Ansai 2015.

Methods	Study design: RCT Number of study arms: 3 Length of follow‐up: 4 months
Participants	Setting: São Paulo, Brazil Number of participants: 69 Number analysed: 68 Number lost to follow‐up: 1 Sample: community‐dwelling Age (years): mean 82.4 (SD 2.4) Sex: 68% female Inclusion criteria: aged > 80, community‐dwelling, sedentary, able to walk alone, available to attend training site 3 a week Exclusion criteria: presence of any injury listed in the absolute contraindications of the Physical Activity Readiness Medical Examination, relative cognition, neurological or musculoskeletal contraindications making participation in protocols impossible, MMSE score below the cut‐off designated by educational level minus 1 SD
Interventions	1. Group‐based balance, strength and aerobic training: cycle ergometer used for aerobic training, strength exercises (upper limbs, abdominals, squats, ankles) progressed using Borg scale and increments of 1 kg, balance activities with increasing difficulty; 1 hour, 3 a week for 16 weeks 2. Group‐based progressive strength training: leg press, chest press, calf raise, back extension, abdominal and rowing, 3 sets of 10 ‐ 12 RM using gym equipment; 1 hour, 3 a week, 16 weeks 3. Control: no intervention
Outcomes	1. Rate of falls 2. Number of people who experienced 1 or more falls (risk of falling)
Duration of the study	23 weeks
Adherence	1. Group‐based balance, strength and aerobic training group: 35% performed ≥ 24 sessions for 16 weeks (50% intervention) 2. Group‐based progressive strength training group: 56% performed ≥ 24 sessions for 16 weeks (50% intervention)
Notes	Source of funding: Federal University of São Carlos Economic information: not reported 16‐week data used due to proportion of fallers not being clear for longer follow‐up periods Email communication regarding fall data, response received, data not included in review
Risk of bias
Bias	Authors' judgement	Support for judgement
Random sequence generation (selection bias)	Low risk	Computerised random‐number generator
Allocation concealment (selection bias)	Low risk	Opaque, sealed envelope
Blinding of participants and personnel (performance bias) All outcomes	Unclear risk	Participants and personnel unblinded but impact of unblinding unknown
Blinding of outcome assessment (detection bias) Falls	High risk	Blinding of assessor not specified; as falls were reported by telephone or during training, assume assessor not blinded to group allocation
Blinding of outcome assessment (detection bias) Fractures	Unclear risk	Not applicable
Blinding of outcome assessment (detection bias) Hospital admission, medical attention and adverse events	Unclear risk	Not applicable
Blinding of outcome assessment (detection bias) Health related quality of life (self report)	Unclear risk	Not applicable
Incomplete outcome data (attrition bias) Falls and fallers	Low risk	Less than 20% of outcome data are missing (6%) and losses are balanced across groups
Selective reporting (reporting bias)	High risk	Falls were measured, but number of falls and adverse events were not reported
Method of ascertaining falls (recall bias)	Unclear risk	Provided with fall calendar, falls reported by retrospective recall once a month, by telephone or during training