Lay advisor interventions for hypertension outcomes: A Systematic Review, Meta-analysis and a RE-AIM evaluation

Introduction Lay advisor interventions improve hypertension outcomes; however, the added benefits and relevant factors for their widespread implementation into health systems are unknown. We performed a systematic review to: (1) summarize the benefits of adding lay advisors to interventions on hypertension outcomes, and (2) summarize factors associated with successful implementation in health systems using the Reach, Effectiveness, Adoption, Implementation, Maintenance (RE-AIM) framework. Methods We systematically searched several databases, including Ovid MEDLINE, CINAHL, PsycINFO from January 1981 to May 2023. All study designs of interventions delivered solely by lay advisors for adults with hypertension were eligible. If both arms received the lay advisor intervention, the study arm with lower intensity was assigned as the low-intensity intervention. Results We included 41 articles, of which 22 were RCTs, from 7,267 screened citations. Studies predominantly included socially disadvantaged populations. Meta-analysis (9 RCTs; n = 4,220) of eligible lay advisor interventions reporting outcomes showed improved systolic blood pressure (BP) [−3.72 mm Hg (CI –6.1 to −1.3; I2 88%)], and diastolic BP [−1.7 mm Hg (CI −1 to −0.9; I2 7%)] compared to control group. Pooled effect from six RCTs (n = 3,277) comparing high-intensity with low-intensity lay advisor interventions showed improved systolic BP of −3.6 mm Hg (CI –6.7 to −0.5; I2 82.7%) and improved diastolic BP of −2.1 mm Hg (CI –3.7 to −0.4; I2 70.9%) with high-intensity interventions. No significant difference in pooled odds of hypertension control was noted between lay advisor intervention and control groups, or between high-intensity and low-intensity intervention groups. Most studies used multicomponent interventions with no stepped care elements or reporting of efficacious components. Indicators of external validity (adoption, implementation, maintenance) were infrequently reported. Discussion Lay advisor interventions improve hypertension outcomes, with high intensity interventions having a greater impact. Further studies need to identify successful intervention and implementation factors of multicomponent interventions for stepped upscaling within healthcare system settings as well as factors used to help sustain interventions.


Introduction
Hypertension is the leading risk factor for heart disease, and 31.3% of adults worldwide have hypertension (1,2).It is estimated that only 13.8% of patients with hypertension globally achieve hypertension control (2).Traditional clinic-based care has not successfully improved hypertension control rates, which are worse in underserved communities (1).Community-based support improves outcomes in socially disadvantaged populations, especially when delivered by lay advisors who belong to the same social groups (3).Prior reviews of lay health advisors and community health workers (CHWs) have shown improved blood pressure and hypertension control (4)(5)(6).These reviews have been limited by including studies that evaluated lay advisor interventions with team-based care or additional health professional interventions and infrequent inclusion of broader community-based lay advisors such as barbers and faith-based lay advisors.Most health systems do not have the resources and staff to include multilevel interventions as reimbursement structure for team-based care is unclear, and it is difficult to know which level of intervention intensity can improve outcomes and in which contexts.The Community Preventive Services Task Force's (CPSTF) systematic review of CHW interventions for heart disease and stroke prevention reported an evidence gap in incremental effectiveness of CHW interventions (7).Therefore, there is a need to identify the sole benefit of adding lay advisors to improve their adoption into routine healthcare teams, assess their generalizability or external validity, and understand the level of intensity and context needed to have an impact on blood pressure.
Thus, we conducted a systematic review which aims to assess the additional benefit of lay advisor interventions (including varying intensity levels) on hypertension outcomes from a health system perspective.We defined lay advisor interventions as those provided by anyone who does not have a health professional degree, including CHWs, health coaches, hairdressers, and faithbased workers.We aim to summarize reported factors that may inform decisions on implementation choices in clinical settings using the Reach, Effectiveness, Adoption, Implementation, and Maintenance (RE-AIM) framework, which is useful for assessing internal and external validity and context of interventions (8).

Methods
The PRISMA statement was used to report the findings of this systematic review (9).

