Return toAll Abstracts
HSR&D 2004 National Meeting Abstracts

1018. Randomized Encouragement Trial: A Pragmatic Paradigm for Clinical Research
Joel T Braslow, MD, PhD, VISN 22 MIRECC and University of California Los Angeles, N Daun, University of California Los Angeles, JR Weisz, University of California Los Angeles, KB Wells, University of California Los Angeles and RAND, SL Starks, University of California Los Angeles

Objectives: Traditional research designs, such as observational studies and randomized controlled trials (RCTs), fall short in addressing practice-oriented questions critical to care-improvement initiatives like QUERI: How are effective interventions adopted across diverse settings and populations? Do these therapies work in naturalistic settings? How do system-wide interventions affect health outcomes? We describe how these questions are addressed by the randomized encouragement trial (RET), a rigorous yet naturalistic study design.

Methods: The RET is an experimental design that randomizes patients to experimental conditions but encourages, rather than mandates, treatment. We define encouragement as the offer of resources, incentives, or communications designed to increase participantsí propensities to adopt the target treatment. We illustrate the RET using the Partners in Care study (PIC), which evaluated encouragement strategies to improve use of guideline-concordant treatments for depression in forty-six non-academic managed care clinics.

Results: In PIC, the RET permitted experimental evaluation of system-wide encouragement interventions and quasi-experimental evaluation of treatments. Because it preserves many aspects of usual care, including patient and physician choice, it has high external validity. Internal validity is lower than that of RCTs, but markedly higher than that of observational studies. The RET does not mandate treatment adoption; thus, it can measure adoption under usual care and encouragement.

Conclusions: Due to its experimental yet non-invasive nature, the RET can evaluate treatment effectiveness, adoption rates, and system-wide interventions in naturalistic settings.

Impact: The RET may be useful to both clinical and services researchers in their efforts to improve treatment efficacy and effectiveness.