Sensor-controlled digital game for Native American adults in the Lumbee Tribe with hypertension self-management: study protocol for a randomized controlled trial

Abstract

Background

Hypertension is a major risk factor for cardiovascular (CV) health in Native Americans (NAs), contributing to disparities in mortality, hospitalizations, and complications that include stroke and kidney diseases. However, despite the benefits of lifestyle modifications for CV health, systemic and cultural barriers hinder their adoption. To promote self-care behaviors, interventions must be culturally tailored and sustainable. Digital games (DGs) offer a promising, community-based approach to enhance self-care for hypertension (HTN) in NAs, aligning with traditional NA practices in which games foster skill-building and engagement. This study focuses on the Lumbee NA community, which faces significant HTN-related disparities. Using community-based participatory research, we are developing a culturally tailored, native-sensor-controlled digital game (N-SCDG) to support HTN self-care behaviors.

Methods

This is a prospective, randomized (1:1) controlled clinical trial with two groups, to evaluate the impact of a culturally tailored N-SCDG on engagement in HTN self-care behaviors and related health outcomes among Lumbee adults at 3 and 6 months. Adults aged ≥ 18 years from the Lumbee tribal community in Robeson County and diagnosed with HTN will be randomized into an N-SCDG intervention group or a sensor-only control group. Both groups will receive a Fitbit activity tracker to monitor physical activity (PA). The N-SCDG group will engage in the game, which incorporates evidence-based HTN education, while the control group will receive the same HTN education in written format. The primary outcome is the mean daily step count, recorded by the activity tracker at 3 and 6 months. Secondary outcomes include systolic blood pressure (SBP), diastolic blood pressure (DBP), BP control, HTN knowledge, self-efficacy, motivation for self-care, quality of life (QoL), and cardiac hospitalization rates.

Discussion

This evaluation of an N-SCDG to enhance HTN self-care in Lumbee adults will integrate culturally relevant design with evidence-based education and thus address a gap in use of digital health tools for NAs. The findings will provide vital data on the impact of digital health interventions to improve HTN outcomes and advance health equity in underserved NA communities.

Trial registration

ClinicalTrials.gov NCT05671406. Registered on January 9, 2024.

Administrative information

Note: the numbers in curly brackets in this protocol refer to SPIRIT checklist item numbers. The order of the items has been modified to group similar items (see http://www.equator-network.org/reporting-guidelines/spirit-2013-statement-defining-standard-protocol-items-for-clinical-trials/).

Title {1}	Sensor-controlled digital game for Native American adults in the Lumbee Tribe with hypertension self-management: study protocol for a randomized controlled trial
Trial registration {2a and 2b}.	ClinicalTrials.gov (NCT05671406)Date of registration: 01.09.2024 https://clinicaltrials.gov/study/NCT05671406?tab=results
Protocol version {3}	Version 1, June 2024
Funding {4}	The research is funded by the National Heart, Lung, and Blood Institute of the National Institutes of Health under award R01HL162598. The content is the sole responsibility of the authors and does not necessarily represent the official views of the National Institutes of Health.
Author details {5a}	• K. Radhakrishnan: School of Nursing, The University of Texas at Austin, Austin, TX, USA • C. Im: School of Nursing, The University of Texas at Austin, Austin, TX, USA • J. Brooks: School of Nursing, The University of North Carolina at Chapel Hill, Chapel Hill, NC, USA • G. Currin: School of Nursing, The University of North Carolina at Chapel Hill, Chapel Hill, NC, USA • A. Rangel: School of Nursing, The University of Texas at Austin, Austin, TX, USA • C. Julien: Department of Computer Science, Virginia Tech University, Blacksburg, VA, USA • M. O’Hair: Good Life Games, LLC, Austin, TX, USA • H. Liang: Department of Management Information Systems, University of Memphis, Memphis, TN, USA • J. Lowe: School of Nursing, The University of Texas at Austin, Austin, TX, USA
	H. Liang: Department of Management Information Systems, University of Memphis, Memphis, TN, USA J. Lowe: School of Nursing, The University of Texas at Austin, Austin, TX, USA
Name and contact information for the trial sponsor {5b}	National Heart, Lung, and Blood Institute of the National Institutes of Health
Role of sponsor {5c}	The content is the sole responsibility of the authors and does not necessarily represent the official views of the National Institutes of Health.

Introduction

Background and rationale {6a}

Hypertension (HTN) is a significant risk indicator for the cardiovascular (CV) health of Native Americans (NAs); it increases their risk for mortality, hospitalizations, stroke, and renal complications [1]. Lifestyle enhancements recommended for managing HTN include engagement in physical activity (PA), along with weight control, reduced sodium intake, a balanced diet, and moderate alcohol consumption—all of which alleviate HTN-related complications and enhance the effectiveness of treatment [2, 3]. However, despite these recommendations, NAs face obstacles in adopting healthier behaviors. Obesity rates in NAs are escalating, and only a fraction of NA adults are able to meet PA guidelines [4]. NA adults’ PA levels, based on current recommendations, remain low at 14.7%, with minimal improvement since 1998 [5].

Although community-based interventions targeting PA, particularly among NA elders, have demonstrated promise in enhancing both self-reported well-being and PA levels, their effectiveness has been hampered by small sample sizes, subjective outcome measurements, and challenges to cultural adaptation [6–8]. Only three trials have evaluated PA interventions among NA adults; none was culturally adapted, and none incorporated objective PA measurements [8]. To bridge this gap, there is a need for culturally relevant and sustainable interventions to promote self-care behaviors among NAs with HTN. An emphasis on understanding motivations and cultural contexts within the NA community is vital for effective CV health promotion.

