JSM 2012 Home

JSM 2012 Online Program

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: 134
Type: Contributed
Date/Time: Monday, July 30, 2012 : 8:30 AM to 10:20 AM
Sponsor: Biopharmaceutical Section
Abstract - #305696
Title: Comparing the Performance of Disease Risk Score Analysis, Propensity Score Analysis, and Multivariate Logistic Regression in a Simulated Data Set for Estimating Odds Ratios (OR)
Author(s): In-Lu Amy Liu*+ and Jiaxiao M Shi and Wansu Chen
Companies: Kaiser Permanente and Kaiser Permanente and Kaiser Permanente
Address: 1272 Wesley Ave, Pasedena, CA, 91104, United States
Keywords: disease risk score ; propensity score ; multivariate logistic regression ; odds ratio

Disease risk score (DRS) and propensity score (PS) are popular tools to balance covariates in observational studies. However, their strengths and weaknesses compared to the multivariate logistic regression (MLR) remain unknown. We conducted simulation to evaluate the performance of DRS, two DRS modifications, the probability and logit-based PS and MLR. When the association between the outcome and exposure is null (OR=1) or when the outcome is extremely rare (1%), the logit-based PS performs the best in terms of both bias and MSE, while the probability-based PS and MLR behave equally well. Bias and MSE were large for all versions of DRS. When the association between the outcome and exposure is not null (OR=2 or 3) and the outcome is not extremely rare (5% or 10%), MLR had the smallest bias as long as the models included all the outcome related covariates. However, one of the modified DRS slightly outperformed the MLR in some scenarios. PS methods performed poorly when OR is not null and outcome is not extremely rare due to the fact that they estimate marginal OR instead of conditional OR. Users need to be aware of the type of OR to be estimated before choosing an appropriate method.

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.