| Methods |
RCT (individually randomised) |
| Participants |
Setting: 1 long‐term care facility (high‐level care), USA
N = 48 participants included in review (total of 124 in the study)
Sample: 73% women
Age (years): mean 89 (SD 6)
Inclusion criteria: life expectancy > 6 months; able to swallow medications; resident for > 3 months
Exclusion criteria: taking glucocorticoids; anti‐seizure medications; pharmacological doses of vitamin D; calcium metabolism disorders; severe mobility restriction; fracture within previous 6 months |
| Interventions |
200 IU of vitamin D2 daily for 5 months (not included in review) 400 IU of vitamin D2 daily for 5 months (not included in review) 600 IU of vitamin D2 daily for 5 months (not included in review) 800 IU of vitamin D2 daily for 5 months Control: placebo daily for 5 months |
| Outcomes |
Rate of falls Number of people falling |
| Duration of the study |
5 months |
| Notes |
Mean baseline serum vitamin D level for 800 IU group and control group combined was 53 nmol/L |
| Risk of bias |
| Bias |
Authors' judgement |
Support for judgement |
| Random sequence generation (selection bias) |
Low risk |
Quote: "... computer‐generated randomisation list." |
| Allocation concealment (selection bias) |
Low risk |
Pharmacy conducted randomisation and supplied medication in blister packs with name and patient identification number only |
| Blinding of participants and personnel (performance bias)
All outcomes |
Low risk |
Judgement comment: double‐blind |
| Blinding of outcome assessment (detection bias)
All outcomes |
Low risk |
Nursing staff completing incident forms blinded to treatment status because blister packs and tablets identical in appearance. Also, quote: "a programmer, not involved with this study and not aware of participant study group assignments, created the falls dataset linking the participant identification number with falls reported during the study period" |
| Incomplete outcome data (attrition bias)
All outcomes |
Low risk |
Judgement comment: low loss to follow‐up and ITT analysis performed. |
| Selective reporting (reporting bias) |
Low risk |
Judgement comment: no protocol identified, however all expected outcomes (falls, rate of falls and fallers) reported. |
| Method of ascertaining falls |
Low risk |
Judgement comment: falls concurrently recorded and clearly defined |
| Baseline imbalance |
Unclear risk |
Judgement comment: few differences at baseline; however baseline cognition, medical comorbidities and function not reported |
| Other bias |
Low risk |
None identified. |
