Video-based Intervention to Reduce Treatment and Outcome Disparities in Adults Living with Stroke or Transient Ischemic Attack (VIRTUAL): protocol for a randomized controlled trial

Abstract

Background

Racial and ethnic disparities in post-stroke blood pressure (BP) control persist, and effective interventions to address post-stroke care inequities are needed. We designed a randomized comparative effectiveness trial to evaluate the Video-based Intervention to Reduce Treatment and Outcome Disparities in Adults Living with Stroke or Transient Ischemic Attack (VIRTUAL) model of care for post-stroke BP reduction.

Methods

The study will enroll 534 stroke survivors in a randomized trial to receive either the VIRTUAL intervention or enhanced standard care. Individuals with ischemic stroke, hemorrhagic stroke, or transient ischemic attack (TIA) are enrolled before hospital discharge and randomized (1:1) to VIRTUAL or ESC for their post-stroke care. The VIRTUAL care model is a social risk-informed telehealth intervention that incorporates remote BP monitoring and multidisciplinary clinical care from a clinical provider, pharmacist, and social worker. Telehealth (TH) based clinical visits occur 7, 30, 90, and 150 days after hospital discharge with the multidisciplinary care team. Pharmacists monitor and manage BP between telehealth visits for 6 months after enrollment. Patients randomized to ESC receive standard post-stroke follow-up, a BP monitor (without remote capabilities), and pharmacist-engaged care (monthly calls and communication to primary care). The primary outcome is BP control (< 125/75 mmHg) assessed with 24-h ambulatory BP monitoring (ABPM) 6 months after hospital discharge. The secondary outcomes are 24-h ABPM-assessed BP control (< 125/75 mmHg) at 12 months, 6- and 12-month mean systolic and diastolic ambulatory BP, 12-month composite recurrent vascular events, insurance coverage at 3 and 6 months, hospital readmission rates, and acute healthcare utilization (emergency room and urgent care visits) at 3, 6, and 12 months after hospital discharge.

Discussion

The VIRTUAL care model represents a novel approach to addressing post-stroke BP control disparities. The intervention aims to improve BP control and reduce disparities in a diverse patient population by integrating telehealth with a multidisciplinary team approach and social risk-informed care. Findings from this study will inform evidence-based strategies for enhancing post-stroke care delivery, particularly in underserved populations, and may contribute to reducing healthcare disparities among racial and ethnic groups.

Trial registration

ClinicalTrials.gov. NCT05264298. Registered on March 3, 2022. URL of trial registry record: https://clinicaltrials.gov/study/NCT05264298?cond=stroke&term=virtual%20&rank=2.

Trial status

Protocol version 1.5, approved May 15, 2024. Recruitment started on March 29, 2022, and was completed on April 28, 2025.

Supplementary Information

The online version contains supplementary material available at 10.1186/s13063-025-09003-5.

Background and rationale

Advancements in stroke prevention, acute treatment, and organized systems of care for acute stroke have contributed to declines in stroke mortality over the past decade [1]. However, with increasing survival after stroke and expected increases in stroke incidence related to population aging, the prevalence of stroke in the U.S. is projected to rise to 10 million by 2030 [1, 2]. Stroke survivors (SS) face heightened risks of hospital readmission, recurrent stroke, other cardiovascular events, and poorly controlled risk factors in the early post-stroke phase. However, there are no established models of care for effective secondary stroke prevention [3–7].

There are persistent racial disparities in stroke incidence, mortality, and risk factor control. Black adults in the U.S. have a higher risk of stroke than White adults [8–11]. Projected increases in stroke prevalence also vary by race and ethnicity, with the most significant rises expected in men and women of Hispanic ethnicity and Black race [2]. By 2030, approximately 35% of the population is projected to be Hispanic (21.1%) or Black (13.8%) [12]. Systems of care for SS must be developed with a health equity lens to mitigate racial and ethnic disparities in stroke outcomes.

Hypertension is the most critical risk factor for ischemic and hemorrhagic stroke [1]. Small reductions in systolic blood pressure (BP) after stroke (5 mmHg) are associated with greater than 20% reduction in recurrent stroke risk [13]. Available data suggest that less than half of patients have controlled BP 6 months after ischemic and hemorrhagic stroke. Since the BP target for stroke survivors was lowered to < 130/80 mmHg in the most recent U.S. guidelines for secondary stroke prevention, rates of post-stroke BP control are likely lower than these prior estimates [14]. A recent meta-analysis found an association between more intensive BP treatment and a reduced risk of recurrent stroke [15]. Greater reductions in both systolic and diastolic BP led to larger decreases in the risk of recurrent stroke, which highlights the pressing need for models of care to improve BP control post-stroke, especially among higher-risk populations.

