Hypertension (HTN) has become the largest driver of morbidity and mortality (M&M) worldwide, affecting nearly 1 billion persons, the vast majority living in low- and middle-income countries (LMICs). While clinical research has identified highly-efficacious and inexpensive options to control HTN, rates of awareness, treatment and control of HTN are abysmally low. Implementation strategies that can effectively reach and engage patient populations while feasible within the constraints of frail health systems are urgently needed. In Peru, less than 10% of those with HTN are optimally controlled and thus avoidable M&M continues to increase at unacceptable levels. In the proposed ANDES strategy, ANDES research team will study a vulnerable, impoverished indigenous Andean population that has a high prevalence of HTN (18.5%) and Type 2 Diabetes (T2D) (7.4%); living at high altitude. The study has two phases: UG3/UH3. During the formative UG3 phase (ended 12/31/22) we completed 3 aims, including: a) adapting community-based service delivery models for detection/control of HTN/T2D using stakeholder-engaged approaches; and b) assessed short-term implementation and service outcomes of pilot Health Fairs to detect HTN/T2D and a CHW-led model for engagement and treatment of HTN/T2D in two communities. A number of stakeholder in-depth interviews revealed significant limitations to HTN control. Discrete Choice Experiments (DCEs) surveys helped understand attributes for a community health worker (CHW)-led HTN care program that patients would value. Human-Centered Design was used to refine CHWs manual prototype, frequency and content of training sessions on patients, training CHW, supervision and integration into the healthcare system; facility-based health worker training content and approaches; identified opportunities for CHWs to improve health care system by helping to promote hypertension diagnosis, treatment and control. All these activities were used to refine the protocol that was tested in the pilot study where 1,079 subjects underwent blood pressure monitoring under a Standard Operating Procedure using electronic tablets to construct a REDCap database.

The UH3 phase will test the adapted ANDES strategy. ANDES is a two-arm, individually randomized superiority trial with parallel assignment. The overall objective of ANDES is to evaluate the effectiveness of a home-based Health Agent-led intervention in reducing blood pressure and improving diabetes control in patients from under-resourced communities in Puno, Peru. Potential ANDES study participants will be identified, screened, and recruited via 2 different mechanisms: 1) community health fairs and 2) healthcare facility-based enrollment. Health fair planning, will be conducted in collaboration with DIRESA & EsSalud and local healthcare facility leaders. All health fair attendees will be screened for blood pressure measurements. ANDES research study staff will do screening at tents in the health facilities after doctors refer patients (or for any patients to visit who want) to inquire about their willingness for an in-person screening visit at their home. Recruitment will be rolling over 24 months during which 1068 participants in total (534 participants per arm) will be enrolled. To detect a difference in systolic blood pressure of 2.5 mmHg between intervention and control arms with 95% confidence, 90% power and a standard deviation of 12 mmHg, ANDES research team would need 485 participants per arm. Assuming a conservative lost-to-follow-up of 10% at 12 months, ANDES research team would need to randomize 1068 participants in total (534 participants per arm). The primary outcome is to determine if the ANDES intervention impacts SBP in comparison to the usual care arm 12 months after randomization. ANDES participants will initiate one of two care pathways according to the randomization assignment. Participants in the usual care group will be referred to their local healthcare facility for evaluation and/or to receive medical therapy per typical standard of care and at the discretion of the treating physician for the entirety of the ANDES study. Participants in the ANDES intervention group will receive the ANDES implementation package consisting of health agent home visits for 12 consecutive months. Implementation package delivery will end after month 12, at which time intervention participants will be advised to obtain standard care from their local healthcare facility. Study outcome measures will be obtained in both the usual care and ANDES intervention group at months 12 and 18 post-randomization. All participants will receive visits at baseline, 12 and 18 months for study outcome assessments. Intervention participants will additionally receive a total of 12 visits by health agents as part of the intervention package, for a total of 18 visits for those assigned to the intervention group.

INTRODUCTION

Study Rationale

Hypertension (HTN) is the major contributor to the global burden of disease and is the leading cause of morbidity and mortality in Peru. Completion of the proposed randomized trial will provide compelling and robust evidence on the acceptability, appropriateness, fidelity, and sustainability of the multicomponent ANDES intervention that is aimed at reducing blood pressure and improving diabetes control in patients from high-risk and under-resourced communities in Puno, Peru. These data are urgently needed to inform policies in low- and middle-income countries on the health gains that can be achieved by multicomponent implementation strategies to reduce hypertension.

