The views expressed here are those of the individual authors and not necessarily those of the JSM sponsors, their officers, or their staff.
Online Program Home
Abstract Details
Activity Number:
|
507
|
Type:
|
Contributed
|
Date/Time:
|
Wednesday, August 1, 2012 : 10:30 AM to 12:20 PM
|
Sponsor:
|
Section on Statistical Consulting
|
Abstract - #304246 |
Title:
|
Zero-Inflated Mixed Regression to Predict Self-Reported Frequency of Problems with Female Condoms Use Among Women at High Risk of Sexually Transmitted Disease
|
Author(s):
|
Resmi Gupta*+ and Rhonda D Vandyke and Maurizio Macaluso
|
Companies:
|
Cincinnati Children's Hospital Medical Center and Cincinnati Children's Hospital Medical Center and Cincinnati Children's Hospital Medical Center
|
Address:
|
3333 Burnet Avenue, Cincinnati, OH, 45249, United States
|
Keywords:
|
zero-inflated Poisson mixed ;
repeated measure count data ;
generalized linear mixed models
|
Abstract:
|
Repeated measures data analysis involving zero-inflation often poses implementation and interpretability challenges to both statisticians and biomedical researchers. In addition to accounting for intra-subject correlation, the issue of excessive zero-counts must be addressed. Modeling count data with extra zeros with standard zero-inflated Poisson (ZIP) models may be inappropriate in such settings. A ZIP mixed regression model with random effects (ZIP-MR) allows specification of correlation and zero-count structures. We illustrate our approach using an epidemiologic study to evaluate self-reported frequency of problems with condom use in a group of women at high risk for contracting sexually transmitted disease who participated in follow up study of the impact of condom promotion intervention. A number of women reported no problems with use, resulting in zero-counts. ZIP-MR model performance was compared to standard ZIP using fit statistics (Akaike and Bayesian information criteria) and the likelihood ratio test (LRT); ZIP-MR model provided substantially better fit than the ZIP model based on model fit statistics.
|
The address information is for the authors that have a + after their name.
Authors who are presenting talks have a * after their name.
Back to the full JSM 2012 program
|
2012 JSM Online Program Home
For information, contact jsm@amstat.org or phone (888) 231-3473.
If you have questions about the Continuing Education program, please contact the Education Department.