Abstract WP179: Global Risk Reduction: A Novel Method To Estimate The Impact Of Multiple Intervention Trials To Reduce Stroke Disparities

Abstract

Introduction: Reduction of recurrent stroke risk requires a comprehensive suite of proven interventions but most community-based trials are insufficiently powered to compare observed rates of “hard” outcomes such as time to recurrent stroke. Demand exists for innovative approaches to estimate the cumulative impact of interventions addressing multiple risk factors. Hypothesis: estimation of the cumulative relative effect on recurrent CVD events of multi-component risk factor interventions is feasible. Methods: For each of four stroke risk factors we identified meta-analyses of RCTS and observational studies. We modeled global risk reduction assuming a multiplicative relationship for the relative risks for each risk factor. We then used data on observed changes in systolic blood pressure (SBP) and LDL cholesterol and on medication adherence from the SUSTAIN trial (a multi-ethnic randomized community trial) to model estimated relative reduction in risk for recurrent stroke due to the SUSTAIN intervention Estimates of uncertainty capture two sources of variation in the outcome of estimated CVD risk: bootstrap methods capture sample-related uncertainty, and probabilistic sensitivity analysis capture uncertainty related to the benefit associated with each risk factor. Results: Among 272 SUSTAIN participants with complete 12-month data, most were non-white (73%) and Spanish speaking (62%). Changes in individual risk factors and medications, including SBP, smoking, aspirin and warfarin were similar in intervention and control groups (p>0.2) with the exception of LDL cholesterol, which was lowered by 11.4 mg/dL more in the intervention group (p=0.04). The estimated combined reductions in the relative risks of stroke were 0.42 and 0.37 in the intervention and control groups, respectively, with a non-significant relative risk difference of 5% (uncertainty interval -6 to 21). Conclusions: Data from community trials can be used to estimate the combined impact of multi-component interventions on risk of recurrent stroke and stroke disparities. We propose this global risk approach as an a priori secondary outcome for SUCCEED, an ongoing community-based trial.