Search strategy and study selection
(See Supplementary material S1 for the detailed search strategy for the Ovid MEDLINE database)

Data sources
Librarians with expertise in screening citations for systematic reviews searched English language articles from 1981 through May 2023, using Ovid MEDLINE, Cochrane Central Register of Controlled Trials, Cochrane Database of Systematic Reviews, CINAHL, PsycINFO, Scopus, World Health Organization International Clinical Trials Registry Platform (WHO ICTRP), ClinicalTrials.gov,and Sociological Abstracts.We reviewed references in published reviews for any additional articles.Two reviewers (SJP and VB) independently screened citations and confirmed the final included studies.

Search terms
Groups of search terms included keywords for (1): lay lead, peer, community health worker, promotora, expert patient, barber, hairdresser, volunteer aide, faith-based, and (2); hypertension, high blood pressure, blood pressure.

Population and setting
Randomized and non-randomized studies published in English where the lay advisor intervention was evaluated as a sole additional intervention in adults with hypertension were included.As this review is designed for upscaling lay advisor interventions for hypertension care from a health system perspective, we did not include populationlevel screening studies that excluded adults with hypertension or population-level studies that did not report outcomes for the proportion of individuals diagnosed with hypertension.We excluded studies focused on pregnancy related hypertension disorders (preeclampsia, gestational hypertension).

Intervention
We defined lay advisor interventions as those including navigation, education, or support provided by anyone who does not have a health professional degree, as they typically belong to the same social groups as study participants (10).Common lay advisor interventions include promotoras, health coaches, peer supporters, faith-based workers, hairdressers, and community health workers.We excluded studies that included additional health professional intervention, including physician education or intervention components, as it is typically uncompensated time.We excluded studies of blood pressure screenings in the community or health insurance linkages where patients with hypertension were excluded or there was no follow-up information on the group of patients with confirmed clinical diagnosis of hypertension.If both arms received the lay advisor intervention, the study arm with lower intensity was assigned as the low-intensity intervention.We assigned low-intervention intensity when the lay advisors delivered a synchronous intervention targeting hypertension education or management.It was not considered an intervention if the lay advisors only checked BP or collected data.

Comparator
We included control groups where the only difference between the intervention and control group was the lay advisor delivered intervention.We included studies even if the control group received any form of low-intensity lay advisor interventions to provide insight into the incremental benefit of low-intensity versus high-intensity lay advisor interventions.Pre-post, process evaluations, and non-randomized studies were included.Studies that compared lay advisor interventions with active comparators such as health professionals or research staff were excluded.

Outcomes
For quantitative outcomes, the primary outcome was reduction in blood pressure (BP).We included change in systolic BP and diastolic BP as our joint primary outcome.Secondary outcome was the difference in the change in the proportion of patients with controlled hypertension from baseline to post intervention between intervention and control arms.If reported, we used the proportion of patients with BP <140/90 mmHg to define controlled hypertension if the study did not explicitly state the proportion of patients with controlled hypertension (2).For RE-AIM dimension outcomes, we looked at the characteristics and presence or absence of each RE-AIM dimension component.

Quality assessment
Two authors assessed study quality using the Cochrane Collaboration's risk of bias tool for RCTs (11).The primary author (S.J.P.) made final decisions where conflicts existed after reviewing all the articles independently.component as multicomponent intervention.For example, if an intervention included education sessions and recurring follow-up telephone calls, it was considered a multicomponent intervention.
Quantitative values and measures of statistical variation for BP and hypertension control rates were extracted from baseline and at the end of the study.When there were multiple study arms, we included quantitative values for the two arms, where the only difference was the lay advisor-led intervention or varying levels of lay advisor intervention.
Internal and external validity indicators using RE-AIM coding and scoring: A previously published tool was used to code eligible articles on the degree to which internal and external validity indicators of the RE-AIM framework were reported (12).We looked at protocols if referenced in the main articles.Supplementary Table S1 details how each dimension and component of RE-AIM was defined and measured.