A promising strategy involves the use of digital games (DGs) [9, 10], which offer an accessible community-based intervention that can enhance engagement in HTN self-care behaviors. Our DG intervention is integrated with an activity sensor that tracks PA and triggers game progression and rewards based on participants’ behaviors in real time. DGs, in contrast to traditional electronic media, combine social connections [11], instantaneous behavior feedback [12], interactive education [13], motivational benefits [14], and appealing stories [15]. Such elements can encourage engagement and foster the development of habits and lifestyle behaviors to improve HTN [16] and, potentially, health outcomes [17, 18]. DGs naturally align with NA culture, in which games have historically served to develop skills within the community [19, 20].

The Lumbee NA community, which constitutes a majority of the population in Robeson County, North Carolina, experiences notable disparities related to HTN and PA behaviors owing to historical trauma and displacement [21]. Approximately 32% of the 45,000 Lumbee adults (those aged 18 and above) are being treated for CV disease, with HTN as a leading disorder. In 2018, age-adjusted incidences of CV disease per 100,000 population (217) were higher in Robeson County than in the state (160) [21]. In this clinical trial, we will culturally tailor our sensor-controlled DG (SCDG) to facilitate engagement in HTN self-care among those in the Lumbee NA tribe, employing community-based participatory research [22]. Subsequently, we will assess the efficacy of the adapted intervention, named the Native-SCDG (N-SCDG), among adults with HTN from the Lumbee tribe.

Objectives {7}

Our study, guided by the Native Reliance model [23], is designed as a prospective, randomized (1:1) controlled trial with two arms to identify the differential benefits of an N-SCDG intervention group (IG) versus a sensor-only control group (CG). In a sample of 220 Lumbee individuals with HTN (n = 110/arm), benefits of the IG versus CG will be evaluated by monitoring enhancements in HTN-related proximal outcomes (daily PA, HTN self-care behaviors, motivations, knowledge, self-efficacy) and distal outcomes (quality of life [QoL]; blood pressure [BP]; and cardiac hospitalizations). In this clinical trial, we hypothesize that, at 3 and 6 months following baseline assessments, NAs with HTN in the N-SCDG intervention will demonstrate elevated PA; enhanced HTN knowledge, self-care behaviors, self-efficacy, motivation, and QoL; and decreased systolic and diastolic BP and cardiac hospitalization in comparison with those in the sensor-only CG.

Trial design {8}

This study, guided by the Native Reliance model [23], is designed as a prospective two-arm superiority randomized (1:1) controlled clinical trial with 2 groups to evaluate the N-SCDG’s impact on Lumbee adults’ engagement in HTN self-care behaviors and HTN-related health outcomes at 3 and 6 months. An N-SCDG playing IG will be compared with a sensor-only CG; both groups receive an activity sensor that monitors PA behaviors. Standard evidence-based HTN education [24, 25] incorporated within the N-SCDG will be provided to the sensor-only CG in written format. We hypothesize below: (1) Compared to the sensor-only CG, participants in the IG receiving the N-SCDG will exhibit increased PA, improved HTN knowledge, self-care behaviors, self-efficacy, motivation, and QoL, as well as fewer cardiac hospitalizations; (2) Compared to the CG, the IG will demonstrate a greater reduction in systolic and diastolic blood pressure at 3 and 6 months. These hypotheses align with the Native Reliance framework and support the evaluation of the N-SCDG intervention’s effectiveness among Lumbee adults with HTN.

Methods: participants, interventions, and outcomes

Study setting {9}

Participants will be recruited in local community clinics associated with the University of North Carolina (UNC) at Chapel Hill, such as the UNC Health Southeastern clinics, which are located in Robeson County, to facilitate access and communication with the local tribal community.

Eligibility criteria {10}

Eligible participants will be adults in the NA Lumbee tribal community in Robeson County aged ≥ 18 years with systolic BP (SBP) ≥ 130 mmHg and/or diastolic BP (DBP) ≥ 80 mmHg on two different measures, or who are using medications for HTN. Participants must also pass a mini-cognitive screen [26] with fluency in English and be able to ambulate independently without a walker or human help [27].

Criteria for exclusion: (1) severe visual or sensory difficulties that prelude the use of a smartphone or sensor’s equipment; (2) chronic renal disease stages 4–5; (3) an indication of terminal illness (e.g., cancer or heart failure); (4) prior cardiac transplantation or placement of a mechanical heart support device due to need for unique self-care; and (5) pregnancy due to unique self-care needs. Participation in the study is safe for pregnant women, but this group will be excluded because of their unique self-care needs. Pregnancy of potential participants will be determined by self-report. If a participant reports pregnancy during the study, the participant will be allowed to rejoin the study at a later date if the participant chooses. The study is registered at ClinicalTrials.gov (NCT05671406).

Who will take informed consent? {26a}

Informed consent will be obtained from potential participants by research personnel at UNC Chapel Hill before the completion of baseline surveys. The research personnel will first read the consent form and solicit questions regarding the consent form over the phone. Participants will then be emailed the informed consent form before their in-person visit. This will ensure adequate time for them to raise questions and hold discussions with family or friends prior to providing informed written consent in person. At the first in-person visit, participants will be asked to sign their consent form, and researchers will emphasize that during any phase of the study, they can respond to any individual questions and/or stop their participation. Both the IG and CG will receive and sign the same informed agreement. Participants will then complete a baseline survey and be blinded to their group assignment.

Additional consent provisions for collection and use of participants’ data and biological specimens {26b}

No blood samples will be collected for this research. Each participant will provide the research team with written, informed consent before any relevant data are shared with other researchers involved in the study or with regulatory agencies. Participants will receive an explanation of this information, and it will be available on their consent forms. All participants must agree with the above.

