Abstract NS4: Challenges and Lessons Learned in an Ongoing Randomized Controlled Trial to Test the Effectiveness of a Multicomponent Intervention in Improving Delirium Outcomes in Acute Stroke

Delirium in acute stroke has higher morbidity and mortality than those without delirium. This two-group randomized controlled trial (RCT) tests if a multicomponent intervention improves delirium outcomes in stroke patients at a comprehensive stroke center. This presentation describes the challenges in managing a RCT in acute stroke. Scientific rigor requires coordinating staff, enrolling adequate sample size, assuring intervention fidelity, and fostering data integrity. A sample of 282 subjects is required for 80% power (α=0.05) to determine a 10% reduction in incident delirium. Eligibility includes acute stroke, aged ≥ 50 years, no aphasia or delirium on admit. Subjects randomized to Usual Care (UC) or Delirium Care (DC). Both groups receive standardized Stroke Care. DC subjects receive a multicomponent intervention: 1) pharmacist recommendations using Anticholinergic Drug scale scores; and 2) therapeutic activities. NIHSS, Montreal Cognitive Assessment (MoCA), Confusion Assessment Method (CAM), and mRS are used to determine primary (delirium) and secondary endpoints (LOS, neurological deficit, functional status). A total of 513 patients screened over 289 consecutive days required unbudgeted staff. The 310 excluded [aphasia (94), baseline delirium (36), critically ill (92), LOS <2 days (69), other causes (19)] unexpectedly impacted enrollment. Of eligible, 66% (133/203) consented; UC (n=65), DC (n=68). Hospital volunteers engaged DC subjects in therapeutic activities twice daily, including holidays. Two pharmacists independently made recommendations for each group. Outcome data were validated at daily rounds. Stroke-related cognitive dysfunction required more time than norm to complete the MoCA. Both the CAM and team consensus was used to confirm delirium. Lessons learned: 1) study staff attending rounds was key to enrollment, intervention fidelity, and data integrity; 2) securing a pool of on-call volunteers to sustain therapeutic activities was required; and 3) additional and unanticipated resources were needed. In conclusion, successful conduct of a RCT in acute stroke patients requires a dedicated well-trained study staff 7 days a week, including holidays.