Abstract 119: Assessing Modifications to Increase Power in a Stepped Wedge Designed Trial Targeting Improved Medication Adherence for Ischemic Heart Disease Patients

Background: Stepped wedge trial designs are often used to avoid simultaneous roll out of an intervention to all sites, yet time and monetary constrains may prevent all sites from receiving the intervention. In a multi-faceted trial using proactive real-time monitoring of patients to improve medication adherence for ischemic heart disease (IHD) patients, control data are passively collected through the standard electronic health record. Thus, study power may be augmented in a cost-efficient manner by modifying the traditional stepped wedge design to include sites acting strictly as controls, rather than by including more intervention sites.Methods: Based on previously collected data, we simulated models using a continuous adherence outcome and a traditional stepped wedge design rolling out an intervention to 4 sites over 4 waves (plus 1 baseline period). We varied the proportion of total variation attributed to site level variation in baseline outcomes (intraclass correlation or ICC). Power was estimated using theory and simulation with linear mixed models accounting for site level variation and time trend. Control sites were then added and the additional power generated was compared to a strategy in which the number of intervention sites was increased.Results: The addition of 3 or fewer control sites provides more power compared to the addition of a similar number of intervention sites. The effect of additional control sites plateaus and eventually is surpassed by additional intervention sites (see figure). Increasing the ICC decreases both power and the asymptote for the additional control sites; however, power with additional intervention sites continues to increase towards 100%.Conclusions: The use of sites acting strictly as controls in a trial targeting medication adherence for IHD patients provides more power at lower cost, especially with initial additional sites. Simulation provides results needed to achieve a balance between required power for the study, hypothesized site level variation, and resources required for additional control versus intervention sites to determine the number and type of sites to include. ![][1] [1]: /embed/graphic-1.gif