Background

HTN affects ~1.4 billion people worldwide, yet in low-and middle-income countries (LMICs) the vast majority of patients are unaware and untreated, often until the occurrence of devastating complications (e.g., myocardial infarction, stroke). HTN is the leading cause of morbidity and mortality in Peru. Overall age- and sex-standardized prevalence of pre-HTN and HTN among Peruvian adults is 24% and 19%, respectively (1). Rates of premature cardiovascular disease (CVD) and stroke mortality are 87 and 27 per 100,000 Peruvians (aged 30-69 yrs), respectively, with greater prevalence in men (2). Pre-HTN and HTN are higher in rural- vs. urban-dwelling Puno residents (44 vs 31%, mean age 55 years) (3). The PERU MIGRANT study found HTN awareness, treatment, and control rates of 48%, 40%, and 30%, respectively; but only 5% of HTN being treated and controlled (4-6). The Puno CRONICAS study cohort found HTN control rates of 19% and 1% among urban vs. rural dwellers, respectively (7). Thus, very low HTN control rates in Peru represent an important health care gap.

In Peru, indigenous Andean communities (i.e, Aymara and Quechua) represent a population with a high prevalence of HTN and Type 2 Diabetes (T2D) (18.5% and 7.4% respectively for those ≥55 years old); yet access to chronic care is virtually non-existent despite the availability of efficacious and affordable treatment. To address this gap, ANDES research team propose to develop and test the Addressing HyperteNsion and Diabetes through Community-Engaged Systems in Puno, Peru (ANDES) strategy for HTN/T2D diagnosis, and HTN treatment and control, which will be designed for remote settings where health care has low penetration.

Trial Design & Objectives

ANDES is a two-arm, individually randomized superiority trial with parallel assignment. The overall objective of ANDES is to evaluate the effectiveness of a home-based Health Agent-led intervention in reducing blood pressure and improving diabetes control in patients from under-resourced communities in Puno, Peru. Health agents are community health workers (CHWs) and/or nurse technicians that act as liaisons between the public health system and community members.

METHODS: PARTICIPANTS, INTERVENTIONS, OUTCOMES

This study will be conducted in Puno, Peru, a remote region in the Andean 'high plains' that suffers a high burden of HTN/T2D and where access to health care is limited. Puno is the capital of the Puno Region and the Puno Province (2017 census pop: 1,172,697) located in southeastern Peru, at 3,825 meters (13,500 feet) above sea level. The population is primarily indigenous, consisting of Quechua (pop: 537,972) and Aymara (pop: 318,363) Andean people, which together comprise ~72% of the population. Puno city, the urban provincial capital (150,000 residents) borders the shores of Lake Titicaca and northern Bolivia. Rural Puno comprises 14 large districts with each district having up to 20 communities. The public health system in Puno is managed by the Regional Health Administration (DIRESA, by its acronym in spanish), under the Peruvian Ministry of Health (MINSA, by its acronym in spanish), the governing body that leads, regulates, and promotes national health policy.

ANDES Implementation Strategy

The "ANDES multicomponent implementation strategy" is an adaptation of the World Health Organization (WHO) HEARTS package (26) and consists of the following components:

- Health agent-managed home care

• BP monitoring
• Health coaching
• Health system navigation
• Medication adherence support

Specifically, the ANDES strategy is aligned with four WHO HEARTS technical package components.

HEALTH AGENT-managed home care As part of the ANDES implementation strategy, ANDES intervention participants will be provided BP monitoring and health coaching by health agents in their home. Additionally, all participants will be assessed for T2D by having their HbA1c measured during the study. All participants deemed to have T2D will be referred to the Health Care System for further evaluation. Health coaching is defined as "multiple sessions for patient-centered health education and motivation delivered with the goal of facilitating lifestyle modification and/or medication adherence." (27) See section "Endpoint Assessment" for a description of BP monitoring. (Note: all participants will also have their BP monitored by a separate team of research fieldworkers at baseline,12-, and 18-month follow-up visits. See Section "Study Schematic & Timeline" for a detailed schedule of activities.)

