Community Hypertension Screening and Care Referral With Blood Pressure-Measuring Kiosks, Digital Education Modalities, and Text Messages

Hypertension is a major modifiable cause of cardiovascular disease (CVD) that affects 46% of US adults.^1,2 Its annual societal costs in the United States are $131 billion and among all diseases, hypertension contributes to the greatest proportion of disability-adjusted life years.^3,4 Blood pressure (BP) lowering medications are inexpensive, safe, well-tolerated, proven therapeutics that greatly reduce the CVD risk related to hypertension.⁵ Despite the personal and societal costs of hypertension, BP control in the United States is abysmal. An estimated 40 million adults have BP ≥140/90 mm Hg, of which 13 million are unaware that they have hypertension.^6,7 The burden of hypertension prevalence, inadequate control, and complications are disproportionate among certain subgroups in the United States. For example, Black US adults have the highest hypertension prevalence of any group in the United States, with earlier onset, poorer control, and consequently suffer more CVD events than White US adults.^8,9

A critical first step in reducing CVD risk among those with hypertension is identification of elevated BP in individuals and providing education about the importance of clinical care of hypertension with BP-lowering medications. BP measurement kiosks are commonly found in pharmacies, shopping centers, and grocery stores in the United States and can provide a measurement in a few moments. There is an obvious opportunity to use these devices to also educate users and refer those with elevated BP toward care.

In the current issue of the American Journal of Hypertension, Dr Steven Shea and colleagues report the results of the pilot for the Retail Outlet Health Kiosk Hypertension Trial (ROKHYT).¹⁰ This feasibility trial enrolled 140 email-registered adult users of a commercial, automated BP measurement kiosk with ≥2 measurements in the year prior to study initiation, 1 of which had to be within 90 days of enrollment, with weight <300 lbs. At the time of kiosk use, the device presents BP measurements classified in BP-level groups as defined in the American College of Cardiology/American Heart Association (ACC/AHA)-led 2017 guidelines (e.g., normal BP for BP <120/80 mm Hg, hypertensive crisis for BP ≥180/90 mm Hg). Those with hypertensive crisis are advised to seek immediate care by the kiosk.

Of the 2,000 registered kiosk users from across the United States invited to participate, 177 provided consent and 140 were ultimately randomized to either intervention or usual care. The intervention group received a multifaceted intervention. Electronically delivered educational content covered topics related to elevated BP, the use of BP-lowering medications, diet, and exercise in hypertension management, and details about accessing care and adherence to therapies. Participants in this group were also contacted with interactive text messages that included a copy of their most recent BP measurement. Those with prior kiosk-measured BP above the ACC/AHA 2017 stage 2 hypertension threshold (140/90 mm Hg) were advised via text messages to seek care for treatment of high BP and return for a kiosk-measured BP in 2 weeks, with reminder messages for those who did not return for repeat measurement. At that repeat measurement, they reported if they saw a clinician for management of high BP. Participants with prior BP <140/90 mm Hg received educational text messages related to BP optimization and the role of clinical management of BP. Participants in the intervention group were also asked to follow-up at a kiosk at 1 month for a repeat BP measurement, then again at 3, 6, and 12 months. The usual care group did not receive any educational content or text messages except for reminders to have repeat kiosk BP measurements at 3, 6, and 12 months. The primary outcome was the proportion of adults with BP <140/90 mm Hg at 12 months. With the Covid-19 pandemic’s emergence in Spring 2020, these kiosks were closed, and the 6-month BP measurements were used in place of the 12-month BP measurements for those who had not yet completed the 12-month measurements.

ROKHYT participants reported a mean (SD) age of 52 (12) years, 48% were female, 71% reported White race, 12% Black race, and 12% Hispanic, Latino, or Latina. Participants overall had a high level of education as only 9% reported educational attainment of high school or less. There was relatively low proportion without access to care as only 11% reported no health insurance, and 14% reported no regular healthcare provider. As is expected in a pilot study assessing trial feasibility, there was no observed difference in the primary outcome. Specifically, the proportion of participants with BP <140/90 mm Hg was 37% in the intervention group and 27% in the usual care group (difference 10.3%; 95% confidence interval −6.2% to 26.8%).

