Making inferences on treatment effects from real world data: propensity scores, confounding by indication, and other perils for the unwary in observational research
- 11 November 2013
- Vol. 347 (nov11 3), f6409
- https://doi.org/10.1136/bmj.f6409
Abstract
Propensity score based methods are used increasingly to evaluate the effectiveness of treatments when evidence from randomised trials is not available. However, users need to be aware of their strengths and limitationsKeywords
This publication has 12 references indexed in Scilit:
- Propensity scores: From naïve enthusiasm to intuitive understandingStatistical Methods in Medical Research, 2011
- Eplerenone in Patients with Systolic Heart Failure and Mild SymptomsThe New England Journal of Medicine, 2011
- Effect of statins on a wide range of health outcomes: a cohort study validated by comparison with randomized trialsBritish Journal of Clinical Pharmacology, 2008
- In Defense of Pharmacoepidemiology — Embracing the Yin and Yang of Drug ResearchThe New England Journal of Medicine, 2007
- Mortality Associated With Aprotinin During 5 Years Following Coronary Artery Bypass Graft SurgeryJama-Journal Of The American Medical Association, 2007
- The Risk Associated with Aprotinin in Cardiac SurgeryThe New England Journal of Medicine, 2006
- Eplerenone, a Selective Aldosterone Blocker, in Patients with Left Ventricular Dysfunction after Myocardial InfarctionThe New England Journal of Medicine, 2003
- A Comparison of Rate Control and Rhythm Control in Patients with Atrial FibrillationThe New England Journal of Medicine, 2002
- The Effect of Spironolactone on Morbidity and Mortality in Patients with Severe Heart FailureThe New England Journal of Medicine, 1999
- The central role of the propensity score in observational studies for causal effectsBiometrika, 1983