Skip to main content

Table 2 Estimates (95 % CI) from the complete case and standard joint modellinga analyses

From: Modelling variable dropout in randomised controlled trials with longitudinal outcomes: application to the MAGNETIC study

Model Longitudinal outcome Dropout
β 1 (95 % CI) HR (95 % CI) γ (95 % CI)
Complete case analysis (n = 402) −0.086 (−0.297 to 0.125)
Standard joint model (n = 502, overall dropouts = 100) −0.193 (−0.381 to −0.010) 1.869 (1.238 to 2.772) −0.211 (−0.442 to −0.017)
  1. a95 % bias-corrected percentile CIs are obtained from 1000 bootstrap resamples