« Previous Next » Journal of Clinical Epidemiology
Volume 59, Issue 5 , Pages 437.e1-437.e24 , May 2006

A review of the application of propensity score methods yielded increasing use, advantages in specific settings, but not substantially different estimates compared with conventional multivariable methods

Til Stürmer
- Division of Pharmacoepidemiology and Pharmacoeconomics, Brigham and Women's Hospital, Harvard Medical School, 1620 Tremont Street, Suite 3030, Boston, MA 02120, USA
- Division of Preventive Medicine, Brigham and Women's Hospital, Harvard Medical School, Boston, MA, USA
- Corresponding author. Tel.: 617-278-0627; fax: 617-232-8602.
Manisha Joshi
- Division of Pharmacoepidemiology and Pharmacoeconomics, Brigham and Women's Hospital, Harvard Medical School, 1620 Tremont Street, Suite 3030, Boston, MA 02120, USA
Robert J. Glynn
- Division of Pharmacoepidemiology and Pharmacoeconomics, Brigham and Women's Hospital, Harvard Medical School, 1620 Tremont Street, Suite 3030, Boston, MA 02120, USA
- Division of Preventive Medicine, Brigham and Women's Hospital, Harvard Medical School, Boston, MA, USA
Jerry Avorn
- Division of Pharmacoepidemiology and Pharmacoeconomics, Brigham and Women's Hospital, Harvard Medical School, 1620 Tremont Street, Suite 3030, Boston, MA 02120, USA
Kenneth J. Rothman
- Division of Pharmacoepidemiology and Pharmacoeconomics, Brigham and Women's Hospital, Harvard Medical School, 1620 Tremont Street, Suite 3030, Boston, MA 02120, USA
- Department of Epidemiology, Boston University School of Public Health, Boston, MA, USA
Sebastian Schneeweiss
- Division of Pharmacoepidemiology and Pharmacoeconomics, Brigham and Women's Hospital, Harvard Medical School, 1620 Tremont Street, Suite 3030, Boston, MA 02120, USA

,Accepted 15 June 2005.

Nonoverlap of the propensity score distributions among exposed and unexposed subjects. In this example, subjects with low propensity scores are never exposed, and subjects with high propensity scores

Nonoverlap of the propensity score distributions among exposed and unexposed subjects. In this example, subjects with low propensity scores are never exposed, and subjects with high propensity scores are always exposed.
- Full-Size Image
- Download to PowerPoint
Number of medical research studies listed in PubMed and Science Citation Index using propensity score methods to control for confounding, according to year of publication.

Number of medical research studies listed in PubMed and Science Citation Index using propensity score methods to control for confounding, according to year of publication.
- Full-Size Image
- Download to PowerPoint