Kyrdalen 2014.

Methods	Study design: RCT Number of study arms: 2 Length of follow‐up: 3 months
Participants	Setting: 11 communities in southeast Norway Number of participants: 125 Number analysed: 94 Number lost to follow‐up: 31 Sample: community‐dwelling Age (years): mean 82.5 (SD 5.7) Sex: 73% female Inclusion criteria: home‐dwelling, at increased fall risk (defined as answering yes on either criterion 1 or 2 below, and in addition yes on 2 or more of criteria 3 ‐ 9: 1) had fallen at least once during the previous 12 months; 2) had self‐reported balance or gait problems; 3) had Parkinson’s disease or had suffered a stroke; 4) had 4+ concomitant diseases; 5) needed a handrail or support while rising from a chair; 6) used 4+ prescribed medications; 7) had reduced cognitive function as assessed by a geriatrician; 8) had BMI < 20, and 9) had reduced vision for their age Exclusion criteria: a score of 23/30 or less on the MMSE or not able to walk without support from another person
Interventions	1. Group‐based Otago Exercise Programme: 45 minutes 2 a week for 12 weeks plus outdoor walking for 30 minutes, ≥ 3 a week for 12 weeks 2. Individual Otago Exercise Programme: 30 minutes, 3 a week for 12 weeks, plus outdoor walking for 30 minutes, ≥ 3 a week for 12 weeks Both groups received 4 home visits to check programme plus 4 telephone calls
Outcomes	1. Number of people who experienced 1 or more falls (risk of falling) 2. Number of people who experienced 1 of more falls requiring hospital admission 3. Health‐related quality of life 4. Number of people who died
Duration of the study	12 weeks
Adherence	Adherence measured as session attendance 1. Group‐based Otago Exercise Programme: attended mean of 21.9 out of 24 sessions (SD 2.7) 2. Individual Otago Exercise Programme: attended mean 32.8 out of 36 recommended sessions (SD 2.8)
Notes	Source of funding: Norwegian Fund for Post‐Graduate Physiotherapy Training Economic information: not reported 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	Quote: "A Web‐based block randomization procedure with varying group size, developed by the Applied Clinical Research Unit at the Norwegian University of Science and Technology, was used"
Allocation concealment (selection bias)	Low risk	Centralised "web‐based" randomisation procedure
Blinding of participants and personnel (performance bias) All outcomes	High risk	Participants and personnel not blinded
Blinding of outcome assessment (detection bias) Falls	High risk	Baseline to 3 months: fall calendars collected by unblinded exercise instructors at intervention sessions. 3 ‐ 6 months: falls collected retrospectively at 6‐month interview with blinded assessor
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	Method of ascertaining hospital admission is unclear
Blinding of outcome assessment (detection bias) Health related quality of life (self report)	High risk	Participants not blinded to group allocaiton
Incomplete outcome data (attrition bias) Falls and fallers	High risk	More than 20% of fall outcome data are missing (25%)
Selective reporting (reporting bias)	High risk	Falls were measured, but number of falls was not reported
Method of ascertaining falls (recall bias)	High risk	Baseline to 3 months: falls were recorded on fall calendars which were collected by unblinded exercise instructors during twice‐weekly group sessions (intervention group) or at home visits in weeks 1, 2, 4 and 8 (control group). Non‐returns or incomplete calendars were followed up with the participant or next of kin; the person collecting this information unclear. 3 ‐ 6 months: falls collected retrospectively at 6‐month interview with blinded assessor