Strategies to improve medication adherence and blood pressure among racial/ethnic minority populations: A scoping review of the literature from 2017-2021

Abstract

Purpose of Review:

To identify intervention strategies that were effective in promoting medication adherence and HTN control among racial/ethnic minority groups in the US.

Recent Findings:

Twelve articles were included in this review and 4 categories of intervention strategies were identified as counseling by trained personnel, mHealth tools, mHealth tools in combination with counseling by trained personnel, and quality improvement. The findings show that interventions delivered by trained personnel are effective in lowering BP and improving medication adherence, particularly for those delivered by health educators, CHWs, medical assistants, and pharmacists. Additionally, the combination of mHealth tools with counseling by trained personnel has the potential to be more effective than either mHealth or counseling alone and report beneficial effects on medication adherence and BP control

Summary:

This review provides potential next steps for future research to examine the effectiveness of mHealth interventions in combination with support from trained health personnel and its effects on racial disparities in HTN outcomes.

INTRODUCTION

Hypertension (HTN) prevalence in the United States (US) remains unacceptably high. HTN affects nearly half (45%) of American adults each year.[1•] It is a leading cause of cardiovascular disease (CVD),[2] which kills approximately 1 in 3 American adults annually, with minority populations being disproportionately affected.[3, 4] According to national data, the prevalence of HTN is highest among Blacks, followed by Whites, Asians, and Hispanics, respectively.[5•] An analysis of 2013-2014 New York City (NYC) Health and Nutrition Examination Survey data found that non-Hispanic Blacks (43.5%); Asians (38.0%); and Hispanics (33.0%), all experience greater HTN prevalence than Whites (27.5%).[6] Furthermore, although HTN control is imperative for the prevention of CVD, it is lowest among racial/ethnic minority groups. A recent analysis of National Health and Nutrition Examination Survey data found that White adults had the highest HTN control rates (55.7%), followed by non-Hispanic Blacks (48.5%), Hispanics (47.4%), and Asians (43.5%).[7] These findings underscore the importance of understanding and collating targeted strategies to improve HTN outcomes among racial/ethnic minority groups.

Medication adherence is fundamental to achieving HTN control. It is defined by the World Health Organization as “the extent to which a person’s behavior-taking medications, following a diet, and/or executing lifestyle changes corresponds with agreed recommendations from a healthcare provider.”[8] Recent definitions build on this, adding initiation and implementation of, as well as adherence to, a healthcare provider-recommended treatment plan.[9••, 10] Prior research has established a positive relationship between adherence and HTN control: adults with better adherence to antihypertensive medications are 45% more likely to achieve HTN control than their middling or low-adhering counterparts.[11] Despite this evidence, racial/ethnic minorities with HTN often exhibit lower rates of medication adherence than Whites with HTN.[12, 13]

Existing research demonstrates the positive relationship between medication adherence and HTN control, and prior reviews elucidate the intervention strategies most efficacious in improving medication adherence among hypertensive participants.[14•] However, despite persistent disparities in HTN outcomes among racial/ethnic minority groups, to date, only a few reviews[14•] examine medication adherence interventions that target HTN disparities specifically among these groups. The purpose of this review was two-fold: (1) to critically evaluate the recent hypertension literature (2017-2021) and identify intervention strategies that were effective in promoting medication adherence and HTN control among racial/ethnic minority groups in the US, and (2) examine if the outcomes of medication adherence and blood pressure reduction differed by race. Findings from this study will advance the understanding of efficacious strategies that could reduce HTN disparities among these groups.

METHODS

Following the PRISMA guidelines for Scoping Reviews,[15] we reviewed available studies on hypertension and medication adherence in minority populations. In February 2021, a trained medical librarian (T.R.) searched the MEDLINE, Embase, and Cochrane Central Register of Controlled Trials (CENTRAL) databases using the Ovid search platform and the CINAHL database using the EBSCO platform. The searches included a combination of subject headings and keywords on hypertension medication adherence along with terms for health disparities and racial/ethnic minorities. The complete Ovid Medline search is included as an appendix to this article. The search was not limited by language but was limited by year from 2017-present.

The criteria for inclusion in the review were as follows: 1) primary research studies conducted between 2017-2021; 2) had a medication adherence focus with hypertension/high blood pressure as an outcome; 3) primarily targeted racial/ethnic minorities; 4) were US-based, and 5) included participants ages 18 years and older. Randomized and non-randomized studies were included, as well as single and multicomponent intervention designs. We included studies that targeted medication adherence across multiple levels of influence (i.e., organizational, community). Studies were excluded from the review if they were systematic reviews, commentary, or opinion articles. Studies that targeted chronic disease and medication adherence, but omitted HTN as an outcome, were also excluded.

Three authors [DO, AB & CC] underwent three review phases in Covidence, an electronic systematic review management system: 1) abstract and title screening, 2) full-text review, and 3) data extraction. All studies were reviewed for alignment with inclusion and exclusion criteria. Initially, each study was independently reviewed by two of the three authors. Articles which achieved a consensus independently were either automatically moved to full-text review or excluded from the study altogether. All 3 authors convened to discuss articles that did not achieve a consensus during independent review. Upon unanimous agreement, the remaining articles were then either moved to full-text review or excluded from the study. In the case of a disagreement between 2 authors, the author who did not review the study served as the tie-breaker.

During the full-text review, the qualifying articles underwent an in-depth review against inclusion and exclusion criteria. In keeping with the title and abstract review process, articles that reached independent consensus were automatically included or excluded. The remaining articles were collectively reviewed by all 3 authors and, upon unanimous consensus, articles which met the inclusion criteria were extracted.

During extraction, the authors convened to discuss and finalize the extraction plan. The plan specified data elements for extraction, which included: demographics, methods, intervention characteristics, outcome measures, and results for the total population by race/ethnicity. Two authors independently reviewed and extracted qualifying studies. All 3 authors convened to review and finalize the extracted data for analysis.

