Figure 3From: Alternative analyses for handling incomplete follow-up in the intention-to-treat analysis: the randomized controlled trial of balloon kyphoplasty versus non-surgical care for vertebral compression fracture (FREE) Mean squared error for all follow-up visits for six outcomes, mean of the estimates from all 4 methods used as reference value. Five outcomes assessed were: short form-36 physical component summary, EuroQol 5-dimension, Roland-Morris Disability, visual analogue scale and restricted activity. CC, complete case; LOCF, last observation carried forward; MI, multiple imputation; MM, mixed-effects model; SE, standard error.Back to article page