Interventions

Explanation for the choice of comparators {6b}

Both the N-SCDG playing IG and the sensor-only CG will receive a Fitbit activity tracker that monitors PA behaviors. Standard evidence-based HTN education [24, 25] incorporated within the N-SCDG will be provided to the sensor-only CG in written format.

Intervention description {11a}

N-SCDG intervention

The N-SCDG intervention, developed for both Android and IOS mobile platforms, will be integrated with the Fitbit activity tracker, which we selected for its accuracy and ease of connection with third-party applications [28–30]. The Fitbit tracker has a long battery life, is waterproof, and represents the user’s daily PA using a digital watch paradigm that most people are comfortable with.

At a location of the participant’s choice, an intervention research specialist will install the Fitbit activity tracker along with the tracker-related Fitbit Health and Fitness app and the N-SCDG app on the participant’s smartphone. Following installation, the intervention research specialist will provide training on how to use the tracker and play the N-SCDG intervention. The N-SCDG (Fig. 1) involves a narrative that assists an avatar in traveling to cultural landmarks in the game, utilizing points acquired through the participant’s real-time actions and in-game chores that help maintain the avatar’s ideal health state Game components are guided by the Native Reliance model’s concepts of being responsible, confident, and disciplined [23]. For example, responsible attitudes toward one’s health will be fostered through experiencing consequences of real-time health behaviors within the game. Positive consequences would include active progress in the game toward reaching a cultural landmark, positive feedback, and game rewards (e.g., coins to buy low-salt food items or cultural accessories for the game avatar). In contrast, negative consequences would include stalled progress and adverse health outcomes for the game avatar. Confidence in HTN self-care will be fostered through personalized behavior goals and evidence-based educational content within the game written for those with low literacy [24, 25]. Discipline will be cultivated through tailored game alerts, messages, and incentives to achieve real-time HTN self-care behavior goals based on data from the behavior sensors. Figure 2 presented content for schedule of procedures were presented by using the schematic diagram from SPIRIT [31].

Fig. 1.

Heart Health Mountain N-SCDG

Fig. 2.

Content for the schedule of enrollment, interventions, and assessments [31]*

An instruction manual with frequently asked questions related to troubleshooting will be provided to IG participants. Both remote and in-person technical support will be available to participants to address any issues in using the tracker and the N-SCDG app. If a participant does not own a smartphone, a research specialist will deliver an Android smartphone pre-loaded with the N-SCDG app, connected with the activity tracker, and provide training to use the smartphone and tracker and play the N-SCDG intervention on the smartphone. Game progress, evidence-based HTN educational content, rewards, and incentives will change with participants’ objective behavioral data from the activity tracker, which will automatically sync with the N-SCDG app multiple times daily to conserve battery power and data usage. Bite-sized chunks of information on medication adherence and low-salt diet will be part of the N-SCDG content on HTN self-care. The IG participants will play the N-SCDG intervention for 6 months, with its difficulty level set such that the game character will avoid hospitalization and travel a path based on increases in participants’ PA behaviors from baseline. The participant’s initial step count will be modified to fit the participant’s tolerable PA level.

Sensor-only control

An intervention research specialist will install the Fitbit activity tracker and the tracker-related Fitbit and Fitness app on the participant’s smartphone at a location of the participant’s choice. The research specialist will train participants to use the tracker as in the IG, and personal technical support will also be available to participants if they face any issues installing the tracker and the app.

Similarities between the IG and CG

If a participant does not own a smartphone, a research specialist will deliver an Android smartphone pre-loaded with the Fitbit app for syncing with the activity tracker and will provide training on how to use the smartphone and the tracker. In the IG, the N-SCDG and Fitbit app will transmit participants’ game playing and activity behavioral data to the research team. In the CG, the Fitbit app will transmit only the activity behavioral data to the research team.

Criteria for discontinuing or modifying allocated interventions {11b}

Participants can withdraw from the study at any time. If individuals are discovered to have two elevated BP readings at least 10 min apart and do not wish to continue participation, they will be advised to follow up with their primary care provider for management. If they choose not to engage in the study or are found ineligible but would like information on lifestyle management of HBP, they will be referred to resources at the American Heart Association [32] or the National Heart, Lung, and Blood Institute [33].

Strategies to improve adherence to interventions {11c}

A research team member will remotely track user engagement after baseline data collection every day during the 6 months of participation, using a dashboard built by our computer science team to review behavioral data from the Fitbit app for IG and CG participants. The research specialist will call participants in both groups if behavioral data have not been synced in 3 days.

Relevant concomitant care permitted or prohibited during the trial {11d}

Any concomitant interventions and procedures are permissible during the study.

Provisions for post-trial care {30}

There are no guarantees for auxiliary care, post-trial care, or payment. CG participants are treated the same way as they would be under conventional care, and the intervention is expected. Following trial completion, participants in the CG will not be offered access to the full intervention, as efficacy of the game intervention is yet to be established. However, participants in the CG will be allowed to retain their activity monitors. Participants in the IG will be allowed to retain access to the intervention game and activity monitors if they express interest and the study resources allow for continued access. However, no additional auxiliary or clinical care beyond standard of care will be provided, and no financial compensation will be offered for post-trial participation.

Outcomes {12}

The primary outcome will be the mean of daily steps on PA sensor logs, measured at 3 and 6 months. Secondary outcomes will be SBP, DBP, BP control, HTN knowledge, HTN self-efficacy, motivation for HTN self-care behaviors, QoL, and cardiac hospitalization. Table 1 shows how variables are measured and their time points in data collection.

Table 1.

