Skip to main content

Table 2 Illustration of different methods by example of data from Diamond JA et al. [4]

From: Problems in dealing with missing data and informative censoring in clinical trials

   Responders+
  Number of Patients Observed Estimated
(a)&
Estimated
(b)&
Estimated
(c)&
Early withdrawal 24(d) ? 0 24 19.2
Completed 6 months 30 24(cr) 24(cr) 24(cr) 24(cr)
Total (%) 54(100) 24(cr)+? (?) 24 (44.4) 48 (88.9) 43.2 (80)
  1. +Responder: effective control of BP and no side effects at month 6. & (a) and (b) assume two extreme informative missing models, (c) assumes missing completely at random (MCAR) model.