Methods for Constructing and Assessing Propensity Scores

• Published in: Health services research Vol. 49; no. 5; pp. 1701 - 1720
• Main Authors: Garrido, Melissa M., Kelley, Amy S., Paris, Julia, Roza, Katherine, Meier, Diane E., Morrison, R. Sean, Aldridge, Melissa D.
• Format: Journal Article
• Language: English
• Published: United States Blackwell Publishing Ltd 01-10-2014; Health Research and Educational Trust; BlackWell Publishing Ltd
• Subjects: administrative data uses; Adult; Aged; Aged, 80 and over; Analysis; Cancer; Cancer Care Facilities - statistics & numerical data; Confounding Factors (Epidemiology); Data Interpretation, Statistical; Female; Health aspects; Homeopathy; Humans; Male; Materia medica and therapeutics; Methods; Middle Aged; Models, Statistical; Neoplasms - epidemiology; New York; Observational data/quasi-experiments; Ohio; Outcome and process assessment (Health Care); Outcome and process assessment (Medical care); Outcome Assessment (Health Care) - methods; Palliative care; Palliative Care - statistics & numerical data; patient outcomes/function; Pennsylvania; Propensity Score; Propensity scores; Research Design; Studies; Therapeutics; Treatment Outcome; Virginia; Wisconsin; Young Adult; New York; Ohio; Virginia; Wisconsin; Pennsylvania; administrative data uses; Observational data/quasi-experiments; patient outcomes/function
• ISSN: 0017-9124; 1475-6773
• DOI: 10.1111/1475-6773.12182

Objectives
To model the steps involved in preparing for and carrying out propensity score analyses by providing step‐by‐step guidance and Stata code applied to an empirical dataset.
Study Design
Guidance, Stata code, and empirical examples are given to illustrate (1) the process of choosing variables to include in the propensity score; (2) balance of propensity score across treatment and comparison groups; (3) balance of covariates across treatment and comparison groups within blocks of the propensity score; (4) choice of matching and weighting strategies; (5) balance of covariates after matching or weighting the sample; and (6) interpretation of treatment effect estimates.
Empirical Application
We use data from the Palliative Care for Cancer Patients (PC4C) study, a multisite observational study of the effect of inpatient palliative care on patient health outcomes and health services use, to illustrate the development and use of a propensity score.
Conclusions
Propensity scores are one useful tool for accounting for observed differences between treated and comparison groups. Careful testing of propensity scores is required before using them to estimate treatment effects.