TY  - EJOUR
AB  - Randomized controlled trials provide essential evidence for the evaluation of new and existing medical treatments. Unfortunately, the statistical analysis is often complicated by the occurrence of protocol deviations, which mean we cannot always measure the intended outcomes for individuals who deviate, resulting in a missing-data problem. In such settings, however one approaches the analysis, an untestable assumption about the distribution of the unobserved data must be made. To understand how far the results depend on these assumptions, the primary analysis should be supplemented by a range of sensitivity analyses, which explore how the conclusions vary over a range of different credible assumptions for the missing data. In this article, we describe a new command, mimix, that can be used to perform reference-based sensitivity analyses for randomized controlled trials with longitudinal quantitative outcome data, using the approach proposed by Carpenter, Roger, and Kenward (2013, Journal of Biopharmaceutical Statistics 23: 1352–1371). Under this approach, we make qualitative assumptions about how individuals’ missing outcomes relate to those observed in relevant groups in the trial, based on plausible clinical scenarios. Statistical analysis then proceeds using the method of multiple imputation.
AU  - Cro, Suzie
AU  - Morris, Tim P.
AU  - Kenward, Michael G.
AU  - Carpenter, James R.
DA  - 2016-04
DA  - 2016-04
DO  - 10.22004/ag.econ.340129
DO  - doi
EP  - 463
EP  - 443
ID  - 340129
IS  - 2
JF  - Stata Journal
KW  - st0440
KW  - mimix
KW  - clinical trial
KW  - protocol deviation
KW  - missing data
KW  - multiple imputation
KW  - sensitivity analysis
L1  - https://ageconsearch.umn.edu/record/340129/files/Cro.pdf
L2  - https://ageconsearch.umn.edu/record/340129/files/Cro.pdf
L4  - https://ageconsearch.umn.edu/record/340129/files/Cro.pdf
LA  - eng
LA  - English
LK  - https://ageconsearch.umn.edu/record/340129/files/Cro.pdf
N2  - Randomized controlled trials provide essential evidence for the evaluation of new and existing medical treatments. Unfortunately, the statistical analysis is often complicated by the occurrence of protocol deviations, which mean we cannot always measure the intended outcomes for individuals who deviate, resulting in a missing-data problem. In such settings, however one approaches the analysis, an untestable assumption about the distribution of the unobserved data must be made. To understand how far the results depend on these assumptions, the primary analysis should be supplemented by a range of sensitivity analyses, which explore how the conclusions vary over a range of different credible assumptions for the missing data. In this article, we describe a new command, mimix, that can be used to perform reference-based sensitivity analyses for randomized controlled trials with longitudinal quantitative outcome data, using the approach proposed by Carpenter, Roger, and Kenward (2013, Journal of Biopharmaceutical Statistics 23: 1352–1371). Under this approach, we make qualitative assumptions about how individuals’ missing outcomes relate to those observed in relevant groups in the trial, based on plausible clinical scenarios. Statistical analysis then proceeds using the method of multiple imputation.
PY  - 2016-04
PY  - 2016-04
SN  - 1536-8634
SP  - 443
T1  - Reference-based sensitivity analysis via multiple imputation for longitudinal trials with protocol deviation
TI  - Reference-based sensitivity analysis via multiple imputation for longitudinal trials with protocol deviation
UR  - https://ageconsearch.umn.edu/record/340129/files/Cro.pdf
VL  - 16
Y1  - 2016-04
T2  - Stata Journal
ER  -