RESULTS

In total 825 articles were identified from the various databases for potential screening (See Prisma Table). After duplicates were removed, the title and abstracts of 561 unique articles were screened, of which 494 were deemed ineligible. Thus, a total of 66 articles were extracted for full review. Fifty-four of these articles were excluded for the following reasons: there was no intervention (n=30); lacked a disparity focus (n=6); the study was still ongoing (n=6); medication adherence was not included (n=6); did not focus on patients with HTN (n=3); was not based in the US (n=2); and the study targeted pregnant women (n=1). Thus, 12 articles were included in this review. Table 1 shows the characteristics of the included studies. The most common study designs included randomized controlled trials (RCT, n=6),[16•-21] followed by pre-post assessments (n=4)[22-25•]; prospective interventional clinical trial (n=1)[26•]; and prospective cohort (n=1)[27](see PRISMA table). Sample sizes ranged from 23 to 525 (median: 54), and study duration ranged from 8 weeks up to 24 months (median: 6 months).

Table 1-.

Study Characteristics

Study Reference	Study Design	Intervention Duration	Sample Size	Age	Females	Education	Income	Race	Intervention Objective
Li et al, 2020	Pre-Post	8 weeks	23	Not specified	65.2%	Less than HS: 69.5% HS or higher: 30.4%	<= $9,999: 52.2% $10,000-$99,999: 4.3%	Asian	Community Health Coaching for Chinese immigrants to improve blood pressure control and medication adherence using
Cene et al, 2017	Prospective cohort	24 months	525	58	68%	HS or less: 73% Higher than HS: 27%	<= $40,000: 78% > $40,000: 22%	African American Caucasian	Multicomponent practice-based quality improvement intervention to lower SBP among patients with uncontrolled HTN and examine differences by race
Taber et al, 2018	Prospective Interventional clinical Trial	6 months	60	59	41.7%	Completed HS: 47.4%	< $50,000: 81.5% >= $50,000: 18.5%	African American Other	Pharmacist-led, technology-aided intervention on improving CVD risk factor control within kidney transplant recipients while also assessing if the potential improvements in CVD risk factor control differ by race
Chandler et al, 2019	RCT	9 months	54	46.5	39%	HS or less: 71% College or higher: 29%	<= $50,000: 100% > $50,000: 0%	Hispanic/Latino	mHealth Medication Regimen Self-Management using SMASH app (Smartphone Med Adherence Stops Hypertension) Program for Blood Pressure control and medication adherence among Hispanics)
Rashid et al, 2017	Pre-Post	9 weeks (Chicago) 3 months (Houston)	97	Chicago: 45-59 Houston: 55+	Chicago: 73% Houston: 41%	Chicago: Graduated HS: 36% Some college: 33% Houston: Middle School or lower: 8.5% College Graduates: 10.6%	Chicago: Not specified Houston: Not specified	African American Hispanic/Latino Asians	Two-city intervention with virtual training institutes where intervention staff learned cultural competency methods of adapting effective interventions. Health educators delivered the Health Empowerment Lifestyle Program (HELP) in Chicago; Community pharmacists delivered the MyRx Medication Adherence Program in Houston
Schoenthaler et al, 2020	RCT	6 months	102	60.8	50%	Less than HS: 58.8% HS or higher: 41.2%	<$20,000: 37.8% >=$20,000: 62.2%	Hispanic/Latino	Systems-level adherence intervention, delivered by medical assistants versus a comparison condition on medication adherence and blood pressure (BP) in hypertensive Latino patients who were initially non-adherent to their antihypertensive medications.
Schoenthaler et al, 2020	RCT	3 months	42	57.6	45.2%	HS or less: 66.7% College or higher: 33.1%	<$20,000: 62.8% >=$20,000: 37.2%	African American	mHealth adherence intervention versus attention control on medication adherence, systolic blood pressure, diastolic blood pressure, and hemoglobin A1c (HbA1c) in patients with uncontrolled hypertension and/or Type 2 diabetes who were initially nonadherent to their medications.
Schroeder et al, 2020	RCT	12 months	295	53.4	59.7%	HS or less: 58.5% College or higher: 41.5%	<$20,000: 60.4% >=$20,000: 39.7%	Alaskan Native Hispanic/Latino Caucasian African American Other	Interactive voice response and text message (IVR-T) intervention for individuals to improve blood pressure (BP), medication adherence and visit keeping among adults with hypertension from multiple racial and ethnic groups in primary care setting
Still et al, 2020	RCT	12 weeks	60	59.5	75%	Not reported	<$20,000: 26.7% >=$20,000: 36.7%	African American	Community and technology-based intervention for hypertension self-management intervention on blood pressure control and health-related quality of life in African Americans with hypertension
Warren-Findlow et al, 2019	Pre-Post	12 months	52	>50	59.6	Not reported	Not reported	African American Hispanic/Latino	2-hour health education class on self-care behaviors to test the acceptability and feasibility of a brief health literacy intervention focused on hypertension self-care and to assess changes in self-care activities.
Wheat et al, 2020	Pre-Post	6 months	33	65.4	75.8%	Not reported	Not reported	African American Native American Caucasian	collaboration between pharmacists and CHWs in identifying and addressing barriers to medication adherence and improving health outcomes for patients diagnosed with hypertension with or without diabetes from diverse backgrounds, including Native Americans and blacks.
Zha et al, 2020	RCT	6 months	25	48.9	83%	Not reported	Not reported	African American Hispanic/Latino	mHealth intervention on changes in systolic and diastolic BP, adherence in monitoring BP, perceived medication adherence self-efficacy, and self-reported health-related quality of life.

