Back to Annual Meeting Page		American Public Health Association 133rd Annual Meeting & Exposition December 10-14, 2005 Philadelphia, PA

Abstract #105713

Jichuan Wang, PhD, Russel S. Falck, MA, and Robert G. Carlson, PhD. Community Health Department, Wright State University, 3640 Colonel Glenn Hwy., Dayton, OH 45435, 937 775-2084, jichuan.wang@wright.edu

Nonparametric standardization and decomposition methods are often used for comparing rate differences between populations. Difference in the crude rate between populations under study can be decomposed into additive component effects, such as the “rate effect,” representing the “real” rate difference, and “composition effects,” representing the outcome difference attributed to compositional differences in the confounding factors between populations. The methods can be applied to a very wide range of other outcome measures, such as percentage, proportion, ratio, and arithmetic mean. When decomposition analysis is conducted, using sample rather than population data, it is desirable to take uncertainty into account. That is, the standard errors of the component effects need to be estimated for the purpose of significance testing. Unfortunately, none of the current standardization and decomposition methods take sampling variability into account. The authors of this study have developed a computer program, DECOMP, that provides an opportunity to estimate standard errors of component effects via bootstrapping so that significance testing for component effects becomes possible. The DECOMP enables researchers to conduct standardization and decomposition analysis simultaneously for multiple populations/samples, including longitudinal data sets. This study demonstrates how to use the DECOMP to compare sex risk behaviors among drug users in three different samples (i.e., urban crack users, rural stimulant users, and MDMA users). Differences in outcome measures between the samples are simultaneously decomposed into “real” differences and component effects of confounding factors.

Learning Objectives: At the conclusion of the presentation, participants (learner) will know

• fundamentals of standardization and decomposition methods.
• how to apply the methods to study outcome differences in the fields of public health research.
• how to conduct standardization and decomposition analysis with multiple samples simultaneously.
• how to use the DECOMP program.

Presenting author's disclosure statement:

I wish to disclose that I have NO financial interests or other relationship with the manufactures of commercial products, suppliers of commercial services or commercial supporters.