Baseline characteristics of participants
| PBT group (n=70) | Control group (n=70) | |
| Age (years), mean (SD) | 72.7 (4.7) | 72.0 (4.7) |
| Sex, no. female (%) | 41 (59) | 38 (54) |
| Frailty,* median (IQR) | 2 (1–3) | 2 (1–3) |
| Living arrangement, no. living alone (%) | 23 (33) | 24 (34) |
| Daily activity level, sitting down† (hours) | 4 (6) | 6 (9) |
| Medication, median (IQR) | 2 (0–4) | 3 (0–4) |
| Function of daily activities,‡ median (IQR) | 2 (1–2) | 2 (1–2) |
| Fall within the past 12 months, no. fall (%) | 28 (40) | 29 (41) |
| Cognition,§ median (IQR) | 26 (24–28) | 26 (24–28) |
| Physical function,¶ median (IQR) | 12 (11–12) | 12 (11–12) |
| Habitual gait speed (m/s), mean (SD) | 1.3 (0.2) | 1.3 (0.2) |
| Executive function,** median (IQR) | 47.3 (26.1–63.5) | 40.7 (29.1–61.8) |
*Tilburg Frailty indicator (score 0–15; lower is better).
†International Physical Activity Questionnaire—time spent sitting down on average per day.
‡The Vulnerable Elderly-13 Survey (score 0–10; lower is better).
§The Short Orientation-Memory-Concentration Test (score 0–28; higher is better).
¶The Short Physical Performance Battery (0–12; higher is better).
**Trail Making Test part A subtracted from part B (lower score implied better performance).
PBT, perturbation-based balance training.