Study Population

The average age of the population included in the studies ranged from 45 to 65 years, with majority being female. Eight of the 12 studies included in this review reported on education level of the participants, with majority having a high school diploma or less.[16-18, 21-23, 26, 27] Similarly, income was also reported in 8 of the 12 studies, with majority of participants earning less than $40,000 per year.[16-19, 21, 23, 26, 27]

Racial minority groups made up the majority of the study population for this review. Five of the 12 studies included just one racial minority group (i.e. Asians or Hispanic/Latino, or African American).[16•, 17, 19, 21, 23] Four studies included two racial groups: Two studies included African American and Latino/Hispanic,[20•, 24] one study included African Americans and Caucasians,[27] while the last study included African Americans and non-African Americans (not specified).[26] Finally, three of the 12 studies included three or more racial groups[18, 22, 25] (See Table 1).

Assessment of Medication Adherence and Blood Pressure

Nine of the 12 studies defined medication adherence as an increase in self-reported adherence score post-intervention.[16-18, 20, 21, 23, 24, 26, 27] For the other three, one did not report a definition,[22] the second defined it as a lower score on a 4-point Likert-type scale to indicate better medication-taking adherence behaviors,[19] and the third study defined adherence in terms of barriers to medication adherence identified by patients.[25•] To measure medication adherence, four studies used the 8-Item Morisky Medication Adherence Scale (MMAS-8).[16, 17, 21, 26, 27] Another study used adapted versions of the MMAS-8 in combination with other scales (i.e. adapted from Morisky, Green, and Levine scales) (n=1).[23] The remaining studies used the following instruments to assess medication adherence: the Voils scale (n=1),[18] an electronic medication tray with timestamped data (n=1),[16]] an electronic monitoring device (n=1),[17] Hill-Bone Compliance to High Blood Pressure Therapy Scale(n=1),[19] Two scales for medication adherence developed using 4 survey items, with a weight of 50 points for each item[22] and a final study using the Drug Adherence Workup Tool, which was an interview-style tool between a Community Health Worker (CHW) and patient (n=1).[25]

Blood pressure (BP) was assessed using different versions of automated BP machines. Most measurements occurred on-site at a clinic and were measured by trained personnel (nurse, community health worker, pharmacist, etc.). Nine studies defined uncontrolled HTN as systolic BP >=140 mm Hg and diastolic BP as >=90 mm Hg. [16-21, 23, 26, 27] Three studies used a lower threshold to define poor HTN control (i.e., systolic BP >=130 mm Hg and diastolic BP >=80 mm Hg).[22, 24, 25] Table 2 includes a description of the medication adherence and BP assessments included in each study.

Table 2-.

Study Characteristics Cont’d

Study Reference	Intervention Delivery	Adherence Measure	BP Definition	Disparity being addressed	Intervention improved Medication Adherence	Intervention improved Blood Pressure	Differential Outcome by Race
Li et al, 2020	Community Health Coaches	Medication Adherence Scale adapted from Morisky, Green, and Levine scales	140/90 mm Hg	Improving blood pressure control in Chinese immigrants	No	Yes- There was a significant reduction in both systolic and diastolic blood pressure from baseline to week 8	N/A
Cene et al, 2017	Quality Improvement	8-item Morisky Medication Adherence Scale	140/90 mm Hg	Hypertension control (SBP) and observed differences by race	No- But at 12- and 24-months attendees were more likely than non-attendees to be older, AA, and have moderate-high medication adherence.	Yes- BP was fairly well-controlled with no significant difference in mean SBP by race	The intervention was not differentially more effective in African Americans than whites.
Taber et al, 2018	mHealth + pharmacist	Patient-reported medication adherence using a 8-Item Medication Adherence Scale	140/90 mm Hg	Assessing if the impact of the intervention varied by race	Yes- Self-reported medication adherence significantly improved over the course of the study.	Yes- Overall, patients demonstrated improvements in blood pressure	African Americans demonstrated a significant reduction from baseline in SBP and DBP compared to non-African Americans
Chandler et al, 2019	mHealth	Medication tray timestamped data AND Morisky medication adherence scale	140/90 mm Hg	Hypertension control among Hispanic/Latinos to lower SBP	Yes- Mean medication adherence (Morisky self-report) improved among intervention group from Baseline to 9 months	Yes- hypertension was controlled over 9 months in the intervention group	N/A
Rashid et al, 2017	Health educators	Two scales for medication adherence developed using 4 survey items, with a weight of 50 points for each item	130/80 mm hg	Type 2 Diabetes, HTN and Obesity by race (among Black, Hispanic and Asia)	Yes- Not Significantly	Yes- In both groups BP was lowered but only significantly in the group who received intervention from Health educators	N/A
Schoenthaler et al, 2020	Medical Assistants	Electronic-monitoring device (EMD) & self-reported medication adherence measured by the eight-item Morisky Medication Adherence Scale	140/90 mm Hg	Blood pressure control among Latinos	Yes- The intervention group had a greater improvement in self-reported adherence when measured using the MMAS-8.	Yes- SBP and DBP significantly improved in both groups across the 6 months, but there was greater reduction in the intervention group	N/A
Schoenthaler et al, 2020	mHealth	Morisky Medication Adherence Scale	140/90 mm Hg	Racial disparities in BP and glycemic control	Yes- There was a significant improvement in adherence across the 3-month study for both group, however, there were no between-group differences.	Yes- SBP significantly improved in the intervention group compared to the usual care group. DBP improved and there was greater reduction in the intervention group but this was not significant.	N/A
Schroeder et al, 2020	mHealth	Voils instrument	140/90 mm Hg	Improving Hypertension care among American Indians or Alaska Natives and other vulnerable populations in Albuquerque, New Mexico	Yes- Self-reported medication adherence improved to a comparable extent between baseline and 6months in both study arms, but not at 12 months	No	N/A
Still et al, 2020	mHealth + Nurse Counselling	Hill-Bone Compliance to High Blood Pressure Therapy Scale & Medisafe app	140/90 mm Hg	HTN self-management & improving blood pressure control in African Americans	Yes- Moderate significant medication adherence in the intervention group	Yes- only DBP improved	N/A
Warren-Findlow et al, 2019	Health educators		130/80 mm Hg	Improve health literacy related to medication adherence and weight management activities among under- and uninsured primary care patients with hypertension.	Yes- Not significantly	Yes- Not significantly	Hypertension self-care adherence levels were overall higher in Hispanics than in non-Hispanics. Two-thirds of Hispanics were adherent to medication regimens as compared to less than 40% of non-Hispanics.
Wheat et al, 2020	Community Health workers +Student Pharmacists	Drug Adherence Workup Tool (interview style between CHW and patient)	130/80 mm Hg	Improving medication adherence and patient outcomes in Native American or black minority groups.	Yes	Yes- There was a statistically significant reduction in mean blood pressure between the initial and final patient encounters for the patients who participated in the CHW-pharmacist collaboration pilot program	N/A
Zha et al, 2020	mHealth + primary care provider	Medication Adherence Self-Efficacy Scale	140/90 mm Hg	The effectiveness of strategies to address disparities in access to care and improve health outcome	Yes- Not significantly	Yes- Not significantly	N/A

