Adaptive Designs for Comparative Effectiveness Trials

• Activity Number: 618
• Type: Contributed
• Date/Time: Thursday, August 2, 2012 : 8:30 AM to 10:20 AM
• Sponsor: ENAR
• Author(s): John Kairalla*+ and Christopher S Coffey and Mitchell Thomann and Keith E. Muller
• Companies: University of Florida and University of Iowa and University of Iowa and University of Florida
• Keywords: Adaptive Designs ; Comparative Effectiveness ; Sample Size Re-estimation ; Power Analysis

Abstract

There is a clearly increasing level of national interest in comparative effectiveness research (CER). Due to the newness of the field and also confusion about what exactly CER is, reliable evidence based methods on the comparative effectiveness (CE) of treatments for medical and health policy decision makers are currently inadequate. RCTs must have a prominent place in CER due to their reliable information and well respected standard. However, in randomized CE trials, there may be limited information to guide initial design choices including the patient population, the primary outcome, or the target effect size. In the general RCT setting, adaptive designs have been proposed to address these concerns. There are potential advantages to expanding adaptive designs to within the CE context. Although there are many similarities between the two, CE trials have some fundamental differences from standard clinical trials. For one, the heterogeneity in the population studied in CE creates higher variability in outcomes. CE studies could be underpowered if they use planning values obtained from tightly controlled clinical trials. Additionally, the concept of a 'minimum clinically meaningful difference' is hard to define in the CE context. Even assuming equal cost and safety, a range of meaningful effect sizes could be defined with upper limit the largest effect with reasonable chance of being observed and lower limit the minimal effect deemed sizable enough to change practice in the study context. We first review the current state of clinical CER. Then, we identify areas of CER that seem particularly strong candidates for the development of novel adaptive design methodology and application. We describe the evaluative process to determine the usefulness of these designs in CER in a number of useful two group comparison situations. Illustrative analytic results are used to explore properties of various adaptive sample size re-estimation designs tailored for use in CE trials. We summarize results, make recommendations, and identify areas needing future research.