Statistical analysis and data synthesis
A descriptive synthesis of the study setting, participants, lay advisors, intervention components, and control group was performed and reported as a study description table.Proportions of total, RCTs, and nonRCT studies reporting each of the RE-AIM dimensions and components are reported as a table.Quantitative synthesis: If we had three or more eligible studies of added lay advisor intervention or varying levels of lay advisor interventions, the primary author (SJP.)performed the statistical analysis using Comprehensive Meta-analysis Software version 3 (Biostat Inc., Englewood, NJ).We adjusted sample sizes for cluster RCTs using the documented intra-cluster coefficient (ICC) (13).We used the random-effects model to compute conservative effect sizes incorporating both within-study and between-study variations.We calculated the difference in means with 95% confidence intervals, and we considered a p-value of <0.05 statistically significant for all analyses other than the Q statistic.A correlation coefficient of 0.5 was assumed between initial and final values.Heterogeneity among studies was evaluated using the Q statistic, with a p-value <0.10 indicating heterogeneity, and using I 2 statistics (I 2 values <40% may indicate less substantial heterogeneity and 75-100% indicates substantial heterogeneity) (14).If substantial heterogeneity existed, we planned a priori meta-regression if we had 20 or more studies or subgroup analysis if we had <10 studies.We identified the following study characteristics that may explain between-study variability: presence or absence of intention to treat analysis; presence or absence of home visits; settings in developed or developing countries; lay advisor training duration; study duration; and intervention components of group education, individualized intervention, or combined intervention.Publication bias was assessed with funnel plots and the Egger regression test (15).We conducted sensitivity analysis by removing one study at a time.

Results
Of 7,267 unique citations, 41 studies were eligible for inclusion in our review.See PRISMA Flow Diagram.(Figure 1) All study characteristics are shown in Table 1.
Most studies mentioned lay advisors were matched demographically with study participants.The least intense interventions were outreach phone calls to promote access to care (29).Most intense intervention included monthly group education with home visits every other month with follow-up biweekly phone calls (35).Other than two studies, all lay advisor interventions were multicomponent.Seven studies compared low-intensity interventions with high-intensity lay advisor interventions.No studies specifically compared stratified or stepped care models of modifying lay advisor intervention intensity based on patient characteristics or hypertension control state with usual care.

Study quality
Supplementary Table S2 summarizes the assessment of risk of bias for individual randomized studies.All studies had at least one domain judged as unclear risk of bias and 18 studies had at least one domain, mainly blinding or intention to treat, regarded as high risk of bias.No studies had all domains regarded as low risk of bias.Dropout rates varied from 0 to 31% and 6 RCTs had dropout rates of >20% with higher dropouts from intervention groups.

Outcomes
Of RCTs where control groups did not receive any lay advisor interventions, nine reported systolic BP outcomes, eight reported diastolic BP outcomes and six reported hypertension control outcomes at baseline and end of study.See Table 2 for improvement in BP and hypertension control noted in all included RCTs and Table 3 for improvement in BP and hypertension control in included non-RCTs.

Effect on blood pressure outcomes
The overall pooled effect of lay advisor interventions from nine RCTs (n = 4,220 participants) showed a mean improvement in systolic BP of −3.7 mmHg (CI -6.1 to -1.3; p 0.002, I 2 88%).(Figure 2) A sensitivity analysis where each study was removed had no significant impact on the results.(Supplementary Figure S1: Forest plot with each study removed) The pooled effect from eight RCTs (n = 3,056) of lay advisor interventions which measured diastolic BP showed an improvement of −1.8 mmHg (CI -2.5 to -1.0; p < 0.001, I 2 7%).(Figure 3) A sensitivity analysis where each study was removed had no significant impact on the results.(See Supplementary Figure S2).