Among SS, isolated behavioral and educational interventions do not improve BP control. The most effective interventions for BP reduction and control address multi-level barriers, utilize home BP monitoring, and involve a team-based integrated multidisciplinary care team using a pharmacist, nurse, or case manager, and/or utilize telehealth [16–24]. A recently published study among Black and Hispanic SS showed dramatic reductions in SBP in the nurse case management and home BP telemonitoring arm. Still, a large proportion of SS remained with uncontrolled BP [25] according to the most recent guidelines for BP control after ischemic stroke, TIA, or intracerebral hemorrhage (< 130/80 mmHg) [14, 26]. There are currently no large interventional studies that have shown a reduction in disparities in BP control, comparing Black and non-Hispanic White SS [27].

Social determinants of health (SDoH) and health-related social needs (HRSN) contribute to disparities in BP control in the general population and SS. Disparities in BP control are medication cost, access to healthcare, access to providers, medication complexity, patient beliefs and perceptions, patient educational achievement and socioeconomic status, patient depression and demoralization, perceived racism and discrimination, social networks and support, physician prescribing practices, and neighborhood segregation [16, 17, 28–35]. Among Hispanic groups, language barriers, acculturation, U.S.-born status, and country of origin may also play a role in BP control [36–40]. Socioeconomic status, medication adherence, self-efficacy, marital status, and level of independence are associated with BP control in SS, who may face additional challenges because of physical and cognitive disability and limited resources from loss of income [41–45].

Despite the clear relationship between social factors, BP control, and disparities in control, SDoH and HRSN are not routinely or explicitly targeted in BP control interventions or models of post-stroke care for SS. Social risk-informed care and social risk-targeted care are strategies that facilitate the integration of SDoH assessments into clinical care and may be key to improving BP control and decreasing racial and ethnic differences in disparities [46, 47]. In social risk-informed care, information about patient living environments, employment, health literacy, and care access is used at the point of care to inform treatment decisions. In social risk-targeted care, clinical encounters are leveraged to address social and economic barriers to health. These care models require organizational change and can be incorporated into multicomponent behavioral interventions; however, there is limited data on the impact of these approaches on health disparities [46, 47].

Objectives

The objective of the Video-based Intervention to Reduce Treatment and Outcome Disparities in Adults Living with Stroke or Transient Ischemic Attack (VIRTUAL) randomized comparative effectiveness trial is to utilize a multidisciplinary approach to post-stroke care, integrating social risk-informed and social risk-targeted care to improve BP control among a diverse population of SS. The VIRTUAL clinical trial is a two-arm prospective randomized, blinded endpoint (PROBE) comparative effectiveness study whose primary aim is to compare the VIRTUAL model of care to enhanced standard care (ESC) for a primary outcome of BP control 6 months after hospital discharge for an ischemic or hemorrhagic stroke or TIA. We used the SPIRIT checklist to ensure adherence to standards for clinical trial design [48].

Study design

Study settings

The trial is conducted at three Memorial Hermann Hospital System sites in Houston, Texas. The system includes four comprehensive stroke centers, the largest in the Texas Medical Center (TMC). Over 1800 patients with stroke are treated at the Memorial Hermann Hospital TMC location yearly, and this serves as the primary site for recruitment. The city of Houston is one of the most diverse in the U.S. when considering race, ethnicity, language, and socioeconomic disadvantage, making it an ideal setting for an interventional study focused on health equity.

Study participants

The study will include 534 adults with ischemic stroke, hemorrhagic stroke, and TIA who meet specific eligibility criteria. The eligibility criteria allow for an inclusive participant population. Participants are recruited from sites with a diverse population (30% Black, 30% Hispanic, and 40% Non-Hispanic White), thereby allowing an assessment of the impact of the intervention on hypertension disparities.

Inclusion criteria

1. Age ≥ 18

2. Presence of hypertension (by clinical history or hospital BP ≥ 140/90 mmHg on two occasions)

3. Plan to discharge home from hospital or inpatient rehabilitation center after ischemic stroke, hemorrhagic stroke, or TIA.

4. Ability to provide consent (patient or caregiver) in English or Spanish. Patients with cognitive impairment or aphasia limiting participation are included if they have a caregiver to assist with monitoring and telehealth visits.

