PMID- 20936625
OWN - NLM
STAT- MEDLINE
DCOM- 20120112
LR  - 20120130
IS  - 1539-1612 (Electronic)
IS  - 1539-1604 (Linking)
VI  - 10
IP  - 3
DP  - 2011 May-Jun
TI  - The role of the minimum clinically important difference and its impact on 
      designing a trial.
PG  - 250-6
LID - 10.1002/pst.459 [doi]
AB  - The minimum clinically important difference (MCID) between treatments is 
      recognized as a key concept in the design and interpretation of results from a 
      clinical trial. Yet even assuming such a difference can be derived, it is not 
      necessarily clear how it should be used. In this paper, we consider three 
      possible roles for the MCID. They are: (1) using the MCID to determine the 
      required sample size so that the trial has a pre-specified statistical power to 
      conclude a significant treatment effect when the treatment effect is equal to the 
      MCID; (2) requiring with high probability, the observed treatment effect in a 
      trial, in addition to being statistically significant, to be at least as large as 
      the MCID; (3) demonstrating via hypothesis testing that the effect of the new 
      treatment is at least as large as the MCID. We will examine the implications of 
      the three different possible roles of the MCID on sample size, expectations of a 
      new treatment, and the chance for a successful trial. We also give our opinion on 
      how the MCID should generally be used in the design and interpretation of results 
      from a clinical trial.
CI  - Copyright (c) 2010 John Wiley & Sons, Ltd.
FAU - Chuang-Stein, Christy
AU  - Chuang-Stein C
AD  - Pfizer Inc, Kalamazoo, MI, USA.
FAU - Kirby, Simon
AU  - Kirby S
FAU - Hirsch, Ian
AU  - Hirsch I
FAU - Atkinson, Gary
AU  - Atkinson G
LA  - eng
PT  - Journal Article
DEP - 20101008
PL  - England
TA  - Pharm Stat
JT  - Pharmaceutical statistics
JID - 101201192
SB  - IM
MH  - Clinical Trials as Topic/*statistics & numerical data
MH  - Data Interpretation, Statistical
MH  - Humans
MH  - Models, Theoretical
MH  - Probability
MH  - *Research Design
MH  - Sample Size
MH  - *Treatment Outcome
EDAT- 2010/10/12 06:00
MHDA- 2012/01/13 06:00
CRDT- 2010/10/12 06:00
PHST- 2010/10/12 06:00 [entrez]
PHST- 2010/10/12 06:00 [pubmed]
PHST- 2012/01/13 06:00 [medline]
AID - 10.1002/pst.459 [doi]
PST - ppublish
SO  - Pharm Stat. 2011 May-Jun;10(3):250-6. doi: 10.1002/pst.459. Epub 2010 Oct 8.