Effect on hypertension control outcomes
Meta-analysis of six RCTs (n = 3,762) showed a pooled odds ratio of 1.2 (CI 0.75 to 2.0; p = 0.        with lay advisor interventions compared to the control group.(See Figure 4).Subgroup analyses of studies grouped by presence of intention to treat did not show any significant differences in BP between groups.(Supplementary Figures S3, S4) There were not enough studies to conduct subgroup analyses of studies grouped by developed or developing country setting, and mode of interventions.The funnel plot and Eggers regression test (p = 0.008) indicate publication bias for systolic BP outcomes but not for diastolic BP or hypertension control outcomes.(Supplementary Figures S5, S6).

Effect on systolic blood pressure, diastolic blood pressure, and hypertension control with high-intensity compared to low-intensity lay advisor interventions
Seven RCTs from the United States compared low-intensity interventions with high-intensity interventions, of which six reported BP outcomes.Pooled effect from these six RCTs (n = 2,644) showed a mean improvement in systolic BP of −3.6 mmHg (CI -6.7 to −0.46; p 0.02, I 2 82.7%) and in diastolic BP of −2.1 mmHg (CI -3.7 to −0.4; p 0.01, I 2 70.9%) in high-intensity lay advisor interventions compared to low-intensity interventions (See Figures 5, 6).The funnel plot and Eggers regression test (p = 0.4) did not indicate publication bias for these pooled BP outcomes.A sensitivity analysis where each study was removed showed reduced significance of results.(See Supplementary Figures S7, S8) Meta-analysis of seven RCTs (n = 3,277) showed a pooled odds ratio of 1.29 (CI 0.79 to 2.1; p = 0.3, I 2 90.79%) for controlled hypertension with high-intensity lay advisor interventions compared to the low-intensity lay advisor intervention group (Supplementary Figure S10).There were not enough studies in groups to conduct subgroup analyses for high intensity compared to low intensity interventions.

RE-AIM criteria reporting In studies
There was no significant difference in the frequency of reporting of RE-AIM components between randomized and non-randomized studies other than qualitative assessments of efficacy, which were more frequently reported in nonRCTs.See Table 4 and Supplementary Table S1.One study specifically reported study results in RE-AIM format (45) and one recent study specifically reported reach and adoption of peer coaching intervention in primary care practices (33, 57).

Reach
Six of the nine studies with sample sizes >1000 were conducted in developing countries (18-20, 23, 25, 34) and three were done in the United States (27,31,45).Participation rate varied from 2 to 98% with clinic-based recruitment showing higher participation rates compared to population-level recruitment.When reported, nonparticipating individuals had higher systolic BP or a lower proportion of their BP controlled at baseline, but this information was not reported in most studies (18,29).Clinic-based recruitment showed higher participation rates as fewer patients needed to be approached in clinics (denominators were lower) with higher recruitment success (21).

Efficacy/effectiveness reporting
Eight non-randomized studies reported qualitative assessments to understand outcomes (39,41,42,(50)(51)(52)(53)56); one For multi-arm studies, we limited the sample size intervention, and control group descriptions to groups where the only difference between groups was either a lay advisor intervention or a low-intensity versus a high intensity lay advisor intervention.c original article for BP improvement included physician education component so was excluded but the current included article focused on lifestyle modifications supported by only CHWs with text messages.BP and hypertension control outcomes were not included in the review for the original article due to physician education component for medication management.d Report of two RCTs with comorbid DM and HTN, included study information where CHW intervention was tailored for diabetes as the RCT with hypertension is separately included.of these was mainly a process evaluation of a sustained peer leader program for Veterans (39).One pragmatic randomized study included process evaluation in their protocol, but published article mentions challenges with balancing external validity and intervention assessments (19,58).Few studies reported reasons for lay advisor withdrawals, which included personal reasons of health issues or relocation as well as an inability to perform certain required intervention tasks such as properly reading BP measurements (25, 39).

Adoption
Method to identify the target delivery agent was reported mainly as the selection and nomination of volunteers with matching sociodemographic characteristics to participants.Two studies reported adding activities to a pre-existing program, but repeated visitor rate (43%) was only reported in one study (39,44).One author of a randomized study shared a follow-up process evaluation using mixed methods assessment with surveys and focus group discussions (18,59).

