Discharge educational strategies for reduction of vascular events (DESERVE): design and methods

Abstract

Rationale: Stroke and vascular risk factors disproportionately affect minority populations, with Blacks and Hispanics experiencing a 2·5- and 2·0-fold greater risk compared with whites, respectively. Patients with transient ischemic attacks and mild, nondisabling strokes tend to have short hospital stays, rapid discharges, and inaccurate perceptions of vascular risk.

Aim: The primary aim of the Discharge Educational Strategies for Reduction of Vascular Events (DESERVE) trial is to evaluate the efficacy of a novel community health worker-based multilevel discharge intervention vs. standard discharge care on vascular risk reduction among racially/ethnically diverse transient ischemic attack/mild stroke patients at one-year postdischarge. We hypothesize that those randomized to the discharge intervention will have reduced modifiable vascular risk factors as determined by systolic blood pressure compared with those receiving usual care.

Sample size estimates: Given 300 subjects per group and alpha of 0·05, the power to detect a 6 mmHg reduction in systolic blood pressure is 89%.

Design: DESERVE trial is a prospective, randomized, multicenter clinical trial of a novel discharge behavioral intervention. Patients with transient ischemic attack/mild stroke are randomized during hospitalization or emergency room visit to intervention or usual care. Intervention begins prior to discharge and continues postdischarge.

Study outcomes: The primary outcome is difference in systolic blood pressure reduction between groups at 12 months. Secondary outcomes include between-group differences in change in glycated hemoglobin, smoking rates, medication adherence, and recurrent stroke/transient ischemic attack at 12 months.

Discussion: DESERVE will evaluate whether a novel discharge education strategy leads to improved risk factor control in a racially diverse population.

Keywords: TIA; discharge; health education; mild; stroke.