JSM 2013 Home
Online Program Home
My Program

Abstract Details

Activity Number: 360
Type: Contributed
Date/Time: Tuesday, August 6, 2013 : 10:30 AM to 12:20 PM
Sponsor: Section on Statistics in Epidemiology
Abstract - #309726
Title: Evaluation of Propensity Score Methods for Multiple Treatment Groups
Author(s): Lucia Mirea*+ and Junmin Yang and Prakesh Shah and Shoo Lee
Companies: Maternal-Infant Care Research Centre and Maternal-Infant Care Research Centre and Maternal-Infant Care Research Centre and Maternal-Infant Care Research Centre
Keywords: propensity score ; treatment selection bias ; multiple treatments ; matching ; stratification ; weighting
Abstract:

Propensity score (PS) methods are frequently applied to correct treatment selection bias in observational studies comparing two treatments. For multiple treatments, standard PS-matching, PS-stratification and PS-adjustment methods can compare treatments pair-wise, or PS scores estimated from multinomial logistic regression can be used in a PS-weighted approach to compare multiple treatments simultaneously. Motivated by a study of cardiac treatments in neonates, we evaluated the performance of PS methods for multiple treatments using simulations. Data were generated for a binary confounder associated with a 3-level treatment (OR= 2, 5), and a binary outcome (OR=5). Association between treatment and outcome was simulated under the null (OR=1) and alternative (OR=2, 5) hypotheses using 1000 replicate data sets of 5000 subjects. Bias, MSE and 5% type I error were estimated for PS-weighted, PS-matching, PS-stratification and PS-adjustment. Among these methods, PS-adjustment had nominal 5% type I error under the null, with negligible bias (< 0.002) and MSE (< 0.008) in the presence of treatment effects. Additional simulations will provide recommendations over a broad range of scenarios.


Authors who are presenting talks have a * after their name.

Back to the full JSM 2013 program




2013 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.

The views expressed here are those of the individual authors and not necessarily those of the JSM sponsors, their officers, or their staff.

ASA Meetings Department  •  732 North Washington Street, Alexandria, VA 22314  •  (703) 684-1221  •  meetings@amstat.org
Copyright © American Statistical Association.