TY - JOUR
T1 - A GEE approach to determine sample size for pre- and post-intervention experiments with dropout
AU - Zhang, Song
AU - Cao, Jing
AU - Ahn, Chul
N1 - Funding Information:
The work was supported in part by NIH grants UL1TR000451 and P30CA142543 , and a CPRIT grant RP110562-C1 .
PY - 2014
Y1 - 2014
N2 - Pre- and post-intervention experiments are widely used in medical and social behavioral studies, where each subject is supposed to contribute a pair of observations. In this paper we investigate sample size requirement for a scenario frequently encountered by practitioners: all enrolled subjects participate in the pre-intervention phase of study, but some of them will drop out due to various reasons, thus resulting in missing values in the post-intervention measurements. Traditional sample size calculation based on McNemar's test could not accommodate missing data. Through the GEE approach, we derive a closed-form sample size formula that properly accounts for the impact of partial observations. We demonstrate that when there are no missing data, the proposed sample size estimate under the GEE approach is very close to that under McNemar's test. When there are missing data, the proposed method can lead to substantial saving in sample size. Simulation studies and an example are presented.
AB - Pre- and post-intervention experiments are widely used in medical and social behavioral studies, where each subject is supposed to contribute a pair of observations. In this paper we investigate sample size requirement for a scenario frequently encountered by practitioners: all enrolled subjects participate in the pre-intervention phase of study, but some of them will drop out due to various reasons, thus resulting in missing values in the post-intervention measurements. Traditional sample size calculation based on McNemar's test could not accommodate missing data. Through the GEE approach, we derive a closed-form sample size formula that properly accounts for the impact of partial observations. We demonstrate that when there are no missing data, the proposed sample size estimate under the GEE approach is very close to that under McNemar's test. When there are missing data, the proposed method can lead to substantial saving in sample size. Simulation studies and an example are presented.
KW - Dropout
KW - McNemar's test
KW - Pre-post intervention
KW - Sample size
UR - http://www.scopus.com/inward/record.url?scp=84883174200&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=84883174200&partnerID=8YFLogxK
U2 - 10.1016/j.csda.2013.07.037
DO - 10.1016/j.csda.2013.07.037
M3 - Article
C2 - 24293779
AN - SCOPUS:84883174200
SN - 0167-9473
VL - 69
SP - 114
EP - 121
JO - Computational Statistics and Data Analysis
JF - Computational Statistics and Data Analysis
ER -