Kim 2014.

Methods	Study design: RCT Number of study arms: 2 Length of follow‐up: 12 months
Participants	Setting: Tokyo, Japan Number of participants: 105 Number analysed: 103 Number lost to follow‐up: 2 Sample: community‐dwelling women Age (years): Intervention mean 77.83 (SD 4.21), Control mean 77.83 (SD 4.15) Sex: 100% female Inclusion criteria: age ≥ 70 years; experienced at least 1 fall incident in the previous year; and no missing fall‐related baseline data Exclusion criteria: severe knee or back pain; severe walking disability; and unstable cardiac conditions
Interventions	1. Group‐based balance and strength: increased difficulty of exercises, used resistance bands or ankle weights for strength training; 60‐minute, 2 a week for 3 months; plus 1‐hour exercise classes 1 a month during 1‐year follow‐up; home programme encouraged ≥ 3 a week during 1‐year follow‐up 2. Control group: Health education. 60‐minute class once a month for 3 months, a total of 3 times
Outcomes	1. Number of people who experienced 1 or more falls (risk of falling) 2. Number of people who experienced 1 or more fall‐related fractures
Duration of the study	52 weeks
Adherence	Adherence measured as session attendance, frequency of home exercises, mean exercise time 1. Group‐based balance and strength group: mean attendance rate during intervention, 75% (range 64 – 86%); mean frequency home exercises 3.4 a week; mean exercise time 24.9 minutes
Notes	Source of funding: Ministry of Health and Welfare of Japan, Japan Society for the Promotion of Science Economic information: not reported
Risk of bias
Bias	Authors' judgement	Support for judgement
Random sequence generation (selection bias)	Low risk	Computer‐generated random numbers
Allocation concealment (selection bias)	Unclear risk	Quote: "The allocation process was blinded". Insufficient information to permit judgement.
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	Low risk	Fall diaries were collected at 1‐year follow‐up Quote: "The investigators evaluating the effects of the exercise treatment were blind to intervention allocations"
Blinding of outcome assessment (detection bias) Fractures	High risk	Participants were asked about fractures by face‐to‐face interview at baseline, 3 month and 1 year. No radiological confirmation
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 fall outcome data are missing (2%). Missing data were balanced between the exercise (n = 2) and control (n = 1) groups, with reasonable reasons for loss of data in the 2 groups (exercise: reduced motivation = 1, hospitalisation = 1; control: moved house = 1)
Selective reporting (reporting bias)	High risk	Falls were measured, but number of falls was not reported. Adverse events were not reported
Method of ascertaining falls (recall bias)	Unclear risk	Falls diary, distributed at 3‐month assessment and collected at 1‐year follow‐up