Overview of intervention strategies

All 12 studies included in this review utilized one of four main strategies to improve medication adherence and reduce BP. Five studies (42%) trained different personnel to deliver interventions: these included health educators, pharmacists, and CHWs. Three studies (25%) utilized mobile health (mHealth) tools to virtually manage adherence and BP (e.g. smartphone apps, text and voice messaging systems, etc.) and three studies (35%) used a combination of mHealth tools and support from trained personnel to deliver the intervention (e.g. patients were able to monitor BP virtually and providers intervened when necessary). One study (8%) utilized a practice-based quality improvement strategy to deliver a medication adherence and BP intervention.[27] Below, we describe the findings of these studies, categorized by the intervention type. Tables 2 – 4 also include details on each of the intervention strategies.

Table 4:

Medication Adherence

Study Reference	Adherence Measure	Adherence Baseline	SD	Adherence Follow-Up	SD	Adherence Change	p-value
Li et al, 2020	Medication Adherence Scale adapted from Morisky, Green, and Levine scales	10.56	3.24	10.89	3.95	−0.33	0.86
Cene et al, 2017	8-item Morisky Medication Adherence Scale	Low adherence: Overall: 40% AA- 42% W- 34%	Not reported	Not reported	Not reported	Not reported	Not reported
Taber et al, 2018	Patient-reported medication adherence using a validated survey instrument	59.6% with high adherence	NA	89.5% with high adherence	Not reported	+29.9% with high adherence	Not reported
Chandler et al, 2019	Medication tray timestamped data AND Morisky medication adherence scale	SMASH: 6.83 ESC: 6.99	1.99 2.39	SMASH: 9.81 ESC: 6.84	1.31 1.52	SMASH: +2.97 ESC:−0.15	p<0.001
Rashid et al, 2017	Two scales for medication adherence developed using 4 survey items, with a weight of 50 points for each item	Chicago: Not Reported Houston: 190.2	Chicago: Not reported Houston: 14.53	Chicago: Not reported Houston: 195.9	Chicago: Not reported Houston: 6.78	Chicago: Not reported Houston: +5.7	Chicago: Not reported Houston: 0.15
Schoenthaler et al, 2020	Electronic-monitoring device (EMD) & self-reported medication adherence measured by the eight-item Morisky Medication Adherence Scale	Intervention: EMD: 84.1 MMAS-8: 4.48 Control: EMD: 79.0 MMAS-8: 4.75	Intervention: EMD: 26.8 MMAS-8: 1.41 Control: EMD: 23.6 MMAS-8: 1.14	Intervention: EMD: 74.5 MMAS: 6.46 Control: EMD: 72.4 MMAS-8: 6.01	Intervention: EMD: 23.5 MMAS-8: 1.65 Control: EMD: 22.2 MMAS-8: 1.35	Intervention EMD: −9.6% MMAS: +1.98 Control: EMD: −6.6% MMAS: +1.26	EMD: 0.66 MMAS: 0.03
Schoenthaler et al, 2020	Morisky Medication Adherence Scale	Intervention- 4.4 Control- 4.0	1.3 1.3	5.6 5.5	2.0 2.1	+1.2	0.50
Schroeder et al, 2020	Voils instrument	IVR-T: 38.3% UC: 34.5%	Not Reported	IVR-T: 47.9% UC: 47.5%	Not reported	+0.25 0.38	0.65
Still et al, 2020	Hill-Bone Compliance to High Blood Pressure Therapy Scale & Medisafe app	Intervention :12.00 UC: 13.38	3.43 6.09	10.72 11.93	2.20 4.43	−1.28 −1.45	0.023
Warren-Findlow et al, 2019	Hypertension Self-Care Activity Level Effects (H-SCALE) self-report questionnaire & A 4-item measure scale adapted from Morisky, Green, & Levine	18.39	4.41	18.76	4.27	+0.36	0.50
Wheat et al, 2020	Drug Adherence Workup Tool (interview style between CHW and patient)	Not Reported	Not Reported	Not Reported	Not Reported	Not Reported	Not Reported
Zha et al, 2020	Medication Adherence Self-Efficacy Scale	mHealth: 64.85 SC: 64.75	11.44 10.85	mHealth: 69.17 SC: 61.00	7.77 13.08	−4.31 −3.75	0.06

Intervention strategies delivered by trained personnel

Four of the five studies which trained personnel to deliver the intervention reported a within-group improvement in medication adherence post-intervention. Additionally, all five studies reported a within-group improvement in BP control post-intervention. Two studies also reported a significant improvement in BP control in favor of the intervention group as compared to the comparison group.[17•, 22]

In a pre-post study, which targeted first-generation Chinese immigrants and immigrants born in mainland China (who speak Cantonese), community health coaches delivered a culturally and linguistically tailored intervention using video storytelling, group presentation, and a question and answer session.[23] Medication adherence was assessed using the Medication Adherence Scale adapted from Morisky, Green, and Levine scales. The study reported a significant reduction in both systolic and diastolic BP from baseline to 8 weeks (Mean change: −17.33 mm Hg /9.58 mm Hg, p<0.005). However, there was no significant change in medication adherence post-intervention.

