Reach Out Emergency Department: A Randomized Factorial Trial to Determine the Optimal Mobile Health Components to Reduce Blood Pressure

Abstract

Background: Mobile health (mHealth) strategies initiated in safety-net Emergency Departments may be one approach to address the US hypertension epidemic, but the optimal mHealth components or dose are unknown. Methods: Reach Out is an mHealth, health theory-based, 2×2×2 factorial trial among hypertensive patients evaluated in a safety-net Emergency Department in Flint, Michigan. Reach Out consisted of 3 mHealth components, each with 2 doses: (1) healthy behavior text messaging (yes versus no), (2) prompted self-measured blood pressure (BP) monitoring and feedback (weekly versus daily), and (3) facilitated primary care provider appointment scheduling and transportation (yes versus no). The primary outcome was a change in systolic BP from baseline to 12 months. In a complete case analysis, we fit a linear regression model and accounted for age, sex, race, and prior BP medications to explore the association between systolic BP and each mHealth component. Results: Among 488 randomized participants, 211 (43%) completed follow-up. Mean age was 45.5 years, 61% were women, 54% were Black people, 22% did not have a primary care doctor, 21% lacked transportation, and 51% were not taking antihypertensive medications. Overall, systolic BP declined after 6 months (−9.2 mm Hg [95% CI, −12.2 to −6.3]) and 12 months (−6.6 mm Hg, −9.3 to −3.8), without a difference across the 8 treatment arms. The higher dose of mHealth components were not associated with a greater change in systolic BP; healthy behavior text messages (point estimate, mmHG=−0.5 [95% CI, −6.0 to 5]; P =0.86), daily self-measured BP monitoring (point estimate, mmHG=1.9 [95% CI, −3.7 to 7.5]; P =0.50), and facilitated primary care provider scheduling and transportation (point estimate, mmHG=0 [95% CI, −5.5 to 5.6]; P =0.99). Conclusions: Among participants with elevated BP recruited from an urban safety-net Emergency Department, BP declined over the 12-month intervention period. There was no difference in change in systolic BP among the 3 mHealth components. Reach Out demonstrated the feasibility of reaching medically underserved people with high BP cared for at a safety-net Emergency Departments, yet the efficacy of the Reach Out mHealth intervention components requires further study. Registration: URL: https://www.clinicaltrials.gov ; Unique identifier: NCT03422718.