During monthly home visits over the course of one year, community health agents will meet one-on-one with intervention participants, during which the community health agent will complete the following activities:

1. Review personal goals and visit agenda;

2. Conduct health assessments:

   • BP

3. Health Coaching:

   1. Assess the participant's HTN health plan and collect medication adherence data and health services access data

   2. Provide health education and counseling:

      • Provide educational materials, medical appointment log, weekly pill organizer and calendar
      • Review behavioral strategies to lower BP (e.g., diet and particularly salt consumption, physical activity, weight management, stress management, sleep, and alcohol use (28)
      • Review barriers to improving BP as well as strategies to overcome them.

4. Provide healthcare system navigation support:

   • Work with patients to set up, prepare for, and attend doctor's appointments.
   • Empower the patient to advocate for their own health care and emphasize their roles and responsibilities as a member of their healthcare team.

5. Provide advice and support for patients to obtain medications

A complete visit will be determined by the ability of a health agent to cover appropriate issues from the list above. The ANDES research team have developed a checklist as a guide for the health agents to follow. This guide will also be used to evaluate fidelity to the multicomponent strategy.

Text messages As part of the multi-component strategy developed by ANDES for HTN detection and management, a text messaging system was developed directed to health agents. Text messages provide reminders, education, and orientation to facilitate health agents to encourage healthy behaviors in their patients, thus improving chronic disease management and health outcomes (29).

The ANDES texting component is directed to health agents. Messages have been tailored to fit with the characteristics of the population, taking into consideration local foods, belief systems, and traditions. The main objective of messages directed to health agents is to communicate weekly reminders regarding topics of discussion with their patients. Messages will also motivate and provide positive reinforcement to health agents for their dedication to the patients and support of the intervention.

Recruitment

Potential ANDES study participants will be identified, screened, and recruited via 2 different mechanisms: 1) community health fairs and 2) healthcare facility-based enrollment.

Health Fairs. Health fairs are 1-day events offering HTN screening in Puno, Peru. Health fair planning, including the selection of the date, times, and location of health fairs will be conducted in collaboration with DIRESA & EsSalud and local healthcare facility leaders. Outreach campaigns and advertisements (e.g., local radio announcements, posters, fliers) will encourage all adult community members (18+ years) to attend health fair events. All health fair attendees will be asked to complete a Screening Case Report Form (CRF), which includes blood pressure measurements. (See below for Screening CRF details.) If the person has high blood pressure (but does not have a Hypertension diagnosis) the ANDES fieldworkers will ask them to sign a pre-screening informed consent to be able to collect basic personal information (address and phone number) to be able to contact them after they receive a diagnosis.

Healthcare facility-based enrollment. MINSA and EsSalud health personnel (physicians, nurses, and nurse technicians or social workers) perform routine triage in the health centers during routine care. In specific cases, when the patient cannot access or reach the health facility due to illness, mobility issues or other health conditions, health personnel perform home visits. Since participation in the ANDES study is voluntary, if a patient does not wish to participate, there will be no consequences on their access to routine care at the health facilities.

Selected healthcare facilities for the ANDES study will be asked to contact patients with HTN- via phone, home visits or during their routine healthcare assessments. Healthcare staff (physicians, nurses, and nurse technicians) will inquire if a patient with HTN is willing to participate in the ANDES study. If the potential participants show interest in the ANDES study, two things can happen:

1. The patient will be invited by the health personnel to the health center on a specific date for screening (inclusion criteria) and subsequent enrolment to the ANDES study by the ANDES fieldworkers, or
2. If the patient expresses interest but cannot access the health care center (for mobility issues or illness), the ANDES team will go to the participant's home for enrolment.

During the enrolment process, the study staff will administer the Screening CRF and determine if the patient fulfils the inclusion criteria. After the participant is enrolled, the field staff will set a date to carry out all baseline measures and health assessments.

The Screening CRF is used to determine the eligibility of participants captured by health fairs and facility-based enrollment. The Screening CRF confirms each inclusion and exclusion criteria via the following methods:

1. Participant age is verified by government document. Participants must be at least 18 years old.

2. The following criteria are verified by participant report: place of residence within the study area, no plans to move during study period, willingness and availability to receive monthly CHW visits for one year, not currently or planning to become pregnant, no other member of their household already a participant in ANDES, not currently receiving home medical visits and not planning to travel outside of the study area for more than 3 months in the next year; also, patients not on dialysis or expecting to start dialysis within the next 12 months, patients with liver failure (meeting Child-Pugh B or C criteria) and no bedridden patients.

