Medicine and Health Sciences Commons^™

Adjusting For Confounding By Neighborhood Using A Proportional Odds Model And Complex Survey Data, Babette A. Brumback, Amy B. Dailey, Hao W. Zheng

Health Sciences Faculty Publications

In social epidemiology, an individual's neighborhood is considered to be an important determinant of health behaviors, mediators, and outcomes. Consequently, when investigating health disparities, researchers may wish to adjust for confounding by unmeasured neighborhood factors, such as local availability of health facilities or cultural predispositions. With a simple random sample and a binary outcome, a conditional logistic regression analysis that treats individuals within a neighborhood as a matched set is a natural method to use. The authors present a generalization of this method for ordinal outcomes and complex sampling designs. The method is based on a proportional odds model and …