Comparing statistical methods for analyzing skewed longitudinal count data with many zeros: an example of smoking cessation.
H. Xie, J. Tao, G. McHugo, and R. Drake. Journal of substance abuse treatment, 45 (1):
99-108(July 2013)CI: Copyright (c) 2013; JID: 8500909; 2012/03/01 received; 2012/11/27 revised; 2013/01/22 accepted; 2013/02/28 aheadofprint; ppublish;.
DOI: 10.1016/j.jsat.2013.01.005
Abstract
Count data with skewness and many zeros are common in substance abuse and addiction research. Zero-adjusting models, especially zero-inflated models, have become increasingly popular in analyzing this type of data. This paper reviews and compares five mixed-effects Poisson family models commonly used to analyze count data with a high proportion of zeros by analyzing a longitudinal outcome: number of smoking quit attempts from the New Hampshire Dual Disorders Study. The findings of our study indicated that count data with many zeros do not necessarily require zero-inflated or other zero-adjusting models. For rare event counts or count data with small means, a simpler model such as the negative binomial model may provide a better fit.
Dartmouth Psychiatric Research Center, Department of Community and Family Medicine, Geisel School of Medicine at Dartmouth, Lebanon, NH 03766, USA. Haiyi.Xie@Dartmouth.Edu
%0 Journal Article
%1 Xie2013
%A Xie, Haiyi
%A Tao, Jill
%A McHugo, Gregory J
%A Drake, Robert E
%D 2013
%J Journal of substance abuse treatment
%K DataInterpretation Humans LongitudinalStudies Models NewHampshire PoissonDistribution ResearchDesign SmokingCessation SmokingCessation:statistics&numericaldata Statistical
%N 1
%P 99-108
%R 10.1016/j.jsat.2013.01.005
%T Comparing statistical methods for analyzing skewed longitudinal count data with many zeros: an example of smoking cessation.
%U http://www.ncbi.nlm.nih.gov/pubmed/23453482
%V 45
%X Count data with skewness and many zeros are common in substance abuse and addiction research. Zero-adjusting models, especially zero-inflated models, have become increasingly popular in analyzing this type of data. This paper reviews and compares five mixed-effects Poisson family models commonly used to analyze count data with a high proportion of zeros by analyzing a longitudinal outcome: number of smoking quit attempts from the New Hampshire Dual Disorders Study. The findings of our study indicated that count data with many zeros do not necessarily require zero-inflated or other zero-adjusting models. For rare event counts or count data with small means, a simpler model such as the negative binomial model may provide a better fit.
%@ 1873-6483; 0740-5472
@article{Xie2013,
abstract = {Count data with skewness and many zeros are common in substance abuse and addiction research. Zero-adjusting models, especially zero-inflated models, have become increasingly popular in analyzing this type of data. This paper reviews and compares five mixed-effects Poisson family models commonly used to analyze count data with a high proportion of zeros by analyzing a longitudinal outcome: number of smoking quit attempts from the New Hampshire Dual Disorders Study. The findings of our study indicated that count data with many zeros do not necessarily require zero-inflated or other zero-adjusting models. For rare event counts or count data with small means, a simpler model such as the negative binomial model may provide a better fit.},
added-at = {2023-02-03T11:44:35.000+0100},
author = {Xie, Haiyi and Tao, Jill and McHugo, Gregory J and Drake, Robert E},
biburl = {https://www.bibsonomy.org/bibtex/2c882c0bd782cea4c7c117b4cbe040ed4/jepcastel},
city = {Dartmouth Psychiatric Research Center, Department of Community and Family Medicine, Geisel School of Medicine at Dartmouth, Lebanon, NH 03766, USA. Haiyi.Xie@Dartmouth.Edu},
doi = {10.1016/j.jsat.2013.01.005},
interhash = {7eadc616b34320065f7ff01a2f788995},
intrahash = {c882c0bd782cea4c7c117b4cbe040ed4},
isbn = {1873-6483; 0740-5472},
issn = {1873-6483},
journal = {Journal of substance abuse treatment},
keywords = {DataInterpretation Humans LongitudinalStudies Models NewHampshire PoissonDistribution ResearchDesign SmokingCessation SmokingCessation:statistics&numericaldata Statistical},
month = {7},
note = {CI: Copyright (c) 2013; JID: 8500909; 2012/03/01 [received]; 2012/11/27 [revised]; 2013/01/22 [accepted]; 2013/02/28 [aheadofprint]; ppublish;},
number = 1,
pages = {99-108},
pmid = {23453482},
timestamp = {2023-02-03T11:44:35.000+0100},
title = {Comparing statistical methods for analyzing skewed longitudinal count data with many zeros: an example of smoking cessation.},
url = {http://www.ncbi.nlm.nih.gov/pubmed/23453482},
volume = 45,
year = 2013
}