Data collected, operationally defined, with times collected (reliability and validity footnoted)

	Instrument/Measure	Time Point(Month)
Baseline characteristics
• Sociodemographic and clinical data	Sociodemographic and clinical characteristicsquestionnaire	BS+
• Native Reliance scale	24-item Native Self-Reliance Questionnaire [23]	BS+, 3, 6
•Digital comfort	9-item smartphone sub-scale of media,technology usage, and attitudes scale [34]	BS+
•Depressive symptoms	8-item PROMIS Emotional Distress ShortFormb,c [35]	BS+, 3, 6
• Being confident
• HTN knowledge	17-item Hypertension Lifestyle and Managementscale (HELM) [36]	BS+, 3, 6
•HTN self-efficacy	20-item sub-scale of High Blood Pressure Self-Care Profile (α = 0.91)b,c [37]	BS+, 3, 6
Being disciplined
• Average PA steps (primary)#	Fitbit app logs in intervention and control groups	3, 6
• HTN self-care behaviors	20-item sub-scale of High Blood Pressure Self-Care Profileb,c [37] (α = 0.83)	BS+, 3, 6
• Engagement with SCDG	Backend game logs in intervention group	3, 6
Patient outcomes
• SBP	Digital sphygmomanometerSphygmamometer by RA	BS+, 3, 6
• DBP	Digital sphygmomanometerSphygmamometer by RA	BS+, 3, 6
• BP control	Defined as BP <130/80 mmHg; % of people with BP control	BS+, 3, 6
• QoL	10-item PROMIS Global-10b [38]	BS+, 3, 6
• Cardiac hospitalization	Participant self-report, and clinic healthcareprovider	BS+, 3, 6

Outcome measures

Baseline demographic and clinical data: These include gender, marital status, age, level of education, medications, depressive symptoms, hospitalizations, and comorbid conditions. Participants’ BP will be taken at baseline with an approved FDA digital sphygmomanometer (A&D UA-767), calibrated regularly for accuracy and precision. Participants will be asked to refrain from smoking or consuming alcohol or caffeine for 30 min prior to measurement. Prior to the first measurement, participants will sit silently with the left arm resting on an even surface (at heart level) for more than 5 min. Three evaluations will be made at 1-min intervals, with the second and third measurements averaged.

Follow-up data: BP values at 3- and 6-month follow-ups will be collected as at baseline. Participants’ 3- and 6-month PA data will be extracted from the dashboard data, and group assignments will be removed. These data are available daily via the app and will be aggregated (summed) over each 1-week interval. We will use step count as an objective proxy measure for PA in the IG and CG. See Table 1 for the schedule of outcome measures. The participant timeline is detailed in Fig. 2.

Description of outcome measures

1. Daily physical activity (PA) steps: Daily PA steps are the activity (domain) from participants and will be assessed by the average number of steps per day (measure), as recorded via Fitbit sensor logs integrated within the mobile application (metric). Data will be collected for both the IG and CG at 3 and 6 months, but data at 3 months will be the primary outcome [29]. Group comparisons will be conducted by evaluating the mean difference in change from baseline (aggregation) at each timepoint (between IG and CG) and examining within-group changes from 3 to 6 months (timepoint) to assess longitudinal effects of the intervention.

2. SBP: SBP is a physical health outcome (domain) from participants. It will be measured using an FDA-approved automated digital sphygmomanometer (A&D UA-767) (measure) at baseline, 3 months, and 6 months (timepoint) [39]. The A&D UA-767 is a fully automated oscillometric instrument that has been verified using a mercury sphygmomanometer. Participants will be asked to refrain from smoking or consuming alcohol or caffeine for 30 min prior to measurement. Prior to the first measurement, participants will sit quietly for at least 5 min, resting the left arm on a flat surface (at heart level). Three measures will be performed at 1-min intervals (metric), and the second and third readings in mmHg will be averaged to obtain the mean SBP (aggregation).

3. DBP: DBP is a physical health outcome (domain) from participants. It will be measured using an FDA-approved automated digital sphygmomanometer (A&D UA-767) (measure) at baseline, 3 months, and 6 months (timepoint), following the same procedure described for SBP. Participants will be asked to refrain from smoking or consuming alcohol or caffeine for 30 min before measurement. After sitting quietly for at least 5 min with the left arm resting at heart level, three measurements will be taken at 1-min intervals (metric). The average of the second and third readings in mmHg will be used to determine the mean DBP (aggregation).

4. BP control: BP control is a derived physical health outcome (domain) based on SBP and DBP measurements. BP will be measured using an FDA-approved automated digital sphygmomanometer (A&D UA-767) (measure), following the same procedure described above for SBP and DBP. Measurements will be taken at baseline, 3 months, and 6 months (timepoint). For each timepoint, three readings will be taken at 1-min intervals, and the mean SBP and DBP will be calculated by averaging the second and third readings in mmHg (metric and aggregation). BP control will be defined as “yes” if mean SBP < 140 mmHg and mean DBP < 90 mmHg.

5. HTN knowledge: HTN knowledge is a cognitive/educational outcome (domain) assessed using the Hypertension Knowledge-Level Scale (measure), which includes 22 items across six subdimensions. Each correct response is scored as 1 point, resulting in a total score ranging from 0 to 22 (metric), with higher scores indicating greater hypertension-related knowledge (aggregation). The scale demonstrated good internal consistency, with a Cronbach’s α of 0.82. Participants will complete the scale at baseline, 3 months, and 6 months (timepoint) [36].