Implementation
Four studies mention compensation for lay advisors reflecting the pay scale of the respective countries otherwise studies reported lay advisor compensation for completing study activities (18,25,28,32).One study reported CHWs worked 40 to 60 h per month to care for 120 participants (28).

Discussion
We contribute to the literature by reporting a systematic review that evaluates the additional benefit of lay advisor interventions for hypertension outcomes where the lay advisor interventions are the sole additional intervention.We limited contamination by

Reach
Characteristics or contexts that interact with an individual's willingness to participate may influence the potential of these interventions to improve health disparities, as most studies included socially disadvantaged populations.

Effectiveness
Studies have rarely reported reasons for improvement in outcomes and characteristics of participants who may not benefit from these interventions or continue to have unmet needs.Assessments of multicomponent interventions to identify the least and most efficacious individual components are missing and may help tailor upscaling.

Adoption
It is unclear what characteristics or contextual factors would encourage the uptake of a lay advisor role by individuals not already engaged in community-level leadership.Settings for lay advisor interventions were mostly predetermined with outside funding; hence, the characteristics of settings that otherwise may or may not participate are unclear.

Implementation and maintenance
The time required for interventions' key components and supervision needs to be quantified from the individual and organizational perspectives.Costs from a societal perspective or grant-funded compensation are frequently reported but may not be helpful for health systems with limited resources or budget margins.

Limitations
Our review has several limitations due to the way studies report information and limitations of meta-analyses.We strictly limited our review to studies with lay advisor-delivered intervention without additional health professional or research staff-delivered components.We did not want to combine two interventions with unclear reimbursement structures, and lay advisors are generally not yet part of core healthcare teams; however, lay advisors alone can provide support.Specifically, our exclusion of any physician education or training component limited a few key studies (62)(63)(64)(65).We excluded these studies because physician-directed interventions may individually improve outcomes, and hypertension is routinely managed in time-restricted primary care clinic visits along with multiple concerns and health maintenance (66,67).Secondly, as health systems may or may not be involved in community-level screenings but are typically held accountable for hypertension outcomes in their patient populations, we limited community-level screening studies to those reporting hypertension outcomes.Third, high heterogeneity was noted with diverse intervention components and intensity variations but planned meta-regression and subgroup analysis to explain the variation could not be done due to limited number of eligible studies; nevertheless, the increasing dose-response gradient with increasing intervention intensity supports directionality of the intervention effects.Literature syntheses can make sense of this heterogeneity if studies also report contextual factors affecting individual intervention component acceptability and efficacy.

Implications for future research
Our review has strong implications for future research.Reporting of most and least efficacious components of multicomponent interventions to tailor stepped upscaling of lay advisor interventions is needed.Studies mainly included adults representing the working population's age where stepped-care models may be important to reducing intervention burden and balancing healthcare resource allocations.Tailoring lay advisor services within health systems that serve diverse patient populations has been understudied as most studies targeted socially disadvantaged population groups.Pragmatic trial designs such as hybrid effectiveness implementation trials may be helpful to evaluate not only how the intervention works but also how to successfully implement the interventions in diverse settings.Qualitative assessments of why and how the lay advisor interventions reach the targeted population, improve outcomes, and can be maintained are areas for future research.Future mixed methods assessments need to contribute to understanding the facilitators and barriers to engaging patients in the interventions, retaining a lay advisor workforce, and sustainability of the intervention at an individual and organizational level.

