Figure 2.

Effect of vitamin D supplementation on risk of falling in the past month for all participants and within participant subgroups: analysis of annual survey data using a repeated measures approach to incorporate responses from all five annual surveys. ORs compare vitamin D to placebo and were estimated using logistic regression with adjustment for randomization stratification variables of age, sex and state of residence at baseline. Each annual survey asked participants to report whether or not they had experienced a fall during the previous month. The numbers under ‘no. people’ show the number of participants who answered the falls question on at least one annual survey. Generalized estimating equations with an exchangeable correlation matrix used to account for participant clustering. P‐value from Wald test of significance of interaction term. Abbreviation: CI, confidence interval.