3. Ambulatory status is based on patient report and observation by ANDES study staff.

4. Cognitive capability is based on ANDES staff evaluation of responses to a standard set of questions.

5. HTN status is based on patient report and the following criteria:

   • If the participant reports a HTN diagnosis and DOES have a current hypertension medication prescription AND the participant's blood pressure ≥ 130/80 mmHg (measured by ANDES study staff), the participant will be move directly to the Informed Consent Process if interested in the study.

     i. If the participant's blood pressure is < 130/80 mmHg, the participant will not proceed to the Informed Consent Process. The participant will be thanked for participant's time and be offered counseling on hypertension and health habits.

   • If the participant reports a HTN diagnosis but DOES NOT have a current hypertension medication prescription, one of the following scenarios will occur:

     i. If the participant blood pressure is ≥ 140/90 mmHg (measured by study staff), the participant will be referred to a healthcare facility to be re-evaluated. If the participant are diagnosed with hypertension and/or provided a hypertension medication prescription, the participant will then proceed to the Informed Consent Process.

ii. If the participant's blood pressure is < 140/90 mmHg, the participant will not proceed to the Informed Consent Process. The participant will be thanked for participant's time and be offered counseling on hypertension and health habits

- If the participant has a blood pressure ≥ 140/90 mmHg and DOES NOT report a past diagnosis or current prescription, patient will be referred to a healthcare facility to be evaluated. If patient receive a diagnosis of HTN and/or are provided a prescription, patient will then proceed to the Informed Consent Process.

Recruitment will be rolling over 24 months during which 1068 participants in total (534 participants per arm) will be enrolled. See Sample Size Consideration for details. Informed consent will be obtained from all study participants using standard procedures.

ANDES is a randomized, two-arm intervention individually randomized trial with parallel assignment. ANDES research team will use rolling enrollment with 45 participants per month who meet inclusion/exclusion criteria. All participants will be followed longitudinally for 18 months. To maintain balance of treatment assignments, the sample will be stratified by MINSA/EsSalud health systems.

METHODS: DATA COLLECTION AND MANAGEMENT Study Schematic & Timeline

Following enrollment and a baseline assessment (Visit 0), ANDES participants will initiate one of two care pathways according to the randomization assignment. ANDES participants in the usual care group will be referred to the participant's local healthcare facility for evaluation and/or to receive medical therapy per typical standard of care and at the discretion of the treating physician for the entirety of the ANDES study. ANDES participants in the ANDES intervention group will receive the ANDES implementation package consisting of health agent home visits for 12 consecutive months. Implementation package delivery will end after month 12, at which time intervention participants will be advised to obtain standard care from their local healthcare facility. Study outcome measures will be obtained in both the usual care and ANDES intervention group at months 12 and 18 post-randomization. All participants will receive visits for study outcome assessments at baseline, 12, and 18 months. Intervention participants will additionally receive a total of 12 visits by health agents as part of the intervention package, for a total of 15 visits for those assigned to the intervention group.

No financial incentives will be offered; participants can withdraw from the study at any time without loss of healthcare benefits patients would otherwise be eligible to receive (i.e., usual care). Nonetheless, based on ANDES research team experience with prior studies in Peru, and specifically in the Puno area, ANDES research team anticipate limited attrition (≤ 10%).

ENDPOINT ASSESSMENT Health Screening

The following health assessments will be conducted by fieldworkers hired and trained by the ANDES research team.

Blood pressure measurements: A member of ANDES study staff will measure resting blood pressure in the right arm in triplicate (with at least 1 minute between measurements) using an automatic oscillometric sphygmomanometer (Omron, model HEM-907XL) at the participant's home using procedures detailed in the "Blood Pressure Measurement - Standard Operating Procedures" document. Briefly, the participant will refrain from exercise, consumption/use of tobacco products, alcohol or caffeinated beverages, or cooking using biomass in the 30 minutes prior, and confirm that participant does not have a full bladder. Clothing will be removed to expose the right upper-arm and an appropriately sized cuff will be positioned as outlined in the "Standard Operating Procedure: Blood Pressure and Glucose" document. Before starting the measurement, the participant will be instructed to sit on a chair in a quiet room for five minutes with legs uncrossed, back supported by the chair, and arm supported on a table at the level of the heart (mid-sternum). Heart rate will also be recorded. The participant will refrain from talking or from engaging in other activities such as reading or using their cell phone in the five minutes prior to and during the blood pressure measurements. ANDES study staff will take the average of all three measurements for SBP and DBP as the final measurement. Uncontrolled hypertension is defined as an average SBP ≥140 mm Hg and/or DBP ≥90 mm Hg.