Conclusion
Add-on and high intensity lay advisor interventions may improve blood pressure outcomes in socially disadvantaged populations, but studied interventions are heterogeneous.Future studies need to identify the intervention's most efficacious components and include assessments of stepped upscaling.Future research should focus on mixed methods assessments to identify explanatory processes for effectiveness and engagement at the individual, lay advisor, and setting levels to inform the real-world implementation of these interventions.
Group education: Nine-week program with educational modules in Spanish in 2-h sessions in weeks 1, 2, 3, and 8 for groups of 15-20.Modules focused on patient education, lifestyle modifications, and overcoming barriers, and a hypertension module with photonovela.Phone calls: Yes, during week 4 to 7 phone calls done to answer questions, schedule makeup classes, and discuss lifestyle changes discussed during educational modules Community basedmin home health education visit session with all members of household, reinforcement visits of 30 recruited from 15 veteran service organization posts.Initial 8-h training session followed by monthly/bimonthly 2-h meetings to introduce new health materials and review project activities.Topics covered hypertension, selfmanagement and peer education/ motivation.63 y/o (peer leaders) 78% peer leaders were males Monthly group education: Educational materials, health scripts, BP check stations, and promotion of self-monitoring at routine veteran's service organization meetings.Community health workers recruited from members of senior center.Selection based on enthusiasm and availability.Trained for 12 h by health educators in BP measurement, record keeping, communication to participants.Sessions every other week include checking and recording BP on program and participant record cards followed by information on BP status and recommended action steps Community based CHWs recruited from community based on written test scores and 2 rounds of interviews.Trained for 7 days, 3 h/day (total 21 h) in 1-2-week blocks followed by 5 h supervised field work.Refresher training midway through studyCommunity based: Yes Home visits: yes, 1 h, every 2 months.Behavior change strategy focused on lifestyle, improving healthcare seeking, and addressing barriers to medication adherence.Communication in native language of patient using pictorial information.Community health workers: training and selection not reported.Two interventions (1): home blood pressure support (n = 20) and (2) community health worker support (n = 20).Community based: Yes In (2) community health worker support: Group education: yes, 4 education sessions over 4 weeks.Home visits: yes, 8 visits over 4 weeks for tailored counseling on health behaviorsCHWs recruited from the primary care staff of public care centers.From the community, with similar ethnicity, language, SES, and life experiences.Trained for 2 days in motivational interviewing techniques, measuring BP, behavior change, adherence counseling, lifestyle modifications.Training followed by onsite field testing and certification.CHW and text messaging intervention:Weekly, from web-based platform to participants and family members using a one-way outgoing system.Individualized to promote lifestyle changes and reminders for medication adherence.Community based: Yes Home visits: Yes, monthly for the first 6 months and every other month thereafter.Initial visit lasting 90 min with subsequent visits of 60 min.Family-based intervention to discuss general HTN knowledge at first visit.At subsequent visit, tailored counseling to participants and their families on adherence, home BP monitoring, and lifestyle modifications.Focus on goal setting, problem solving, social support, and motivation.
recruited a team of Faith-based organization (FBO) volunteers.Participated in 2-day train the trainer program Volunteers advertised programs.Worked through co-located senior center, every 2 weeks to quarterly, conducted free blood pressure screening, lifestyle counseling, weight management, coaching for clinical encounters.local community and advocate for local needs.Trained to collect information from individuals on demographics, health habits, BMI, BP.Monthly training sessions throughout to ensure consistency and time for troubleshooting.Community-wide campaign in 10 municipalities.CHWs lead the culturally tailored efforts of social support, risk factor screening, exercise groups, healthy cooking classes.CHWs and Firefighters: CHWs recruited from preexisting Community Health Center-associated individuals.1-week CHW training Firefighters leveraged due to integration into the community via community fire check safety program with training in implementation of the intervention package.CHWs and firefighters supported by tablets for data management and decision support, identified participants with poorly controlled hypertension, recommending evidence-based management strategy, supported lifestyle counseling Group education: no Individual education: yes.Each participant had a tailored intervention designed for their barriers, based on the 3 components above.Four total visits over 6 months.Online: No, but mobile health technology leveraged by CHWs.Study participants recruited from selected religious centers.Role-model patients selected from enrolled patients.Must also have secondary education.Trained for 1 day in HTN, BP measurements, signs of uncontrolled HTN.CHWs helping with implementation trained for 3 days.Peer-support adherence clubs led by role-model patients to motivate and facilitate medication adherence, BP monitoring.Once a month for 6 months in local community center.Group education: Yes, small groups (10-15) of patients with a role-model patient as facilitator Community based: yes Home visits: no None (compared to baseline) coaches provided support to patients as they made lifestyle changes to improve their blood pressure.This included goal planning, connection to community resources etc. Peer coaches received training and certification by staff members of the Southeastern Collaboration to Improve Blood Pressure Control (SEC).The coaches were members of the community just like the patients in the study.They also had chronic medical conditions but were not health professions and were not required to provide medical advice.They worked with patients for 12 months.Assist engagement in hypertension self-management (including dietary changes and physical activity), to carry out the recommendations of the healthcare team (including taking medications and keeping appointments), to provide emotional support, and to link patients to the practice for care.Group education: No Phone calls: Yes, a one-on-one telephone delivered structured program intensive intervention phase of 8 weekly topic-focused sessions followed by monthly check-ins over the 12-month intervention period.Longer booster sessions were offered if BP control slipped after the intensive intervention phase.existing BAME barber network in south London, UK were recruited from 5 barbershops.Barbers were educated on offering BP checks to clients and providing education when needed.Trained online (1.5 h) and face-to-face ( 4, I 2 85.8%) for controlled hypertension