5. Two neurologists must agree on TIA diagnoses.

Exclusion criteria

1. Modified Rankin scale > 4 (severe disability) at time of discharge

2. Life expectancy < 1 year or terminal illness

3. Stroke unrelated to vascular RFs (drug use, trauma, vasculitis)

4. Current pregnancy

5. Symptomatic flow-limiting carotid stenosis without a plan for intervention

6. Long-term BP goal ≥ 130/80 mmHg according to the clinical team

7. Inability to fit standard-size cuff for remote BP monitor (22–42 cm arm circumference).

Recruitment, enrollment and informed consent

Potential patients are recruited from the Memorial Hermann Hospital inpatient stroke service lists. The research coordinators (RC) work with the inpatient stroke clinical team to identify and discuss when the research staff can approach a potentially eligible patient. Physicians of potentially eligible patients are encouraged to describe the study to their patients briefly and obtain permission for research staff to contact the patients. Potentially eligible patients are approached before discharge from enrollment sites to assess their interest and screen them for the trial. Informed consent is obtained by the RC from all participants interested in the trial and eligible to participate. After consent is obtained, baseline assessments and SDoH screens are completed before randomization. Recruitment and enrollment will occur over 3 years.

VIRTUAL intervention

The VIRTUAL intervention aims to address medical and social factors contributing to uncontrolled BP among SS based on a modified theoretical framework (Fig. 1). Upon discharge, participants receive a packet containing a folder with educational materials and study information, a BP device with remote monitoring capability, and a pillbox. They are trained to use the BP monitor and given information about their initial (7–14-day post-hospital discharge) telehealth visit (Fig. 2—Assessment schedule). The assigned pharmacist and social worker (SW) contact the patients within 72 h of discharge to introduce themselves and their role in the study. The pharmacists reconcile medications, and the SW assesses transitions of care barriers (Appendix 1). During the initial telehealth visit, the stroke prevention practitioner (nurse practitioner or neurologist), SW, and pharmacist review hospital records, develop personalized care plans, and address medication adherence and lifestyle factors. At follow-up telehealth visits at 1, 3, and 5 months post-enrollment, providers adjust medications, follow up on studies, provide counseling, and reassess social needs. Pharmacists monitor BP remotely every 2 weeks and adjust medications based on a BP management algorithm operating under a collaborative care agreement (Appendix 2).

Fig. 1.

Framework for blood pressure control in stroke survivors. This framework represents distal and mediating factors on societal, organizational, community, interpersonal, and individual levels that may impact blood pressure control in stroke survivors. Underlined items are associated with disparities in blood pressure control in general population. Green items require additional study In stroke survivors. Abbreviations: obstructive sleep apnea (OSA); continuous positive airway pressure (CPAP); chronic kidney disease (CKD)

Fig. 2.

Study assessment schedule. The figure demonstrates the timeline from enrollment to final outcome assessment for intervention and standard care patients. Not shown in the table: biweekly calls with the pharmacist to review BP in the intervention group and monthly calls with the pharmacist to review BP in the standard care group. All pharmacy calls end at 6 months. Abbreviations: multidisciplinary (Multi-D); telehealth (TH); social worker (SW); case report forms (CRFs); ambulatory blood pressure monitor (ABPM)

Enhanced standard care group

The Enhance Standard Care protocol was designed based on the current standard of care within our institution, which is based on optimal follow-up for SS and recommendations supporting pharmacist-engaged care for post-stroke risk factor control [14]. Participants assigned to the ESC group receive an educational packet and BP monitor (with instruction) before hospital discharge. The BP device received by the ESC group does not have remote monitoring capability. Within 48–72 h, SW and pharmacist calls occur, as previously described. Stroke prevention practitioners (nurse practitioners or neurologists) evaluate them at 7–14 days and again at 3 months if indicated according to the patient’s needs. Per patient preference, standard care visits are conducted via video or in person. Pharmacists contact ESC patients monthly to request their BP log for review and notify primary care providers if BP is out of range. If no primary care provider is available, pharmacists notify the provider managing the patient’s BP (cardiologist, nephrologist, etc.). Monthly pharmacist calls will continue for 6 months, with follow-up primary care provider visits based on preference. This evidence-based approach serves as the study's comparator, aiming to provide multidisciplinary post-stroke care in alignment with current recommendations [14].

Outcomes

The primary outcome measures the proportion of subjects with controlled BP (BP) using 24-h ambulatory BP monitoring (ABPM) (< 125/75 mmHg) 6 months post-discharge.