6. HTN self-efficacy: HTN self-efficacy is a behavioral/psychological outcome (domain) assessed using the self-efficacy subscale of the High Blood Pressure Self-Care Profile (measure). The subscale consists of 20 items scored on a 4-point Likert scale ranging from 1 (not relevant) to 4 (extremely relevant), evaluating participants’ confidence in engaging in hypertension-related self-care behaviors such as physical activity and nutrition (metric). Total scores range from 20 to 80, with higher scores indicating greater self-efficacy in managing hypertension (aggregation). The subscale has demonstrated good internal consistency (Cronbach’s α = 0.83) [37] and concurrent and construct validity. Participants will complete the scale at baseline, 3 months, and 6 months (timepoint).

7. HTN self-care behaviors: HTN self-care behaviors are behavioral outcomes (domain) assessed using the behavior subscale of the High Blood Pressure Self-Care Profile (measure). This subscale consists of 20 items scored on a 4-point Likert scale ranging from 1 (not relevant) to 4 (extremely relevant), measuring participants’ engagement in hypertension-related activities such as physical exercise and nutrition (metric). Total scores range from 20 to 80, with higher scores reflecting more frequent engagement in self-care behaviors for HTN control (aggregation). The subscale has demonstrated excellent internal consistency (Cronbach’s α = 0.91) [37] and strong evidence of concurrent and construct validity. Participants will complete the scale at baseline, 3 months, and 6 months (timepoint).

8. QoL: QoL is a patient-reported health outcome (domain) assessed using the Patient-Reported Outcomes Measurement Information System (PROMIS) Global-10 (measure). This 10-item questionnaire evaluates general QoL across two domains: physical and mental health. Scores for each domain are calculated using standardized T-scores (metric) based on a normative US population distribution (mean = 50, SD = 10), with higher scores indicating better health status (aggregation) [38]. The PROMIS Global-10 is a validated and reliable instrument (Cronbach’s α = 0.82) used across various populations with chronic conditions. Participants will complete the instrument at baseline, 3 months, and 6 months (timepoint).

9. Cardiac hospitalization: Cardiac hospitalization is a clinical outcome (domain) assessed through participant self-reports collected via periodic online questionnaires, corroborated by hospital discharge summaries when available (measure). Data will be collected at baseline, 3 months, and 6 months (timepoint). Hospitalization will be treated as a binary variable (yes/no) indicating whether a cardiac-related admission occurred during each interval (metric). Reports will be verified using submitted discharge documentation when applicable (aggregation).

10. Intensity of daily PA: Intensity of daily PA is a behavioral outcome (domain) assessed using sensor-derived data from the Fitbit app (measure). The duration (in minutes) of physical activity at three intensity levels (e.g., light, moderate, vigorous) will be recorded daily (metric). Average durations will be computed for each week and aggregated at 3-month and 6-month timepoints (aggregation). Data will be collected continuously through the Fitbit app for both the intervention and control groups (timepoint).

Participant timeline {13}

The participant timeline is detailed in Fig 3. by using CONSORT 2010 guideline [40].

Fig. 3.

Study Flow: Participant–Research Staff Interaction using CONSORT 2010 [40]

Sample size {14}

The sample size will be 220. For 1:1 randomization (IG:CG), we will recruit 110 individuals for each group and anticipate 25% dropout (some will be restored by treating missing data). Our previous research shows that the SCDG can produce a small effect on participants’ PA behavior measured by daily steps (Cohen’s f = 0.10). For power estimation, given that participants in this project are different from those in our previous research, we cannot ascertain that the effect size will remain the same. To be conservative, we have used a very small effect size (Cohen’s f = 0.07) to calculate the desired sample size. Given two groups, three waves of data collection, and an α of 0.05, 110 participants in each group for a total of 220 participants will achieve a statistical power of 94% [41]. Even with 20% attrition, 176 participants will yield a power of 88%.

Recruitment {15}

Participants will be members of the Lumbee tribal community in Robeson County aged ≥ 18 years who have been screened for a current diagnosis of HTN by a health care provider and are on hypertensive medication or who will have 2 BP readings > 130/80 on two separate readings taken at least 10 min apart. All qualified participants (N = 220) will be enrolled after providing informed consent. Snowball sampling will be conducted with recruitment at community events such as health fairs, senior centers, and community clinics run by UNC Health Southeastern, where study flyers are posted. A list of participants in co-investigator Dr. Brooks’ prior studies who consented to participate in future studies will also be used for recruitment. We will disseminate recruitment flyers with information about the study in public places in Robeson County. Given the high prevalence of HTN in the NA Lumbee tribal community, we expect to achieve our intended sample size of 220 participants.

Assignment of intervention: allocation

Sequence generation {16a}

Computer-automated randomization will assign eligible participants stratified by sex to the N-SCDG IG or the sensor-only CG to balance the study design. A researcher unconnected with participants’ enrollment will create an allocation sequence for permuted block randomization using SPSS v27 statistical software and random block sizes of 4 and 6.

Concealment mechanism {16b}

A statistical analyst not involved in recruitment has developed the allocation sequence for permuted block randomization with random blocks of 4 and 6. The group allocation has been entered in REDCap, The University of Texas Austin’s (UT Austin’s) survey management system, which allocates individuals to groups following their registration in the research.

Implementation {16c}

A research specialist in data collection who is also a Lumbee community member will screen potential participants for eligibility during a scheduled call, at the participant’s home or at a location of the participant’s convenience, after potential participants have initiated contact. If the individual is found eligible, the research specialist will collect baseline data from the participants using REDCap. Participants will be considered as enrolled in the study once they complete the baseline survey, and they will be blinded to their group assignment.

Assignment of interventions: blinding

Who will be blinded {17a}

We have designed our implementation team to separate and distinguish intervention implementation from evaluation of all survey outcome measures to reduce any potential bias in collecting outcome data. Participants will receive a link to complete the survey outcome measures on their cellphones on the day the follow-up measurements are due.