TABLE 1 (
Continued) Health coaches recruited from health center employees and volunteers with bachelor's degrees.Trained for 16-20 h HTN, HTN medications, lifestyle modifications, adherence counseling.High Intensity intervention of home BP monitoring, weekly health coaching, and home titration of BP medications Phone cells: weekly, to both study arms.Discuss well-being, adherence, BP values.Participants who reported elevated BP and adherence to medications could increase dose according to predetermined algorithm.Health worker advised physician to fax prescription and updated EHR.Community based: no Home visits: no Low intervention: Home BP monitoring and weekly health coaching (no home titration).Lay Hispanic community health workers (promotora) from surrounding communities.Trained for 2 days using the National Heart, Lung, and Blood Institute's promotora curriculum, Salud Para Su Corazon (SPSC).Curriculum consists of educational materials designed to address CVD risk in the Hispanic community in the US.Group education: weekly session with 10-15 participants, 9-week curriculum.Plus, additional evidence-based mental health session and managing medicines session.Community based: Yes Home visits: Yes, reviewed and emphasized group education concepts.

TABLE 1 (
Continued) immigrants fluent in English and native languages.All had at minimum a bachelor's degree and most lived in the same community as participants.Trained for 60 h in core competencies, HTN, lifestyle modifications.Group education: yes, 4, 1x/monthly 90 min sessions consisting of health education using adult learning techniques and other culturally appropriate games.Individual visits between group sessions, 1x/month.Convenient locations including home, employer, community setting.Develop individual goals, remove barriers to access, promote adherence, provide referrals (including mental health or tobacco cessation).Phone: yes, as needed, to follow up with participants.Community based: yesHome visits: yes, as needed for individual visits.

TABLE 1 (
Continued) Educational intervention based on the Health Belief Model delivered by CHW as part of HIV clinic appointments.Group education: No Phone calls: Yes (three in person sessions and two phone sessions).Remind patients of next appointment and to foster therapeutic relationship with CHW.Online: No Community-based: Yes Home visits: Yes.First, third and fifth session were at patient's home but patient could chose alternative location.

TABLE 1 (
Continued)If data on population/setting such as age or urban/rural location was not reported, we did not include the missing components in the study rows.c a Only sample size of patients with hypertension is mentioned.b

TABLE 2
Improvement in BP in RCTs a .

TABLE 2 (
Continued) Original article for BP improvement included physician education component so was excluded but the current included article focused on lifestyle modifications supported by only CHWs with text messages.
a p-values are not reported if they were not reported in the original article.b

TABLE 3
Improvement in BP in non-RCTs a,b .
a p-values are not reported if they were not reported in the original article.b Other than Hovell, 1984 and Reninger, 2021, all non RCTs were pre-to post comparisons.c Only included results for participants diagnosed with hypertension additional requirements.PRISMA checklist attached as a separate Supplementary Table.

TABLE 4
Publications reporting on RE-AIM elements.