Another pre-post study consisting of African Americans, Hispanic/Latino, and Asian patients, used health educators and community pharmacists to deliver culturally tailored lifestyle interventions for HTN and medication adherence in Chicago and Houston.[22] In this study, health educators delivered a health empowerment lifestyle program (Chicago) and community pharmacists delivered a prescription medication adherence program (Houston). Adherence to antihypertensive medications was assessed using 4 survey items with a weight of 50 points for each item. The study reported a non-significant improvement in medication adherence from baseline to 9 weeks for participants in Chicago and baseline to 3 months for participants in Houston. There was a significant reduction in systolic BP from baseline to post-intervention among participants who received the health empowerment lifestyle program in Chicago (Mean change: −9.79 mmHg, p<0.001). Change in diastolic BP was non-significant (Mean change: −2.42 mmHg, p=0.13). The mean change in BP from baseline to post-intervention was non-significant among those who received the medication adherence program from community pharmacists in Houston (−2.80 mm Hg (p=0.86) /−0.20 mm Hg (p=0.39)).

In a third pre-post study consisting of African Americans and Hispanic/Latino patients, health educators delivered a 2-hour health education class to improve health literacy and self-care behaviors.[24] Ten classes were taught with class sizes ranging from 3 to 10 participants per class. Medication adherence was assessed using Hypertension Self-Care Activity Level Effects (H-SCALE) self-report questionnaire and a 4-item measure scale adapted from Morisky, Green, & Levine. At baseline, about 62% of participants had SBP >130 mm Hg and 69% had DBP >80 mm Hg. At the 12-month follow-up, there was a non-significant improvement in self-reported medication adherence from baseline. Of note, about one-third of the study population still reported barriers to medication adherence at 12-months. Also at 12 months, researchers for this study were only able to obtain BP measures for 18 participants. However, only 3 of the 18 were within a 30- to 60-day, post-intervention window; the remainder were 2 to 6 months after the intervention and as such were not reported. [24]

Similar to the study conducted in Chicago and Houston,[22] the fourth pre-post study, consisting of African Americans, Native Americans, and Caucasian patients, utilized a collaboration between CHWs and student pharmacists to identify and address medication adherence barriers encountered by hypertensive patients, and create an action plan.[25] CHWs were also responsible for following up with patients each month throughout the study to document progress and modify the action plan. For this study, medication adherence was assessed using the Drug Adherence Workup Tool (i.e. interview-style interview between CHW and patient). At baseline, patients identified numerous barriers to medication adherence for antihypertensive medication (forgetfulness, adverse effects, refills, cost, and knowledge barriers). At the end of the 6-month intervention, 76% of these barriers had been resolved, as compared to baseline. Additionally, there was a significant reduction in BP from baseline to 6-month follow-up (Mean change: −5.90 (p=0.006) / −4.50 (p=0.008) mm Hg among study participants).

Finally, in an RCT study consisting of Hispanic/Latino patients, Medical Assistants were trained to deliver a culturally tailored system-level intervention targeted to improve adherence to an antihypertensive regimen and BP in Latino patients with uncontrolled HTN.[17•] Patients randomized to the intervention had greater improvements in self-reported adherence at 6-months, as assessed by the MMAS-8, than the usual care group, who received standard coaching at the clinic. BP also improved for both groups; however, the between-group difference was non-significant (Mean Change: −2.70 mmHg (p=0.34) /−0.60 mmHg (p=0.18)). [17•]

Intervention strategies using mHealth tools

Three studies that utilized mHealth tools reported improvements in medication adherence, while two out of the three studies reported improvements in BP control. In an RCT consisting of Hispanic/Latino patients, participants randomized to the intervention arm utilized a mobile application called SMASH (Smartphone Med Adherence Stops Hypertension) in combination with a Bluetooth-enabled BP monitor and electronic medication tray for self-monitoring their HTN.[16] The attention control arm received text messages that included links to PDFs and brief video clips containing healthy lifestyle tips. Medication adherence was assessed using an electronic time-stamped medication tray and the MMAS-8. Compared to the attention control group, medication adherence improved significantly in the intervention group at 9-months (p<0.001). BP also improved significantly for the intervention group SMASH (Mean Change:−30.50 mm Hg/−13.60 mmHg)

In another RCT, African American patients with HTN and type 2 diabetes completed a tablet-delivered intervention that included a tailoring survey based on the Information Motivational Behaviors Adherence Questionnaire to identify patients’ barriers to medication adherence.[21] Results from the survey were used to create an individualized adherence profile for each patient in the intervention group, which described the two prominent adherence barriers and develop a personalized list of adherence-promoting strategies that were matched to the adherence barriers.[21] This was compared to an attention control group who only received the tailoring survey without results and links to health education materials unrelated to medication adherence. Self-reported medication adherence assessed using the MMAS-8. At the 12-week follow-up, medication adherence improved significantly for both groups with no between-group difference. Similarly, systolic BP improved significantly for both groups (Mean change: −4.80 mmHg, p=0.04) with no between-group difference. The intervention did not have an effect on diastolic BP.[21]

