The effect of perturbation-based balance training on balance control and fear of falling

Key points

1. 013 sessions of high-tech treadmill and platform perturbation training produced no significant additional benefit over usual care
2. 02Mini-BESTest scores improved slightly in both groups but changes did not reach the minimal detectable change threshold of 3-4 points
3. 03Fear of falling scores did not change in either group over the 4-week period
4. 04Training adherence was high at 93.7%, with only one minor adverse event reported
5. 05Baseline balance scores were relatively high, which may have limited the ability to detect improvement

How it was conducted

Design

Single-blind randomised controlled trial

Participants

82 community-dwelling adults aged 65 or older with a fall in the previous 3-6 months; median age 73 years (IQR 8 years)

Groups

PBT plus usual care (n=39) vs usual care alone (n=43)

Intervention

3 x 30-minute sessions over 3 weeks using the CAREN system (dual-belt treadmill, motion platform with 6 degrees of freedom, 180-degree VR screen); perturbations during standing and walking

Primary outcomes

Balance control (Mini-BESTest, 0-28 points) and fear of falling (FES-I, 16-64 points), assessed at baseline and 1 week post-intervention

Analysis

Intention-to-treat with Mann-Whitney U tests; multiple regression for confounders

What they found

• Mini-BESTest change scores were not significantly different between groups (p = 0.87); median change was 1 point (IQR 3) in both the control and PBT groups
• FES-I change scores were not significantly different between groups (p = 0.85); median change was 0 points (IQR 3) in both groups
• Baseline Mini-BESTest scores were 23 (IQR 4) in both groups; post-intervention scores were 24 (IQR 4) in the control group and 25 (IQR 5) in the PBT group
• Baseline FES-I scores were 20 (IQR 8) in the control group and 20 (IQR 7) in the PBT group; post-intervention scores were 20 (IQR 7) and 19 (IQR 7), respectively
• Proportion at increased fall risk based on Mini-BESTest dropped from 50.0% to 26.1% in the control group and from 56.8% to 40.5% in the PBT group, but the between-group difference was not significant (p = 0.23)
• Training adherence was 93.7% (104 of 111 scheduled sessions); 83.8% of participants attended all 3 sessions
• One training-related adverse event was reported: transient knee pain resolving within 2 days

Limitations

• COVID-19 restrictions led to protocol adjustments including broadening the fall history inclusion window from 3 to 6 months and conducting some follow-up assessments by phone
• Participants and treating therapists in the PBT group were not blinded to group allocation, introducing potential performance bias
• No direct instrumented measures of reactive balance control during perturbations were collected, so specific training-induced changes in reactive balance cannot be quantified
• Baseline Mini-BESTest scores were relatively high (median 23/28), suggesting a possible ceiling effect that may have limited detectable improvement on this clinical measure

Why it matters

For patients

Older adults who have fallen recently should not expect a short 3-session perturbation training programme to noticeably improve their balance or confidence compared to a physiotherapy referral alone.

For clinicians

PBT delivered in 3 sessions using advanced technology did not outperform usual care on clinical balance or fear-of-falling measures in community-dwelling fallers with relatively preserved baseline balance; more research is needed on optimal dose and outcome selection.

For readers

This well-conducted RCT challenges the assumption that a brief high-tech PBT programme adds meaningfully to standard physiotherapy for community-dwelling older adult fallers, highlighting gaps in training dose guidance and outcome measure sensitivity.