ANDES participants in the intervention group will have their blood pressure assessed by community health agents monthly for 12 months. Blood pressure assessment will follow the same procedure as described above.

Diabetes assessment. The presence of type 2 diabetes will be determined by use of an FDA-approved laboratory based HbA1c testing (Roche cobas c311 analyzer) as a secondary aim to assess whether the multicomponent intervention has any effect on diabetes control. A measurement will be taken from each participant at the time of the baseline, 12-, and 18-month visits. A trained member of ANDES study staff will perform a minimally invasive finger stick to collect up to 30 µl of whole blood; the sample will then be taken for testing at a central health facility laboratory. Alternatively, patients will have the opportunity to go to the central health facility to have the test performed independently of the home visit. All samples will be destroyed after testing. T2D diagnosis will be defined as ongoing treatment with a hypoglycemic drug or insulin, or a HbA1c ≥7%. If the patient has a diagnosis of T2D, documentation will be obtained from the healthcare facility.

Anthropometry: ANDES research team will assess height with a stadiometer (seca 213, seca, Hamburg, Germany) and waist size with a circumference tape (seca circumference tape 203, Hamburg, Germany) at baseline. ANDES research team will assess weight using a digital balance (seca 803, seca, Hamburg, Germany) at baseline, 12 and 18 months. These measurements will be conducted by ANDES study staff. All anthropometric measurements will be measured in triplicate and mean value will be used as the final anthropometric measurement. ANDES study staff will follow standard procedures for the assessment of anthropometry.

Oxyhemoglobin saturation: ANDES research team will assess oxyhemoglobin saturation using the Rad 5v (Masimo, Irvine CA) pulse oximeter. These measurements will be conducted by ANDES study staff. Participants will be at rest for 2 minutes before oxyhemoglobin saturation. ANDES study staff will record values of oxygen saturation at 60, 90 and 120 seconds. The mean value will be used as the final oxyhemoglobin saturation.

Participants in the usual care group will not receive the same benefits as those in the intervention group, which include medication adherence counseling, assistance with appointment and intensive health coaching. To offset this difference, ANDES study staff will offer the following clinical exams to a random selected group of participants as an additional benefit to address their overall non-communicable disease risk:

Lung function: Spirometry will be offer to all participants at baseline, 12 and 18 months. These measurements will be conducted by ANDES study staff. Spirometry will be conducted by trained staff before and after inhaled bronchodilation (400 mcg of salbutamol using a spacer) following standard American Thoracic Society/European Respiratory Society guidelines. ANDES study staff will use the EasyOne Air handheld spirometer (ndd, Zurich, Switzerland), a device that has been validated and used in several large population-based studies in low-income settings by member of ANDES research team. ANDES research team will adapt a standardized grading system for quality control, review, and interpretation. All lung function tests will be read by a pulmonologist on the ANDES research team as part of the quality control/quality assurance procedures. ANDES study staff will record pre- and post-bronchodilator Peak Expiratory Flow (PEF), Forced Expiratory Volume in One Second (FEV1), and Forced Vital Capacity (FVC). If a spirometry measurement does not meet quality standards of acceptability or reproducibility, ANDES study staff will attempt to repeat the procedure on a different day for a maximum of 3 tests in total. Participants with any spirometry abnormalities will be referred to the nearest health center.

Ambulatory blood pressure: ANDES study staff will offer ambulatory blood pressure measurement at 12, and 18 months. These measurements will be conducted by ANDES study staff. ANDES study staff will measure 24-hour BP by fitting participants with a CE/ISO approved 24-hour ambulatory blood pressure monitor [Contec Medical Systems Co., Ltd, Qinhuangdao, China] oscillometric device with an appropriate cuff for mid-upper arm circumference (small, 20-24 cm; medium, 24-32 cm; large, 32-38 cm). The 24-hour ambulatory blood pressure monitors will be fitted on a regular working day (i.e., a day during which participants would be undertaking their regular daily activities). Blood pressure will be automatically measured at 30 minute intervals during awake periods, 60 minute intervals during asleep periods over the 24 hour period.