Secondary outcomes include:

1. Twenty-four-hour systolic BP assessed with ABPM 6 months after hospital discharge.

2. Twenty-four-hour diastolic BP according to ABPM 6 months after hospital discharge.

3. Twenty-four-hour systolic BP according to ABPM 12 months after hospital discharge

4. Twenty-four-hour diastolic BP according to ABPM 12 months after hospital discharge.

5. Twenty-four-hour BP control according to ABPM 12 months post-hospital discharge, defined as < 125/75 mmHg.

6. Composite recurrent vascular event rate (stroke, vascular death, non-elective coronary revascularization, myocardial infarction, heart failure hospitalization) 12 months after hospital discharge

7. Proportion of uninsured SS who have obtained insurance coverage (including Medicaid, state programs, or commercial insurance) by 3- and 6-months following hospital discharge

8. Hospital readmission rates and the number of acute healthcare visits (emergency department, urgent care) 3 months, 6 months, and 12 months after hospital discharge

Additional outcomes

1. Daytime (6 am–10 pm), nighttime (10 pm–6 am) average systolic and diastolic BP at 6 and 12 months.

2. Proportion of current smokers who stop smoking or have at least one quit attempt by 6 and 12 months

3. Proportion of smokers who have been treated for tobacco use disorder at 6 and 12 months

4. Mean change in depressive symptoms using patient's health questionnaire-9 at 6 and 12 months

5. Adherence to the VIRTUAL intervention of telehealth visits (among those randomized to the VIRTUAL arm), defined as at least 2 of the telehealth visits attended

6. Adherence to the VIRTUAL intervention of BP measurements (among those randomized to the VIRTUAL arm), defined as the number of weeks that have met the minimum number of 3 BP measurements per week

7. Cognitive function measured by the Montreal Cognitive Assessment at 6 months and 12 months.

Standard Protocol Items: Recommendations for Interventional Trials (SPIRIT) figure: VIRTUAL trial protocol—schedule of enrollment, intervention and assessments (Fig. 3).

Fig. 3.

SPIRIT figure

Safety outcomes

Safety assessments evaluate adverse events related to aggressive BP reduction, such as syncope, falls, and impaired renal function.

Outcome assessments

The primary outcome is assessed at 6 months, while secondary outcomes are evaluated at 6 and 12 months. Outcome assessments are conducted in person, via home visits, or through TH, depending on patient preferences and needs. Hospital readmissions, acute care utilization, recurrent vascular events, and safety outcomes are assessed at 3, 6, 9, and 12 months through patient or proxy reports and medical record reviews by RC.

Sample size

The sample size justification is based on previous research and preliminary data on BP control post-stroke. Estimates from prior studies suggest that approximately 50% of SS have uncontrolled BP at 3 to 6 months post-stroke. We use a conservative estimate of 50% BP control in the standard care group to ensure an adequate sample size for the primary outcome. It is hypothesized that there will be at least an absolute 15% difference between groups at 6 months (65% in the VIRTUAL arm). With 90% power and alpha set at 0.05, a total of 454 SS will be retained in the study. To account for an estimated attrition rate of 15%, 534 participants will be enrolled over 36 months. While the sample size estimates are based on frequentist statistics, frequentist and Bayesian approaches will be used for analysis. Bayesian methodology will allow for the characterization of the posterior probabilities of potential effect sizes and uncertainties, particularly for secondary outcomes and differential impacts based on race/ethnicity.

Randomization and blinding

Randomization is conducted centrally within the REDCap system. Patients are randomized (1:1) within each site, using a stratified permuted block randomization, with random block sizes of 2 or 4 to the VIRTUAL intervention versus ESC. Stratification variables include insurance status (any versus none) and cognitive score (brief neurophysiological screen < 9 versus 9 +). The randomization sequence is developed by a statistician who is not involved in the enrollment procedure and is programmed into REDCap. After baseline data are collected, the research coordinator enters stratification details into REDCap, and group assignment is provided on the screen.

The study is conducted in a single-blind manner. The physician who analyzes the ABPM data is blinded to the study group and otherwise not involved in this study. They submit their findings to the Data Coordinating Center, and the final data is stored in a form inaccessible to the clinical team.

Data collection and management

Plans for assessment and collection of outcomes

Primary and key secondary outcome assessments are conducted at 6- and 12-month follow-up visits, which may take place in-clinic, at the participant’s home, or via telemedicine. For in-person and in-home visits, participants undergo standard anthropometric measurements including weight, height, and unattended automated blood pressure, performed by the RC using a standardized protocol. Additional outcomes are recorded on paper case report forms (CRFs), including tobacco use, cognitive function assessed using the Montreal Cognitive Assessment (MoCA), depression screen using the PHQ-9, and other patient-centered measures.

