Engagement in Mobile Health Prompted Self-Measured Blood Pressure Monitoring among Participants Recruited from a Safety-Net ED: The Reach Out Trial (Preprint)

BACKGROUND Hypertension, a key modifiable risk factor for cardiovascular disease, is more prevalent among Black and low-income individuals. To address this health disparity, leveraging safety-net Emergency Departments for scalable mobile health (mHealth) interventions, specifically utilizing text messaging for self-measured blood pressure (SMBP) monitoring, presents a promising strategy. This study investigates patterns of engagement, associated factors, and the impact of engagement on lowering blood pressure in an underserved population. OBJECTIVE To identify patterns of engagement with prompted SMBP monitoring with feedback, factors associated with engagement, and the association of engagement with lowered BP. METHODS This is a secondary analysis of data from Reach Out, a mobile-health, factorial trial among 488 hypertensive patients recruited from a safety-net ED in Flint, Michigan. Reach Out participants were randomized to weekly or daily text message prompts to measure their BP and text in their responses. Engagement was defined as BP response to the prompt. k-means clustering algorithm and visualization were used to determine the pattern of SMBP engagement by SMBP prompt frequency, weekly or daily. BP was remotely measured at 12 months. For each prompt frequency group, logistic regression models were used to assess the univariate association of demographics, access to care, and comorbidities with high engagement. We then used linear mixed-effects models to explore the association between engagement and systolic BP at 12 months, estimated using average marginal effects. RESULTS For both SMBP prompt groups, the optimal number of engagement clusters was 2, which we defined as high and low engagement. Of the 241 weekly participants, 189 (78%) were low (mean response rate=20% (sd=23.4), and 52 (22%) were high (mean response rate 86% (sd=14.7) engagers. Of the 247 daily participants, 221 (89%) were low (mean response rate=9% (sd=12.2), and 26 (10%) were high (mean response rate=67% (sd=8.7)) engagers. Among weekly participants, those who were older (>65 years), attended some college (vs. no college), married/lived with someone, Medicare (vs. Medicaid), under the care of a primary care doctor, and took antihypertensive medication in the last 6 months had a higher odds of high engagement. Participants who lacked transportation to appointments ) had lower odds of high engagement. In both prompt frequency groups, participants who were high engagers had a greater decline in BP than low engagers. CONCLUSIONS Participants randomized to weekly SMBP monitoring prompts responded more frequently overall and were more likely to be classed as high engagers compared to participants who received daily prompts. High engagement was associated with a larger decrease in BP. New strategies to encourage engagement are needed for participants with lower access to care. CLINICALTRIAL NCT03422718 INTERNATIONAL REGISTERED REPORT RR2-https://doi.org/10.1186/s13063-020-04340-z