Fig. 6From: Multiple imputation for patient reported outcome measures in randomised controlled trials: advantages and disadvantages of imputing at the item, subscale or composite score levelRMSE in the OKS composite score estimates (unit-nonresponse). Abbreviations: MAR – Missing at random; MI – Multiple imputation; OKS – Oxford knee score; PMM – Predicted mean matching; RMSE – Root mean square errorBack to article page