The primary outcome, 6-month BP, is collected using a 24-h ABPM device, which is distinct from the remote BP monitor used for the VIRTUAL intervention. After completing the CRFs, participants are trained in the use of a 24-h ambulatory blood pressure monitor (ABPM), placed during the visit. Participants are also provided with a diary to record sleep and wake times and a pre-addressed envelope to return both the diary and ABPM device after the 24-h monitoring period.

For telemedicine visits, participants are mailed the CRFs, ABPM diary, and ABPM device in advance. The video MoCA (reference) is conducted during the virtual visit, and RCs provide instructions for proper ABPM use. Participants are asked to return the completed CRFs, diary, and ABPM using a pre-paid FedEx envelope.

Analysis of secondary outcomes will utilize generalized linear models (i.e. 24 h Systolic, Diastolic BP’s and according to ABPM at 6 and 12 months after hospital discharge, BP control (< 125/75 mmHg) according to ABPM 12 months post-hospital discharge, defined as composite recurrent vascular event rate (stroke, vascular death, non-elective coronary revascularization, myocardial infarction, heart failure hospitalization) 12 months after hospital discharge, proportion of uninsured stroke survivors who have obtained insurance coverage (including Medicaid, state programs, or commercial insurance) by 3- and 6-months following hospital discharge, and health-care utilization (i.e., hospital readmission rates and the number of acute healthcare visits (emergency department, urgent care) 3 months, 6 months, and 12 months after hospital discharge. After adjusting for stratification variables, models will evaluate each outcome as a function of treatment. Similar models will evaluate pre-planned subgroup analyses: after adjusting for stratification variables, generalized linear modeling will evaluate each outcome as a function of subgroup variable (i.e., race, ethnicity, sex and gender), treatment, and the interaction of treatment and subgroup variable. Subgroup analyses will utilize both Frequentist and Bayesian inference.

The ABPM data and patient-reported sleep times are analyzed by a physician blinded to the randomization group and independent of the study. This physician reviews and adjudicates BP data and assesses data quality. Summary data, including mean 24-h, daytime, and nighttime BP, and quality metrics are provided to the Data Coordinating Center (DCC) and are used as the primary and secondary BP outcomes summary measures. We make two attempts to repeat ABPM for individuals with poor-quality data. Unattended automated BP is collected for all participants and used to supplement missing data if needed [49].

Recurrent vascular events are assessed at 6- and 12-month visits and additionally at 3- and 9-month intervals along with adverse events and safety outcomes. The 3- and 9-month assessments are completed by the RC via telephone or during scheduled clinical visits (3 months). Medical record review is also conducted at 3, 6, 9, and 12 months to identify recurrent vascular events and safety outcomes (Fig. 3). Recurrent vascular events (and adverse events) may also be captured and recorded at any time by study personnel and clinicians interacting with participants for BP assessment calls, social work calls, or other participant-initiated communications.

Plans to promote participant retention and complete follow-up

All participants receive a BP monitor at enrollment and are offered a monitor to keep following the primary outcome assessment. Contact information for next of kin or emergency contact is obtained from all participants in case of loss to follow-up. All participants are contacted by the research coordinator at stated intervals for the study duration (every 3 months until 12 months). The research coordinator encourages continued participation in these calls. All participants receive a copy of the ABPM report at the 6-month visit to encourage continued participation for 1 year. The research team will make efforts to minimize the undue burden on participants. Participants are compensated for time and travel and are offered home visits for outcome assessments. All participants are offered continued neurological care following the 6-month outcome assessment.

Data management

The database consists of two separate systems, with an application programming interface between them to send/receive information. The first clinical trial management system (CEDAMS) stores identifying information for contacting the participants, study ID generation, a database that stores the remote BP data, and a dashboard that displays the longitudinal BP data for each patient as part of the intervention by the pharmacists. The dashboard includes features to allow the study team to track BP monitoring adherence. The second database is in REDCap and stores de-identified data from the electronic case report forms, implements the randomization procedure, and stores the outcomes, including the adjudication of adverse events.

Data validation and quality control

All study data will be entered into REDCap, which includes built-in options for data validation and range checking to support data quality control. Queries for out-of-range or inconsistent values are automatically generated and sent to research coordinators (RCs) for resolution. Data entry is reviewed and reconfirmed by senior RCs prior to data lock and ahead of Independent Study Monitoring Committee meetings. Clinical outcome events and safety endpoints will be evaluated biannually by an Independent Study Monitoring Committee.