Finally, an RCT study consisting of Alaskan Native, Hispanic/Latino, Caucasian, African American, and other races, used mHealth to deliver an interactive voice response and text message (IVR-T) intervention for individuals with HTN.[18] The mHealth intervention was designed to improve BP, medication adherence, and appointment keeping among adults with HTN from multiple racial/ethnic groups followed in primary care practices at an Urban Indian Health Organization in Albuquerque, New Mexico. Specifically, the intervention group received reminders from the IVR-T on three days and one day before their appointments; they also received message requests to reschedule missed appointments, monthly reminders for medication refills, as well as motivational messages to encourage self-care, appointment keeping, and medication-taking. The usual care arm only received reminders and not the additional motivational messages. Medication adherence was assessed using the Voils instrument, at 6-months. Self-reported medication adherence improved in both study arms with no between-group differences at 6 months and no improvement in self-reported medication adherence at 12 months. There was no effect on BP for both groups.

Intervention strategies using mHealth in combination with trained personnel

Three studies used a combination of mHealth and trained personnel to deliver the intervention. All three studies reported improvements in medication adherence and BP. The first study, which was a prospective interventional clinical trial consisting of African American and Non-African American patients, used a pharmacist-led, technology-aided, education intervention[26•] that was comprised of Bluetooth-enabled home monitoring devices for glucose and BP and monthly, face-to-face, pharmacist-led encounters. This intervention was designed to reduce patient-level barriers for CVD risk factor control, including medication adherence, and to promote the adoption of healthy lifestyle choices. Participants served as their own controls and self-reported medication adherence was assessed via an 8-Item Medication Adherence Scale. Self-reported medication adherence improved significantly over time among participants: odds of self-reported high medication adherence increased by 34% with each month in the study (OR: 1.34, [95% CI 1.10-1.64] p=0.004).[26] There was also a significant improvement in BP among African American patients, who demonstrated a greater monthly reduction in SBP from baseline to 6-month follow-up than non-African Americans (−0.86 mmHg per month (p=0.03) vs −0.13 mmHg per month (p=0.87), respectively).[26]

Another study used a community and technology-based intervention for hypertension self-management (COACHMAN) of HTN in African Americans.[19] The 12-week intervention consisted of web-based education modules, self-monitoring BP with an automatic home BP device, and a medication management app in conjunction with 30-minute nurse-led phone counseling to support self-management behaviors. The control group received printed educational material for HTN management and access to one educational web-based session for BP self-monitoring. Medication adherence was assessed using the Hill-Bone Compliance to High Blood Pressure Therapy Scale and the Medisafe app (www.Medisafe.com). Compared to the control group, the intervention group reported significantly better medication adherence at 12-weeks (p=0.002). There was a non-significant increase in SBP and a non-significant reduction in DBP in both groups (Mean change: +0.55 mm Hg (p=0.87)/−0.8 mm Hg (p=0.67) in the intervention group.

Finally, a study used the iHealth BP7 Wireless BP Wrist Monitor to test, track, and share patients' BP data with health care providers for remote patient monitoring and to promote patient self-monitoring.[20•] Patients included in this study were African Americans and Hispanic/Latino. The usual care group received standard follow-up with health care providers without the use of the BP wrist monitors. Both groups exhibited reductions in BP, with a significant mean change in SBP for the intervention group (Mean change: −8.39 mm Hg (p=0.01)/−2.76 mm Hg (p=0.06)).

Intervention strategies using quality improvement techniques

One study consisting of African Americans and Caucasian patients prospectively evaluated the effect of a multicomponent practice-based quality improvement intervention on systolic BP reduction in a cohort of African American patients with uncontrolled HTN compared to their white counterparts at 12 and 24-months post-baseline.[27] The intervention consisted of practice-level strategies (i.e. engaging health care stakeholders and quality improvement activities at practices) and patient-level strategies (telehealth coaching and home BP monitoring) to improve medication adherence and BP. The practice-level intervention included training practices and community-based coaches on communication and behavior change strategies based on social cognitive theory and motivational interviewing. At the patient level, coaches provided telephone counseling to help patients set HTN care goals, learn appropriate techniques for home BP measurement and monitoring, and discuss BP target values. At 12 months, there was a significant decrease in SBP in both African American and White groups with no significant group differences compared with baseline, with a mean change at 24months of: Overall- 6.40 mm Hg (p<0.001); African American- 6.00 mm Hg (p<0.001); White- 7.20 mm Hg (p<0.001). There was a non-significant improvement in medication adherence, as assessed with the MMAS-8. However, at 12 and 24-month study visits, attendees were more likely than non-attendees to be older, African American, and have moderate-high medication adherence.

Differential effect of intervention strategies on medication adherence and BP by participant race

Table 2 outlines the racial/ethnic differences in outcomes examined by each study. While all studies included racially/ethnically diverse samples, only seven of the 12 studies (58%) included individuals from more than one racial/ethnic group, to make a comparison of outcomes by race possible. Of these seven studies, three of the studies (43%), reported on medication adherence and BP outcomes by race. The Cene et al. study, which utilized a practice-based quality improvement strategy, found no differences in the effectiveness of the intervention on medication adherence and BP control when comparing African Americans and Whites. The Taber et al. study, which utilized a pharmacist-led, technology-aided intervention, reported that African Americans demonstrated a greater reduction in systolic and diastolic BP compared to non-African Americans, however, this reduction was significant when compared to the non-African American group.[26] Finally, the Warren-Findlow et al. study, which utilized health educators to improve health literacy related to medication adherence among under- and uninsured primary care patients with HTN found that two-thirds of Hispanic patients reported being adherent to medication regimens as compared to less than 40% of non-Hispanic patients at the end of the intervention.[24]

