Propensity Score Methods for Bias Reduction in Observational Studies of Treatment Effect.

Authors:
Sindhu R Johnson
Sindhu R Johnson
University of Toronto
Canada
George A Tomlinson
George A Tomlinson
University of Toronto
Toronto | Canada
Gillian A Hawker
Gillian A Hawker
University of Toronto
Canada
John T Granton
John T Granton
University of Toronto
Toronto | Canada
Brian M Feldman
Brian M Feldman
The Hospital for Sick Children
Toronto | Canada

Rheum Dis Clin North Am 2018 05;44(2):203-213

Institute of Health Policy, Management and Evaluation, University of Toronto, 155 College Street, Toronto, Ontario M5T 3M6, Canada; Dalla Lana School of Public Health, University of Toronto, 155 College Street, Toronto, Ontario M5T 3M7, Canada; Division of Rheumatology, Department of Paediatrics, The Hospital for Sick Children, 555 University Avenue, Toronto, Ontario M5G 1X8, Canada.

A challenge to the use of observational data to study treatment effects is the issue of confounding. Noncomparability of exposed and nonexposed subjects can lead to biased estimation of the treatment effect. The propensity score is a balancing score that can be used to form matched groups, or pairs, that are not systematically different and enable nonbiased comparisons between groups. This article reviews propensity score methods with an illustrative example of the application of propensity score matching in an observational study of an uncommon disease (systemic sclerosis).

Download full-text PDF

Source
http://dx.doi.org/10.1016/j.rdc.2018.01.002DOI Listing
May 2018
1 Read

Publication Analysis

Top Keywords

propensity score
16
score methods
8
matched groups
4
form matched
4
score form
4
groups pairs
4
systematically enable
4
enable nonbiased
4
balancing score
4
pairs systematically
4
propensity
4
lead biased
4
subjects lead
4
biased estimation
4
estimation treatment
4
nonbiased comparisons
4
treatment propensity
4
score balancing
4
groups article
4
observational study
4

Similar Publications

A comparison of the ability of different propensity score models to balance measured variables between treated and untreated subjects: a Monte Carlo study.

Stat Med 2007 Feb;26(4):734-53

Institute for Clinical Evaluative Sciences, Toronto, Ont., Canada.

The propensity score--the probability of exposure to a specific treatment conditional on observed variables--is increasingly being used in observational studies. Creating strata in which subjects are matched on the propensity score allows one to balance measured variables between treated and untreated subjects. There is an ongoing controversy in the literature as to which variables to include in the propensity score model. Read More

View Article
February 2007

The performance of inverse probability of treatment weighting and full matching on the propensity score in the presence of model misspecification when estimating the effect of treatment on survival outcomes.

Stat Methods Med Res 2017 Aug 30;26(4):1654-1670. Epub 2015 Apr 30.

4 Department of Biostatistics, Johns Hopkins Bloomberg School of Public Health, Baltimore, Maryland.

There is increasing interest in estimating the causal effects of treatments using observational data. Propensity-score matching methods are frequently used to adjust for differences in observed characteristics between treated and control individuals in observational studies. Survival or time-to-event outcomes occur frequently in the medical literature, but the use of propensity score methods in survival analysis has not been thoroughly investigated. Read More

View Article
August 2017

Estimating effects of nursing intervention via propensity score analysis.

Nurs Res 2008 Nov-Dec;57(6):444-52

Division of Biostatistics, Mayo Clinic, Rochester, Minnesota 55905, USA.

Background: Lack of randomization of nursing intervention in outcome effectiveness studies may lead to imbalanced covariates. Consequently, estimation of nursing intervention effect can be biased as in other observational studies. Propensity score analysis is an effective statistical method to reduce such bias and further derive causal effects in observational studies. Read More

View Article
December 2008

Type I error rates, coverage of confidence intervals, and variance estimation in propensity-score matched analyses.

Authors:
Peter C Austin

Int J Biostat 2009 Apr 14;5(1):Article 13. Epub 2009 Apr 14.

Institute for Clinical Evaluative Sciences, Canada.

Propensity-score matching is frequently used in the medical literature to reduce or eliminate the effect of treatment selection bias when estimating the effect of treatments or exposures on outcomes using observational data. In propensity-score matching, pairs of treated and untreated subjects with similar propensity scores are formed. Recent systematic reviews of the use of propensity-score matching found that the large majority of researchers ignore the matched nature of the propensity-score matched sample when estimating the statistical significance of the treatment effect. Read More

View Article
April 2009