Criteria for discontinuing or modifying allocated interventions

Patients can withdraw from the study at any time and for any reason. Study withdrawal does not impact future medical care. If a patient withdraws from the study, the VIRTUAL and ESC interventions will cease immediately, and we will retain data collected up to the withdrawal date.

Statistical methods

Primary and secondary outcomes

At 6 months, rates of controlled BP, according to ABPM, will be higher in the intervention arm than in the ESC arm ("condition"in the model). Generalized linear modeling will evaluate BP control at 6 months as a function of treatment after adjustment for stratification variables (site, insurance status, and neuropsychological score) using an intent-to-treat analysis. For each $i$:

$$\begin{array}{cc}& Logit\left(P,{robabilityofBPControl}_i\right)={\beta }_0+{\beta }_1\times Sit{e}_i+{\beta }_2\times InsuranceStatu{s}_i\hfill \\ \hfill & +{\beta }_3\times NeuropsychologicalScoreStrat{a}_i+{\beta }_4\times Conditio{n}_i\hfill \end{array}$$

The statistical analysis plan outlines the analytic plans for all secondary outcomes (Appendix 3).

Interim analyses

No interim analyses will be conducted for efficacy and futility. One formal safety interim analysis of recurrent vascular events (composite, adjudicated, but regardless of whether the events were adjudicated as related or not) will be conducted after enrollment of 267 total participants, of which about half will be randomized to the intervention group. Because ascertainment of safety outcomes may differ by group due to the increased touch points with the intervention participants (with possibly undercounting of events in the control group), we will compare the intervention arm only to assess whether recurrent vascular events ($p_{\mathit{intervention}}^{\mathit{RV}}$) occurred at a high rate, defined as above 20% as observed in a 1- year follow up study in patients who had ischemic stroke and stroke mimics [47, 50]. The Bayesian model will assess the posterior probability $P\left(p_{\mathit{intervention}}^{\mathit{RV}},>,0.20\right)$, and logistic regression will estimate the probability of recurrent vascular events in the active treatment after adjusting for stratification variables assuming priors in the log-form ~ Normal (Mean = 0, SD = 100) for the intercept and ~ Normal (Mean = 0, SD = 10) for the treatment coefficient. The study may be discontinued if that probability is above 90% (Table 1).

Table 1.

Bayesian modeling for interim analysis of safety

Probability of recurrent vascular event		Estimated probability of identifying a probability of recurrent vascular events in treatment > 0.20 as a function of posterior probability threshold (K = 1000 Monte Carlo simulations)
Treatment	Comparison	> 95%	> 90%	> 85%	> 80%
0.20	0.20	0.04	0.08	0.12	0.18
0.21	0.20	0.08	0.16	0.23	0.28
0.22	0.20	0.13	0.21	0.31	0.38
0.23	0.20	0.19	0.31	0.40	0.48
0.24	0.20	0.30	0.42	0.51	0.59
0.25	0.20	0.38	0.52	0.62	0.69
0.26	0.20	0.48	0.62	0.70	0.77
0.27	0.20	0.61	0.74	0.81	0.85
0.28	0.20	0.71	0.81	0.87	0.90
0.29	0.20	0.78	0.88	0.92	0.94
0.30	0.20	0.83	0.91	0.94	0.96

Heterogeneity of treatment effects

Per the NIH guidelines on the inclusion of women and minorities as subjects in clinical research, the primary outcome will be analyzed for evidence of differential treatment effects in subgroups. One of the aims of VIRTUAL is to assess whether, at 6 months, the impact of the intervention will be greater in Black and Hispanic SS. Other pre-specified subgroups will include gender and sex as biological variables.

A Bayesian multiple generalized linear regression will be fit with the outcome of 6-month BP control as a function of the study arm and stratification variables. In addition, separately for each subgroup analysis, we will include covariates for the subgroup and the interaction between the subgroup and the treatment. The interaction term will be used to test for heterogeneity of the treatment effect across subgroups, and the subgroup-specific treatment effects and corresponding credible intervals will be constructed to evaluate subgroup effects [51, 52], accompanied by the posterior probability that a non-zero effect exists. Similar analyses will be conducted for secondary outcomes of 6-month diastolic and systolic BP. The baseline value of the outcome will also be included in the model for systolic and diastolic BP outcomes. For example, for 6-month systolic BP, the interaction will be adjusted for the baseline value of systolic BP. Because the trial is powered to detect a main effect of treatment rather than a clinically relevant interaction effect related to these subgroups, a clinically relevant interaction will be interpreted regardless of statistical significance.