DISCUSSION

The American College of Cardiology and the American Heart Association published new HTN guidelines in 2017, defining HTN as 130/80 mm Hg.[28•] As such, the CDC reports that about 45% of adults who live in the US have HTN, and only about 24% of this population exhibit adequate HTN control.[1] HTN remains the leading cause of death for about half a million people in the US.[29]

About 67% of 30 million adults recommended to take anti-hypertensive medication still have a BP greater than or equal to 140/90 mm Hg. While the reasons for poor HTN control are multi-factorial, this data suggests some level of non-adherence to medication as a contributing factor.[1•, 29] The goal of this review was to provide a synthesis of recent interventions targeted at improving BP and medication adherence among racial/ethnic minority patients with HTN, and to document any differential outcomes by race. Twelve articles were included in this review and 4 categories of intervention strategies were identified: Counseling by trained personnel, mHealth tools, mHealth tools in combination with counseling by trained personnel, and quality improvement. The findings show that interventions delivered by trained personnel are effective in lowering BP and improving medication adherence, particularly for those delivered by health educators, CHWs, medical assistants, and pharmacists. Additionally, intervention strategies which combined trained personnel with mHealth tools were more likely to report beneficial effects on medication adherence and BP control.[19, 26] Alternatively, only one of the 3 studies which used mHealth tools alone reported significant results in improved medication adherence and BP control. Finally, quality improvement strategies that target both the practice and patient level also showed positive effects on medication adherence and BP control.

There is already robust literature documenting the positive effects of using trained personnel to deliver interventions targeting medication adherence in patients with HTN. For example, previous studies have noted the effectiveness of using CHWs (including community coaches and health educators) to deliver culturally relevant interventions for HTN control.[30-32] One systematic review of interventions delivered by trained health personnel reported significant improvements in several HTN self-management behaviors including adherence to anti-hypertensive medication, and in some studies, the CHW-led interventions also resulted in reductions in CVD risk.[31] Importantly, these improvements were predominately reported in studies that included racial/ethnic minority groups. We documented similar beneficial effects of utilizing trained personnel to deliver interventions on HTN-related outcomes in racial/ethnic minority patients.[22, 24] Thus, trained personnel like CHWs, health educators, and coaches could serve important roles in improving HTN control and medication adherence, especially among racial/ethnic minority groups.[30]

In recent years, there has been a growth in the number of studies examining the use of mHealth tools as potential intervention strategies to improve HTN control and medication adherence.[33-35] Our review suggests that the evidence base is mixed. For instance, while one study reported positive effects of the SMASH mobile app [16] on both BP and medication adherence another study reported no effect when testing an IVR and text messaging intervention.[18] These varying results may be due to differences in the intervention strategies, study designs, and length of the interventions (i.e., RCT with 9 months duration vs RCT with 12 months duration). This finding mirrors previous studies, which show that early effectiveness of mHealth tools in the short-term dissipate over time, bringing into question the sustainability of using only mHealth strategies.[35•] Additionally, many of these strategies were multicomponent strategies, so it may be difficult to decipher which component was most effective in contributing to the positive outcome. Despite the mixed findings, both studies reported high acceptability and feasibility of using mHealth tools in the target populations.[18]

The combination of mHealth tools with counseling by trained personnel has the potential to be more effective than either mHealth or counseling alone.[36] All three studies that used a combination of mHealth and support from trained personnel reported significant improvements in medication adherence and BP control as compared to a control group.[19, 20, 26•] This was particularly true when the mHealth tool was paired with support from health personnel like pharmacists and nurses.[19, 26] Several previous studies have reported the effectiveness of using trained health personnel to deliver mHealth-enabled care to sustain patient adherence to treatment.[36] Given the shortages of primary care physicians and variations in physicians’ practice behavior, augmenting patient care with mHealth tools could serve as an effective model for enhancing patient self-care and improving health outcomes, especially for racial/ethnic minority groups and underserved communities who are at disproportionate risk for HTN.[36-38] In this review, it is important to note that studies which used mHealth tools also reported improved BP Control and medication adherence outcomes by race/ethnicity.[24, 26•] For example, a pharmacist-led, technology-aided intervention demonstrated improvements in BP control and medication adherence among African Americans when compared to non-African Americans.[26]

There was only one study which utilized a quality improvement strategy for improving BP control and medication adherence. While the authors reported improvements in BP control among African American patients, this was not significantly different from baseline. There were also no improvements in medication adherence.

Implications

Findings from this scoping review have research and public health implications. First, while our findings showed that the use of mHealth tools and trained personnel resulted in improvements in medication adherence and BP, only a few studies examined outcomes by patient race/ethnicity. More research is needed to understand how these intervention strategies can mitigate racial disparities in HTN-related outcomes. This includes recruiting more racially/ethnically diverse patient samples as well as including a focus on comparative effectiveness methodologies to examine the relative impact of using trained personnel and technology-based interventions on medication adherence on BP control among minority populations.[22]

With respect to public health implications, many racial/ethnic minority populations reside in underserved communities, where 4 out of 5 people living in areas of concentrated poverty are either Black or Hispanic.[39] Coupled with a disproportionate burden for HTN in these populations, this can further exacerbate access to care, technological resources, disease management, and overall motivation to adopt a healthy lifestyle. As such, employing innovative, efficient, and accessible mHealth strategies delivered by trusted and knowledgeable community health personnel could improve health promotion practices, patient motivation, and self-care in these populations, which in turn could have an effect on BP control and medication adherence.

Limitations

There were several limitations to this review, despite the comprehensive search. First, our review was limited to articles published between 2017-2021. This limits a longitudinal examination of previous articles examining strategies to improve medication adherence and BP control. Second, for all included articles, medication adherence was measured and reported on different scales, which made reporting specific numerical outcomes difficult. To mitigate against this variety in reporting, we simply reported medication adherence as either improved or not improved (Note: if there was a significant difference, this was also reported). Third, uncontrolled HTN was defined at different thresholds, with some studies using a definition of BP>130/80 mm Hg, and others using BP>140/90 mm Hg. Due to this heterogeneity in BP and medication adherence measurements and reporting, it was difficult to make comparisons across studies.

