balance measures. Weekly prospective fall reporting is more reliable than recall at a follow-up visit, which strengthens the null.
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.
Scan to keep reading on your phone. No account needed to start.
Tomorrow morning, before your first patient
One edition a day for geriatrics, written by the desk, every claim tied to its paper. Six minutes.
Get the app — free