Data analysts, outcome assessors, and our statistician will also be blinded to group assignments. Participants will be unaware of their assigned teams because both groups employ digital tools. Participants will be blinded to their group assignment; all will complete the same form to provide informed consent, which includes generic phrases such as “digital health tools.”

Procedure for unblinding if needed {17b}

Throughout the experiment, participants will remain unaware of their group designation.

Data collection and management

Plans for assessment and collection of outcomes {18a}

Following completion of the baseline survey, participants will take a posttest at 3 months for our initial assessment of the intervention’s impact and a second posttest at 6 months for examination of the maintenance of behavioral modifications. A research specialist trained in data collection and unaware of treatment group allocation will contact participants to collect data at the two follow-ups, which participants will complete at home, in clinics run by UNC Health Southeastern, or at locations of their choice. All participants will retain the activity tracker at the end of the study. We will use REDCap and Mosio for text messaging, phone calls, or in-person guidance for online surveys at baseline and 3 and 6 months.

Plans to promote participant retention and complete follow-up {18b}

To improve participants’ engagement and follow-up, the project coordinator will remotely track user engagement after baseline data collection from week 1 to week 24, using a dashboard built by our computer science team to review behavioral data from participants’ sensor apps. A research team member will call participants in both groups if behavioral data have not been synced for 3 days. This method will facilitate quicker collection of safety data, and data collection at 3 and 6 months will include the health outcome of hospitalizations. This data collection mechanism will enable us to identify adverse events, which participants can also report to the research team by email or phone. All participants will receive $15, $20, and $25 as incentives to complete surveys at baseline, 3 months, and 6 months, respectively.

Data management {19}

Every participant will be assigned a unique study ID number linked to all collected data, which will be maintained in the study database. The one reference table that links each study ID with each participant’s name will be stored securely on the encrypted cloud platform UT Box. All other databases will be free of identifying data. Participants’ contact information will be maintained in a Microsoft Excel file on UT Box that will be password protected and accessible only to relevant team members.

Participants’ informed consent will be kept separately from the research data in a password-protected folder on UT Box. All data will be backed up to a secure server at the PI’s institution. Participants will be given a pseudonym and research code number as identifiers instead of their names, with dummy email IDs to enter on the Fitbit Bluetooth-enabled activity tracker in order to protect patient privacy. This pseudonym and data will not be linked to any of their extant digital identities.

The results of the study will be input into REDCap, a safe web-based platform for gathering data in research projects. REDCap databases will be created to minimize data entry errors, with alerts embedded for items left blank or entries outside an acceptable range for a given variable. The database’s electronic audit log shows any changes made to data after original input, including time, date, and the user who made the changes. Reports will be generated by study staff monthly and discussed with the PI to monitor data quality. Data quality control checks will be performed on 10% of completed surveys by the PI or her designee. Study staff will be observed in their data collection annually by the PI or her designee to ensure adherence to study procedures.

Quality control will be maintained with double data entry (discordant entries will be checked and corrected) and range checks (entries will be flagged for correction if they are outside a specified range). Hard-copy data will be stored in locked files and kept entirely confidential. Data will be backed up on a shared box in UT’s cloud system at least daily.

Confidentiality {27}

Confidentiality will be ensured with statistical coding and the removal of identifying information from all data. Each enrolled individual will be given a distinct identification number. Participants will be given a unique pseudonym to enter into the Fitbit activity tracker in order to protect patient privacy. A separate list with the participants’ names and assigned identifiers will be kept on the research computer in a password-protected file. The data collector will provide the assigned unique subject identifiers to the participants every time they complete online questionnaires. The behavioral data of the participants in the two groups will be organized in digital folders and identified by their unique identification numbers. The research specialist and other investigators will work with questionnaires and qualitative data with unique subject identifiers but no names. Any paper survey data collected will be maintained in the site project manager’s private office. All participants’ ID numbers and names will be kept in protected, secured files by the PI. All data will be accessible only to study personnel, and no participants will be identifiable in any project reports.

Plans for collection, laboratory evaluation, and storage of biological specimens for genetic or molecular analysis in this trial/future use {33}

Not applicable; no biological samples are gathered for genomic or cellular investigation.

Statistical methods

Statistical methods for primary and secondary outcomes {20a}

Descriptive statistics will be calculated for research variables to examine missingness, out-of-range values, and distributional properties such as floor and ceiling effects. To assess and quantify the treatment effect on primary and secondary findings, the IG and CG will be examined first using intention-to-treat analysis. We will also employ an instrumental variable approach to estimate causal relationships if there are substantial dropouts. Cronbach’s α for each multi-item scale will be ≥ 0.70 for adequate consistency. In endpoint assessments, we will account for the impact of biological characteristics such as age and gender, which are possible significant predictors of outcomes. The primary and secondary analyses will be statistically significant at α = 0.05.

Primary outcome

For analysis of our primary outcome (PA behavior measured as mean daily steps), we will use a generalized linear mixed model (GLMM) [42]. The present study is longitudinal, with three waves of measurement to show change over time. In comparison with the more rigid repeated measures general linear model, a GLMM is appropriate because of its flexibility for modeling data structures. With random inclinations and captures to represent relationships across time in repeated measurements among individuals, GLMMs can model changes over time for individual participants while enabling the comparison of group differences. In this way, GLMMs are more flexible than approaches such as generalized estimating equations [43]. We will calculate the main impacts of time and treatment, as well as the period-by-treatment association. Because PA behavior measured with daily steps consists of count data, we will use the log link function to specify the distribution as Poisson or as negative binomial if the data are over-dispersed. Baseline sociodemographic variables, social determinants of health, perceived social support, digital comfort, and depressive symptoms will be included in the GLMM as covariates to control for confounding effects. Post hoc analyses will be conducted to evaluate specific PA differences between the IG and CG at 3 and 6 months.