Personal exposure to PM2.5: ANDES research team will offer air quality assessment at baseline, 12, and 18 months. These measurements will be conducted by ANDES study staff. ANDES study staff will collect between 24 to 48-hour personal exposures of PM2.5 using the UPAS device (Access Sensor Technologies, Fort Collins, CO). The UPAS is a small, lightweight wearable device that is easy to use and has built-in technology for gravimetric (filter-based) sampling. Long battery life and GPS tracking help ensure that data is collected and that the participant is wearing the device over the specified time interval, and the device can quickly be deployed between locations without having to return to a lab environment between tests for calibration.

Study Staff Training

Health agents will receive training on all study procedures and counseling techniques, including educational and motivational counseling techniques, BP measurement, medication management, and behavioral change related to hypertension. The training program has been designed by the ANDES research team and is based on materials from the Pan-American Health Organization's HEARTS technical package. These materials are validated for use in Spanish, as the health agents are all Spanish speakers and will be coaching most of their patients in Spanish as well. Health agents are instructed and assessed on how to properly take clinical measures under supervision of licensed professionals and study staff. Upon completion of this training, the ANDES research team will issue the health agent a certificate indicating that are able to perform clinical measures and healthy lifestyle coaching on behalf of the ANDES Study.

Fieldworkers hired by the ANDES research team will be trained on all study procedures relevant to the collection of study outcome data, including BP measurement, finger stick, spirometry, ambulatory blood pressure, personal exposure to PM2.5, and CRF administration. These fieldworkers will be responsible for collecting outcome measures to be used in analysis. Fieldworkers will be instructed to remain blinded to each participant's group.

Sample Size

To detect a difference in systolic blood pressure of 2.5 mmHg between intervention and control arms with 95% confidence, 90% power and a standard deviation of 12 mmHg, ANDES research team would need 485 participants per arm. Assuming a conservative lost-to-follow-up of 10% at 12 months, ANDES research team would need to randomize 1068 participants in total (534 participants per arm).

Inclusion Criteria:

1. Diagnosis of HTN, defined as one or more of the following:

   1. Physician-prescribed use of antihypertensive medications, and SBP≥130 mmHg and/or DBP≥80 mmHg
   2. Physician diagnosed hypertension defined as a hypertensive blood pressure measurement (SBP≥140 mmHg and/or DBP≥90 mmHg) documented or measured on two or more separate occasions.

2. Willing and available to receive health agent home visits monthly for the next year.

3. Receiving healthcare through MINSA or EsSalud healthcare systems

Exclusion Criteria:

1. Less than 18 years old;.
2. Has a SBP less than 140 mmHg and DBP less than 90 mmHg and is not currently taking medication;
3. Has a SBP less than 130 mmHg and DBP less than 80 mmHg if currently taking hypertension medication;
4. Unwilling, unable, or or not cognitively capable of providing informed consent;
5. Pregnant or plan to become pregnant in the next 18 months;
6. Plan to move out of the study area in the next 18 months;
7. Are not ambulatory, particularly if bedridden;
8. Currently receives home medical visits;
9. Life expectancy less than 6 months.
10. Lives more than approximately 1 hour from Puno city center;
11. Plan to travel for more than 3 months in the next year;
12. Another ANDES participant already lives in their home;
13. Patients on dialysis or expecting to start dialysis within the next 12 months;
14. Patients with liver failure (meeting Child-Pugh B or C criteria)

In the final year of the award, DCC staff will permit qualified researchers access to the ANDES dataset. A de-identified database, meeting HIPAA requirements, will be stored in an NHLBI repository upon the publication of the ANDES primary endpoint paper. It will include a data dictionary, detailing each dataset's content and structure, variable names, labels, and questionnaire mappings. Additionally, secondary and exploratory paper plans will be outlined. A questionnaire will be required for access, ensuring researchers' credentials and analysis goals are clear. ANDES investigators will assume responsibility for adjudicating requests for data for a defined but brief period of time that coincides with the intensive period during which we are writing secondary and exploratory papers. Following that time period, we will cede responsibility for adjudicating data-access requests to NHLBI personnel who oversee the repository.

Randomization will be done at the conclusion of the study