Citation: Archives of Physical Medicine & Rehabilitation. 93(8 Suppl):S127-37, 2012 Aug..Journal: Archives of physical medicine and rehabilitation.ISSN: 0003-9993.Full author list: Horn SD; DeJong G; Deutscher D.UI/PMID: 22840879.Subject(s): Comparative Effectiveness Research/mt [Methods] | *Evidence-Based Medicine/mt [Methods] | Humans | Outcome Assessment (Health Care) | *Randomized Controlled Trials as Topic/mt [Methods] | *Rehabilitation/mt [Methods] | Rehabilitation/st [Standards] | Reproducibility of Results | *Research DesignInstitution(s): MedStar National Rehabilitation NetworkActivity type: Journal Article.Medline article type(s): Journal ArticleOnline resources: Click here to access onlineDigital Object Identifier: http://dx.doi.org/10.1016/j.apmr.2011.10.031 (Click here)Abbreviated citation: Arch Phys Med Rehabil. 93(8 Suppl):S127-37, 2012 Aug.Local Holdings: Available online from MWHC library: 2000 - present, Available in print through MWHC library:1999-2007.Abstract: Sound rigorous methods are needed by researchers and providers to address practical questions about risks, benefits, and costs of interventions as they occur in routine clinical practice such as: Are treatments used in daily practice associated with intended outcomes? For whom does an intervention work best? With limited clinical resources, what are the best interventions to use for specific types of patients? Answers to such questions can help clinicians, patients, researchers, and health care administrators learn from, and improve, real-world everyday clinical practice. In this article, we describe existing research designs to demonstrate clinical usefulness and comparative effectiveness of rehabilitation treatments. We compare randomized controlled trials and observational cohort studies of various types, including those that use instrumental variables or propensity scores to control for potential patient or treatment selection effects. We argue that practice-based evidence (PBE) study designs include features that address limitations inherent in both randomized trials and traditional observational studies, and also reduce the need for instrumental variables and propensity scores methods. We give examples of how PBE designs have been used in various rehabilitation areas to determine better treatments for specific types of patients. Copyright 2012 American Congress of Rehabilitation Medicine. Published by Elsevier Inc. All rights reserved.