Secondary outcomes

We will also employ GLMMs for SBP, DBP, BP control, HTN knowledge, HTN self-efficacy, motivation for HTN self-care behaviors, HTN self-care behaviors, and QoL. Since these can be treated as continuous variables, we will use Gaussian distribution with an identity link function (i.e., using the original values without transformation). To analyze cardiac hospitalizations, we will compare the two groups using the Cox proportional hazards model for time from baseline to first hospitalization.

Interim analyses {21b}

There are no strategies to undertake interim efficacy or safety analyses of the intervention data, especially because the sample size will be low at interim time points [44]. In addition, this is a low-risk behavioral intervention, so no significant adverse effects are anticipated during the trial.

Methods for additional analyses (e.g., subgroup analyses) {20b}

Section 20a provides all analyses, including the proposals for subgroup analyses.

Methods in analysis to handle protocol non-adherence and any statistical methods to handle missing data {20c}

Missing data will be accounted for as absence at random. Because this is a longitudinal study, participants’ dropout at follow-up visits might lead to attrition bias. We will address this issue in two ways. First, we will compare all sociodemographic variables between the full sample at baseline and the sample at each follow-up measurement. A lack of statistically significant differences would suggest that attrition bias is unlikely. Second, if there are substantial dropouts, we will employ a 2-stage contamination-adjusted intention-to-treat analysis [45] to control for attrition bias. In the first stage, we will treat random assignment as an instrumental variable and regress staying with the study on the instrumental variable to obtain the residual. In the second stage, the residual will be included as a control variable in the GLMM analyses to produce an unbiased analysis of the treatment effect [46]. Noncompliance of participants might lead to a differential dosage problem in which different participants attend different numbers of sessions. To handle the dosage problem, we will calculate an overall dosage score based on system logs, which reflect how much time a participant has spent using the DG, and we will include it as a covariate in data analysis. Regarding other types of sporadic missing data often encountered in longitudinal studies, GLMMs can accommodate missing data points without a need for imputation [47, 48].

Plans to give access to the full protocol, participant-level data, and statistical code {31c}

The majority of the internal research protocol’s information is provided in this publication. Even though the final dataset comprising questionnaire, digital, and qualitative data will be cleansed of identifiers before being released for distribution, we agree that there is still the possibility that participants with certain characteristics might be exposed. As a result, we will make the information as well as related materials available to users only under a sharing data arrangement that includes (1) an assurance to use the data solely for research and not to reveal the identity of any individual respondents; (2) a devotion to secure the data using suitable computer technology; and (3) a dedication to eliminate or return the data once analyses have been completed.

Oversight and monitoring

Composition of the coordinating center and trial steering committee {5d}

The research will be conducted by the PI, a site PI, a data manager, graduate research assistants, and a statistician. Key personnel at UT Austin and UNC Chapel Hill meet weekly, and the entire study group meets monthly. Senior investigators on the research team meet weekly to monitor the study, identify problems early, and respond quickly to identified safety problems.

Composition of the data monitoring committee, its role and reporting structure {21a}

A Data Safety and Monitoring Board (DSMB) will (1) examine the study protocol and the data and safety monitoring plans; (2) assess the study’s progress, focusing on the quality of the data, participation rates, the rate of retention, statistics for adverse events, and risk–benefit profiles; (3) generate suggestions for ending the research project if necessary due to safety concerns; and (4) preserve the privacy of data and monitoring. The DSMB will have two members. Their expertise will include CV disease management, NA culture and health disparities, and health research methodology.

DSMB meetings will occur via audio/video conferencing. The first meeting will take place 6 months after the project begins to review the project and establish guidelines. Following that, the DSMB will meet twice a year to review the project’s progress and data and safety monitoring. A review and report will be written by the DSMB Chair after each meeting and submitted to the PI. The report will also be submitted to the UT Austin IRB.

Adverse event reporting and harms {22}

Participants will be followed for adverse outcomes beginning when they sign their consent and ending 30 days after their final research appointment. For each participant, data collection at 3 and 6 months will include the health outcome of hospitalizations. Adverse events will also be identified; significant adverse events might include the death of a participant. If we detect previously unknown or undiagnosed major depression, we will provide a list of available resources with contact information and encourage participants to inform their primary care provider. Notification of adverse events will be directed immediately to the PI. When an adverse event is reported, the PI will forward all appropriate information to the UT Austin Committee on Human Research. If the DSMB notes a rise in HTN-related morbidity (hospitalization) or mortality in the sample to levels above baseline or has other concerns, the DSMB will contact the study team. All adverse events will be coded and classified using the Medical Dictionary for Regulatory Activities (MedDRA) terminology to ensure consistency in reporting and analysis. All serious adverse events (SAEs), as well as those deemed related to the intervention, will be reported in trial publications. Non-serious events will be summarized and reported as appropriate, based on relevance to study outcomes and ethical reporting standards.

Frequency and plans for auditing trial conduct {23}

The team will verify the skills of study research assistants regarding the installation of the study devices and measurement procedures at the beginning of the study and every 6 months in order to minimize differences in intervener effect, maximize standardization of data collection, and thus minimize measurement error including observer/interviewer bias.

Study data will be entered into REDCap, a secure, web-based application designed to support data capture for research studies. The REDCap databases will be created to minimize data entry errors with alerts embedded for items left blank or entries that are outside of an acceptable range for a given variable. The database incorporates an electronic audit trail to show change(s) to data after original entry including the date/time and user making the change. Reports will be generated by study staff on a monthly basis and discussed with the PI to monitor data quality. Data quality control checks will be performed on 10% of completed surveys by the PI or her designee.