Study monitoring

The independent study monitoring committee (SMC), which includes a biostatistician, a cardiologist, and a vascular neurologist, provides external oversight on the safety aspects of the VIRTUAL study for all subjects and safeguards the interests of the study participants. The key responsibilities of the SMC are as follows: (1) reviewing the study protocol; (2) evaluating the quality of ongoing study conduct, including accrual rate, adherence to protocol, attrition rate, and accuracy and completion of data capture; (3) evaluating clinical outcomes at specified interim analysis time-point; (4) assessing adverse events and safety data at scheduled SMC meetings. The SMC meets every 6 months during the clinical trial to review aggregate summary data. The SMC provides a report of recommendations following each meeting. The SMC will review the interim analysis and provide recommendations for study continuation following review.

Data monitoring and protocol adherence

A DCC, led by a biostatistician (JMY) and supported by a data manager, statistician, and programmer, oversees all data monitoring and quality assurance aspects. The DCC supports database development and maintenance, creates range checks for data entry, creates protocols for adjudications, provides data for and participates in SMC meetings, and offers timely feedback on data issues, including reporting unresolved delinquencies and creating appropriate reports. Additionally, the DCC documents procedures for identifying data errors and protocol violations, regularly evaluates the quality control program, and conducts process reviews to ensure robust study implementation and data accuracy. The DCC collaborates with the lead statistician (CG) to establish the Statistical Analysis Plan (Appendix 3). DCC research coordinators, social workers, pharmacists, and clinical providers receive regular training and guidelines to support protocol adherence. Visit adherence is tracked, and the DCC distributes weekly report cards on screening, enrollment, basic demographics, visit adherence, and participant retention to the research team.

Risks and adverse event reporting

RCs give breaks, as needed, to minimize participant burden during survey completion. Participants can decline to answer questions if fatigued. While there are no costs for participation, participants in the intervention arm may experience interference with work due to the frequency of telehealth visits. Efforts are made to accommodate participants’ schedules, and all necessary equipment for study visits will be provided.

Participants may experience adverse events related to aggressive BP reduction, which are closely monitored and managed by trained clinical staff. In cases of uncontrolled BP despite treatment, consultation with a hypertension specialist will be sought to assist with management. Enhanced standard care participants may be at risk for complications associated with elevated BP. However, they receive follow-up from clinic staff, mitigating potential risks. Vulnerable subjects, including children, prisoners, and institutionalized individuals, are not included in the study to minimize risk.

All adverse events of special interest and serious adverse events (Table 2) are reviewed and recorded, regardless of whether they were directly observed by a research team member, self-reported by a participant or proxy, or if the patient is ascribed to the therapy. The report includes, at a minimum, a description of the event, dates of onset, duration and resolution, event type and severity, contributing factors, and any action taken concerning the study. If the patient self-reports an AE, the date of onset may be approximated. Adverse events are assessed 3, 6, 9, and 12 months after enrollment. Hospital and acute care (emergency department and urgent care) visit data are requested and maintained in binders for all reported events. Independent adjudicators review all acute care visits, hospital admissions, and serious adverse events.

Table 2.

Adverse events and safety outcomes

Serious adverse events

5-point major adverse cardiovascular event - Cardiovascular death - Acute myocardial infarction - Ischemic or hemorrhagic stroke - Acute coronary revascularization procedures - Heart failure hospitalization Other clinical events - Death from any cause - Transient ischemic attack Readmissions including safety events - Syncope associated with hospitalization - Fall associated with hospitalization - Readmission for any cause

Adverse events of special interest

Safety events - Falls with or without injury but w/o hospitalization or emergency visit - Syncope with or without injury but w/o hospitalization or emergency visit - Transient impaired renal function Acute care utilization - Urgent care visit, not associated with serious adverse event - Emergency department visit, not associated with serious adverse event

Relevant concomitant care and interventions

Primary care providers and treating physicians are notified of enrollment in the study. Routine clinical care resumes without interference; however, adjustment of BP medications by other providers is tracked. Enrollment in other studies is tracked, and co-enrollment in lifestyle interventions and other studies that impact the primary outcome is avoided.

Potential benefits to subjects

All participants will receive an enhanced stroke education packet, guaranteed standard-of-care visits, and access to a multidisciplinary team of providers. Both groups will have BP monitoring by a clinical pharmacist, and all participants will receive a BP monitor at no cost. Additionally, participants will receive reports of their 24-h ambulatory BP at 6- and 12-month visits.