The authors are to be applauded for their undertaking of this pilot that combined several electronic platforms into 2 arms of a clinical trial: automated BP measurement, targeted recruitment with email, electronic obtaining of consent, digital delivery of educational content via email and text messages, and text-message reminders. Did the authors prove feasibility of this platform for use in a large clinical trial? In our opinion, this is a resounding yes. Even the largest pandemic in a century only required a simple modification in the analysis protocol. This is promising news from a feasibility standpoint. There appears to be a clear path to a larger trial; however, we think that there should be greater representation of persons who are non-White and less educated and those with lower access to care (in terms of insurance and connections to a regular healthcare provider) to realize its full potential. Such groups have been underrepresented in prior NIH-sponsored CVD-related trials.¹¹ An effective kiosk-based BP screening and recommendation into care pathways could improve hypertension identification and control in the United States and potentially improve disparities in hypertension outcomes among groups with excess burden, like Black US adults, if implemented equitably.

We think that there are a few barriers to implementation of the ROKHYT protocol on a national scale, even with positive findings in an adequately powered, large clinical trial. First, there is a sordid history of unvalidated BP-measuring kiosks in the United States, so the accuracy of most kiosk devices is unclear.¹² The authors note that the device used in ROKHYT has not undergone rigorous validation. Inaccurate, falsely elevated BP measurements might drive unnecessary healthcare utilization and mistrust of out-of-office measurements. Falsely nonhypertensive BP measurements are also problematic. We previously described that reassuringly normal BP measurements from an inaccurate BP-measuring device improves user experience and drives users to want to repeat their BP measurement in the future.¹³ Therefore, adults with hypertension might preferentially use kiosks that provide falsely reassuring BP measurements. Such inaccurate, reassuring measurements might serve as justification for disregarding clinical recommendations about BP-lowering medication initiation or intensification based upon elevated in-office BP measurements. Second, the use of these devices needs to be proven among the groups at the highest risk, which entails consideration of structural barriers. The retail landscape suffers from inequities shaped by racial residential segregation, which could limit access to populations who arguably stand to benefit the most. It is well known that racial and ethnic minority residents live in neighborhoods that are disproportionately food deserts^14,15—lacking supermarkets—and pharmacy deserts^16,17—lacking pharmacies—prime locations for kiosk stations. Thus, the geographical distribution of the kiosks may very well drive the demographics of the users represented in the pilot. Although the product used in the pilot is available within 49 states with kiosks located within 5 miles of most of the population (73%, per the authors), deliberate efforts are required to ensure equitable access to those representing the minority. While 5 miles seems like a short distance, swift and drastic changes in health risks related to the neighborhood environment can differ within as few as 3 miles in urban settings.^18,19 From the perspective of health equity, future studies might consider partnering with trusted community organizations, such as churches or community centers, and the manufacturer to place kiosks in locations more easily accessible to populations who have been historically marginalized as both patients and consumers. Finally, equitable outreach might also entail additional considerations of the digital divide. ROKHYT’s protocol uses a mobile-accessible educational platform which may address an initial digital access barrier—85% of the US population owns a smartphone.²⁰ However, individuals with low income disproportionately rely on these devices to access the internet due to lack of broadband at home, which could make viewing and engaging with educational videos problematic for individuals with limited data plans. Human outreach through partnership with trusted entities might improve on both the impersonal nature of digital communications, and also allow for involvement of adults with barriers to digital interventions, like unreliable or complete lack of internet access.

Ultimately, there is a critical need for community-based interventions to improve hypertension screening and care referral for adults with elevated BP. Scalable, accessible technologies like those integrated into the ROKHYT pilot have great promise to move the needle on the atrocious state of hypertension screening and BP control in the United States. Effective interventions are greatly needed, especially for high-risk groups like Black US adults. We look forward to Dr Shea and colleague’s next step in their very promising work.

Funding

This work was supported by National Heart, Lung, and Blood Institute (K01HL152011) and Bloomfield Early Career Professorship in Cardiovascular Research.

DISCLOSURE

The authors declared no conflict of interest.