Comparing treatment effects after adjustment with multivariable Cox proportional hazards regression and propensity score methods

• Published in: Pharmacoepidemiology and drug safety 2008-01, Vol.17 (1), p.1-8
• Main Authors: Martens, Edwin P., de Boer, Anthonius, Pestman, Wiebe R., Belitser, Svetlana V., Stricker, Bruno H. Ch, Klungel, Olaf H.
• Format: Article
• Language: English
• Subjects: Antihypertensive Agents - therapeutic use; confounding; Confounding Factors (Epidemiology); Cox proportional hazards models; Data Interpretation, Statistical; Humans; hypertension; Hypertension - drug therapy; Multivariate Analysis; Observation; observational studies; Observer Variation; propensity scores; Proportional Hazards Models; Prospective Studies; Qualitative Research; Risk Factors; Stroke - epidemiology; Stroke - prevention & control
• ISSN: 1053-8569; 1099-1557
• DOI: 10.1002/pds.1520

Purpose To compare adjusted effects of drug treatment for hypertension on the risk of stroke from propensity score (PS) methods with a multivariable Cox proportional hazards (Cox PH) regression in an observational study with censored data. Methods From two prospective population‐based cohort studies in The Netherlands a selection of subjects was used who either received drug treatment for hypertension (n = 1293) or were untreated ‘candidates’ for treatment (n = 954). A multivariable Cox PH was performed on the risk of stroke using eight covariates along with three PS methods. Results In multivariable Cox PH regression the adjusted hazard ratio (HR) for treatment was 0.64 (CI95%: 0.42, 0.98). After stratification on the PS the HR was 0.58 (CI95%: 0.38, 0.89). Matching on the PS yielded a HR of 0.49 (CI95%: 0.27, 0.88), whereas adjustment with a continuous PS gave similar results as Cox regression. When more covariates were added (not possible in multivariable Cox model) a similar reduction in HR was reached by all PS methods. The inclusion of a simulated balanced covariate gave largest changes in HR using the multivariable Cox model and matching on the PS. Conclusions In PS methods in general a larger number of confounders can be used. In this data set matching on the PS is sensitive to small changes in the model, probably because of the small number of events. Stratification, and covariate adjustment, were less sensitive to the inclusion of a non‐confounder than multivariable Cox PH regression. Attention should be paid to PS model building and balance checking. Copyright © 2007 John Wiley & Sons, Ltd.