Plans for communicating important protocol amendments to relevant parties (e.g., trial participants, ethical committees) {25}

This study has been approved by the UT Austin IRB; it was registered on January 9, 2024, at ClinicalTrials.gov (NCT05671406). Also, it was conducted in accordance with the Declaration of Helsinki and relevant CONSORT recommendations. The study leaders hold frequent meetings, during which the coordinator highlights significant issues and essential revisions to procedures. Patients qualified for recruitment receive full details of the trial, including possible hazards, and they are asked for informed written consent. Any amendments to the protocol are submitted for approval to the IRB.

Dissemination plans {31a}

We will distribute our findings among those who provide care for NA HTN patients and who may profit from lessons learned. We will also disseminate our results through (1) peer-reviewed health technology journals and CV clinical journals, (2) annual scientific research conferences, (3) local and regional clinical organizations in Robeson County, (4) scheduled continuing education or staff meetings conducted by UNC Health Southeastern clinics for their staff, and (5) members of the Lumbee tribe and tribal council and other NA patient support groups or community forums. In compliance with the NIH Public Access Policy, all peer-reviewed accepted publications will be uploaded to the National Library of Medicine’s PubMed Central by either the study’s researchers or the journal.

Discussion

This publication provides the protocol of a clinical trial to evaluate the effect of a culturally adapted DG intervention on self-care in adults in the NA Lumbee tribe with HTN. Contemporary advances in digital health interventions that have motivated self-care behaviors in other communities have not yet been adapted for the NA population to foster engagement in self-care behaviors. The flexibility provided by our SCDG platform enables the cultural adaptation of a digital health intervention to incorporate the unique cultural contexts of NAs to motivate their health behaviors [49]. A culturally adapted intervention with high usability and resources for support, tailored to users’ varying capabilities in the use of technology, can enable NA adults with HTN to access the advantages of innovative digital health interventions. It is time to empower NA participation in healthcare: to proceed from digital inclusion—the “ability of individuals and groups to access and use health information and communications technologies”—to digital citizenship—“regular (and frequent) access and effective use of digital health technology” for full participation in healthcare [50]. DGs that are fun and easy to use offer a window to accessible digital health interventions that can facilitate NA participation in healthcare and increase NAs’ inclusion in health promotion interventions.

This project will honor the strength and resilience of the NA community to address self-care behaviors that can improve health outcomes associated with HTN. Ultimately, the study based on this protocol will fill a vacuum by providing data on the effect of digital health treatments such as the N-SCDG in people with HTN from the NA Lumbee tribe.

Trial status

This study protocol is version 1 from January 1, 2024. It was registered on January 9, 2024, at ClinicalTrials.gov (NCT05671406). The clinical trial commenced on June 1, 2024, and the first participant was registered on June 26, 2024. The current trial status is to recruiting participants to the study. To date, we have recruited 45 participants. The final participant is expected to be enrolled by the second quarter of 2026 (August 15, 2026), with a total recruitment target of 220 participants. Full study findings are expected to be disclosed by 2027.

Abbreviations

IG

Intervention group

CG

Control group

N-SCDG

Native-sensor-controlled digital game

SCDG

Sensor-controlled digital game

HTN

Hypertension

PA

Physical activity

NA

Native American

SBP

Systolic blood pressure

DBP

Diastolic blood pressure

UNC

University of North Carolina

UT

University of Texas

QoL

Quality of life

Authors’ contributions

K.R. is the chief investigator; she conceived the study and led the proposal and protocol development. J.L.B. contributed to the study design and the development of the proposal. H.L. was the lead trial methodologist. C.I. helped write the protocol. All authors read and approved the final manuscript. Authorship for future trial-related publications will be determined based on substantial contributions to the conception or design of the work; acquisition, analysis, or interpretation of data; drafting or revising the manuscript; and final approval of the version to be published, in accordance with ICMJE guidelines. Professional writers may be used to assist in manuscript preparation; their roles will be fully acknowledged, and they will not meet criteria for authorship unless otherwise justified.

Funding

This research is supported by the National Heart, Lung, and Blood Institute of the National Institutes of Health under award R01HL162598. The content is the sole responsibility of the authors and does not necessarily represent the official views of the National Institutes of Health.

Data availability

Even though the final dataset comprising the questionnaire and digital and qualitative data will be stripped of identifiers prior to release for sharing, we believe there remains the possibility of deductive disclosure of participants with unusual characteristics. Thus, we will make the data and associated documentation available to users only under a data-sharing agreement that provides for (1) a commitment to use the data only for research and not to identify any individual participant; (2) a commitment to secure the data with appropriate computer technology; and (3) a commitment to destroy or return the data after analyses are completed. After review to ensure that the data request is consistent with the project’s original goals and verification that the request meets approval from the IRB, deidentified data may be shared with researchers, but no later than within 1 year from the completion of the funded project period for the parent award or upon acceptance of the data for publication, whichever is earlier. We will share data according to the rules set by the Community Partnership Committee and any tribal review boards. We understand that past research has sometimes caused distrust in NA communities. Our data-sharing approach is designed to respect and address the concerns of tribal communities.

Declarations

Ethics approval and consent to participate {24}

The University of Texas Austin IRB has approved the study (Protocol #: STUDY00002092). Written informed consent to participate will be obtained from all participants.

Consent for publication {32}

Informed consent materials have been developed and approved by the University of Texas Austin IRB prior to recruitment. These materials is included as a supplemental file.

Competing interests {28}

M.O. is the owner of Good Life Games, Inc, a company that develops health games for hire, which provided the prototype game images for this study.