Access to data sets

The final dataset will be housed in the DCC (Yamal). Public access to datasets will occur per NIH guidelines.

Dissemination plans

The trial results will be prepared for publication in internationally recognized peer-reviewed journals and will be presented at national and international conferences. Trial participants will be notified of study results via newsletter mailings.

Discussion

The VIRTUAL model of care represents a promising strategy to address the multifaceted needs of SS, with a particular focus on post-stroke BP reduction and the reduction of disparities in care. By leveraging telehealth technology and assembling a multidisciplinary team, the VIRTUAL trial integrates medical management with social support and technological engagement to enhance clinical outcomes, especially within underserved populations. The core of the VIRTUAL intervention, which involves video visits with a multidisciplinary team consisting of a stroke practitioner, SW, and pharmacist, alongside remote BP telemonitoring, provides a holistic approach to ensure that medical interventions are contextualized within the social and environmental realities of patients, with tailored, evidence-based recommendations addressing both medical and SDoH. Existing research studies have laid important groundwork in stroke care delivery, highlighting barriers to implementation and the importance of incorporating social work and pharmacy expertise into stroke care teams. The VIRTUAL trial builds upon these insights, aiming to overcome barriers to care access and utilization through telehealth interventions tailored to the needs of SS. While telehealth has expanded access to care for some SS, disparities in access have been observed. The VIRTUAL trial aims to mitigate these disparities by providing SS in the intervention arm with the necessary tools for telehealth engagement.

Conclusions

The VIRTUAL model of care holds significant implications for post-stroke care delivery and reducing disparities in care access and outcomes. By evaluating the effectiveness of the VIRTUAL model in a rigorous randomized comparative effectiveness trial, this study aims to provide valuable insights into the impact of telehealth interventions for SS, particularly those from underserved populations. The importance of this study lies in its potential to transform post-stroke care practices by addressing not only the medical needs but also the SDoH that impact stroke outcomes. Finally, the findings will be relevant beyond the research setting, informing future stroke care policies and practices and facilitating the widespread adoption of innovative, patient-centered approaches to stroke care.

Abbreviations

BP

Blood pressure

VIRTUAL

Video-based Intervention to Reduce Treatment and Outcome Disparities In Adults Living with Stroke or Transient Ischemic Attack

ESC

Enhanced standard care

TIA

Transient ischemic attack

ABPM

Ambulatory BP monitoring

U.S.

United States

SS

Stroke survivors

SDoH

Social Determinants Of Health

HRSN

Health-related social needs

TMC

Texas Medical Center

SW

Social worker

RC

Research coordinator

SPIRIT

Standard Protocol Items: Recommendations for Interventional Trials

Authors’ contributions

Munachi Okpala, Gabretta Cooksey, Thuy Nguyen, Charles Green, Sarah Cohen, Chigozirim Izeogu, Latonya Bryant, Daphne C. Hernandez, Elmer V. Bernstam, Rhonda Conyers, Mengxi Wang, Michael Gonzales, Charles Cotton, Olasimbo Chiadika, Kristin Varacalli, Sean I. Savitz, Jose-Miguel Yamal, Anjail Sharrief.

Co-authors MO, GC, TN, CG, SC, CI, DH, EVB, RC, MW, MG, CC, OC, KV, SS, JMY, and AS made substantial contributions to the conception and design of the work; MW, MG, CC, and JMY contributed to the creation of new software and programs used for the work; and MO, TN, SC, CI, EVB, SS, JMY, and AS have drafted the work or substantively revised it. Professional writers were not utilized for this study.

All authors read and approved the final manuscript.

Funding

This work was supported by the National Institute of Minority Health and Health Disparities grant number R01MD016465. There was no industry sponsor for this trial. The National Institutes of Health (NIH) is the funder and had no role in the design and conduct of the study; collection, management, analysis, and interpretation of data; writing of the report; or the decision to submit the report for publication.

Data availability

This section is not applicable as the manuscript does not contain any data.

Declarations

Ethics approval and consent to participate

The study has ethics approval from the Center for Protection of Human Subjects at UTHealth (Study HSC-MS-210549) (Original approval date: 8/05/2021). Trained research personnel obtain informed consent from all patients for participation in the trial (Appendix 4). All protocol amendments, protocol deviations, and SMC reports will be submitted to the IRB for review.

Consent for publication

The manuscript does not include any individual person’s data in any form.

Competing interests

AS received consulting opportunities from Abbott Cardiovascular and the Bristol-Myers Squibb/Johnson & Johnson Alliance. The authors declare no other competing interests.