Finally, while all populations included in the review were majorly racial minority groups, comparisons on the effectiveness of the interventions among races were limited. Only a few studies reported on differences in outcomes by race, which limited our reporting.[24, 26, 27] The ability to make comparisons by race might have provided more information on pathways for possible future tailored intervention strategies for these groups.

CONCLUSION

Overall, this review provides potential next steps for future research to examine the effectiveness of mHealth interventions in combination with support from trained personnel (i.e. CHWs, Health Educators & Coaches, Pharmacists, Nurses, etc.) and its effects on racial disparities in HTN outcomes. These preliminary findings show promise, but further efforts are needed to fully understand racial differences in improving blood pressure and medication adherence. Such efforts will advance scalable and sustainable interventions that are effective for BP control and medication adherence among racial minorities.

Table 3-.

Systolic and Diastolic Blood Pressure readings

Study Reference	BP Definition (mm Hg)	Baseline BP SBP/DBP (mm Hg)	SD	BP Follow-Up SBP/DBP (mm Hg)	SD	BP Change SBP/DBP (mm Hg)	p-value
Li et al, 2020	140/90	152.33 / 86.5	12.15/11.78	135 / 76.92	11.27/9.35	−17.33 / −9.58	0.005/0.005
Cene et al, 2017	140/90	AA: 140 SBP W: 138 SBP	Not Reported	AA: 134 SBP W: 131 SBP	Not reported	Overall: −6.0 SBP 12 months AA: −5.0 SBP 12 months W: −7.8 SBP Overall: −6.4 SBP 24 months AA: −6.0 SBP 24 months W: −7.2 SBP	p<.001 p<.001 p<.001 p<.001
Taber et al, 2018	140/90	50% had BP < 140/90	Not Reported	68% had BP< 140/90	Not Reported	+18% with BP<140/90	0.054
Chandler et al, 2019	140/90	SMASH: 152.3 / 86.8 ESC: 150.7/84.6	Not Reported	SMASH: 121.8 / 74.2 ESC: 145.7/79.4	Not Reported	SMASH: −30.5 / −12.6 ESC: −5.0/−5.2	p<0.001/p<0.001 Not Reported
Rashid et al, 2017	130/80	Chicago: 136.26 / 84.66 Houston: 139.8 / 77.3	Houston: 19.8/ 11.0 Chicago: 14.92 / 9.40	Chicago: 126.47/ 82.24 Houston: 137 / 77.1	Houston: 20.00/ 8.9 Chicago: 14.05/ 9.41	Chicago: −9.79/−2.42 Houston: −2.8 /−0.2	C: p<0.001/0.13 H: 0.86/0.39
Schoenthaler et al, 2020	140/90	Intervention: 141.1 /81.3 Control: 142.4/81.1	Intervention: 16.4/ 10.6 Control: 16.3/10.9	Intervention: 135.1 / 77.9 Control: 139.7/80.5	Intervention: 12.9 /9.8 Control: 13.3/12.2	Intervention: −6.0 /−3.4 Control: −2.7/−0.6	0.34/0.18
Schoenthaler et al, 2020	140/90	Intervention: 139.9 / 84.1 Control: 137.4/88.5	Intervention: 18.3 / 14.1 Control: 17.8/10.9	Intervention: 130.9 / 80.2 Control: 135.1/87.4	Intervention: 17.4/ 16.0 Control: 19.5/10.3	Intervention: −9.0 / −3.8 Control: −2.3/−1.1	0.10/0.10
Schroeder et al, 2020	140/90	IVR-T: 132.2 / 81.1 UC: 135.1/82	18.8/ 11.5 20.1/13.6	IVR-T: 132.1 / 82.1 UC: 137.5/83.5	17.2 /12.0 18.6/12.2	0.23 / +1.34 1.66/1.10	0.57 /0.88
Still et al, 2020	140/90	Intervention: 138.51 / 81.97 UC: 141.24/85.28	14.77 / 13.45 16.92/12.61	Intervention: 139.06 / 81.17 UC: 139.07/84.24	16.54/11.45 16.41/12.06	Intervention: 0.55 /−0.8 UC: −2.17/−1.04	0.87 / 0.67 0.55/0.63
Warren-Findlow et al, 2019	130/80	61.5% with SBP >130 / 69.2% with DBP>80	Not reported	Not reported	Not reported	Not reported	Not reported
Wheat et al, 2020	130/80	136/85.7	8.3/8.4	130.1/81.2	12.1/11.2	−5.9/ −4.5	0.006/0.008
Zha et al, 2020	140/90	mhealth- 145.77 / 90.84 SC: 145.67/90.30	5.10/ 7.58 3.68/6.46	mhealth- 137.38 / 88.08 SC: 140.88/88.10	4.86/ 7.45 5.01/9.41	−8.39 /−2.76 −4.79/−2.20	0.01/0.60 0.17/0.66

ABBREVIATIONS

HTN

hypertension

BP

Blood Pressure

SBP

Systolic Blood Pressure

DBP

Diastolic Blood Pressure

MA

Medication Adherence

IVR-T

interactive voice response and text message

mHealth

Mobile Health

SMASH

Smartphone Med Adherence Stops Hypertension

COACHMAN

community and technology-based intervention for hypertension self-management

CHW

Community Health Workers

DO

Deborah Onakomaiya

CC

Claire Cooper

AB

Aigna Barber

TR

Timothy Roberts

JG

Joyce Gyamfi

JZ

Jennifer Zanowiak

NI

Nadia Islam

GO

Gbenga Ogedegbe

AS

Antoinette Schoenthaler