Virtual AppLication-supported Environment To INcrease Exercise (VALENTINE) during cardiac rehabilitation study: Rationale and design

Abstract

Background

In-person, exercise-based cardiac rehabilitation improves physical activity and reduces morbidity and mortality for patients with cardiovascular disease. However, activity levels may not be optimized and decline over time after patients graduate from cardiac rehabilitation. Scalable interventions through mobile health (mHealth) technologies have the potential to augment activity levels and extend the benefits of cardiac rehabilitation.

Methods

The VALENTINE Study is a prospective, randomized-controlled, remotely-administered trial designed to evaluate an mHealth intervention to supplement cardiac rehabilitation for low- and moderate-risk patients (ClinicalTrials.gov NCT04587882). Participants are randomized to the control or intervention arms of the study. Both groups receive a compatible smartwatch (Fitbit Versa 2 or Apple Watch 4) and usual care. Participants in the intervention arm of the study additionally receive a just-in-time adaptive intervention (JITAI) delivered as contextually tailored notifications promoting low-level physical activity and exercise throughout the day. In addition, they have access to activity tracking and goal setting through the mobile study application and receive weekly activity summaries via email. The primary outcome is change in 6-minute walk distance at 6-months and, secondarily, change in average daily step count. Exploratory analyses will examine the impact of notifications on immediate short-term smartwatch-measured step counts and exercise minutes.

Conclusions

The VALENTINE study leverages innovative techniques in behavioral and cardiovascular disease research and will make a significant contribution to our understanding of how to support patients using mHealth technologies to promote and sustain physical activity.

Background

Cardiovascular disease (CVD) afflicts 9.3% of American adults with a high burden of recurrent events.¹ While the frequency of recurrent events varies by demographic characteristics, up to 45% of patients who have had a first myocardial infarction will have a recurrent coronary event or die from coronary heart disease within 5-years.^{1 , 2} There is thus an urgent need for long-term secondary prevention strategies. Cardiac rehabilitation is a medically supervised risk reduction program for patients recovering from a cardiac event that reduces hospital readmissions and all-cause and cardiovascular mortality.^3–5 Furthermore, in-person, exercise-based cardiac rehabilitation programs improve quality of life and increase exercise capacity, the latter a strong predictor of mortality. Despite the known benefits of cardiac rehabilitation, participation is low⁶ and, for those who graduate, there is evidence of behavioral recidivism in the subsequent months to years.^{7 , 8} There is thus a need for scalable interventions that can be used to augment and extend the benefits of cardiac rehabilitation.

Mobile health (mHealth) technology is now used broadly⁹ and has been predicted to revolutionize care for patients with diverse health conditions, with its key value proposition being its ability to support users longitudinally outside of the confines of siloed clinical encounters.¹⁰ Despite much enthusiasm for mHealth technologies, evidence of health outcome improvements over a sustained period is limited.^11–13 This is, in part, from the inability of earlier studied mHealth technologies to leverage contextual environmental data, which diminishes the relevance of digital interventions over time and leads to habituation. Just-in-time adaptive interventions (JITAIs) have the potential to advance the field of mHealth interventions by using contextual data from wearable devices to deliver tailored support to users at times most likely to modify behavior.^{14 , 15}

Herein we describe the VALENTINE study, a prospective, randomized-controlled, remotely administered trial designed to evaluate an mHealth intervention to supplement cardiac rehabilitation for low- and moderate-risk patients using a JITAI. By leveraging important contextual information from wearable devices, the VALENTINE study was designed to help participants establish habits around physical activity while enrolled in cardiac rehabilitation and then reinforce those habits after graduation from cardiac rehabilitation. In addition to examining the overall effectiveness of the JITAI, the short-term impact of notifications on physical activity can be established through microrandomization in order to learn the most effective times and contexts to support users in achieving their physical activity goals.¹⁶ By nesting the microrandomized notifications within a traditional randomized controlled design, the VALENTINE study will allow us to determine (1) the effect of the larger intervention package on clinically meaningful outcomes and (2) how wearable devices combined with behavioral health theory can enable personalized care for patients with CVD.

Methods

Trial overview and objectives

The VALENTINE study is a prospective, randomized-controlled, remotely administered trial designed to evaluate a JITAI to supplement cardiac rehabilitation for low- and moderate-risk patients (ClinicalTrials.gov NCT04587882). The study design is illustrated in Figure 1. After undergoing screening and consent, participants are randomized to the control or intervention arms of the study. Participants in both groups receive a compatible smartwatch (Fitbit Versa 2 for those with Android smartphones or Apple Watch 4 for those with iPhone smartphones) and usual care. Additionally, participants in the intervention arm of the study receive microrandomized, contextually tailored notifications, have access to activity tracking and goal setting through the mobile study application, and receive weekly activity summaries via email. Participants are followed for 6-months for the remotely assessed primary and secondary endpoints of 6-minute walk distance and step count. No extramural funding was used to support this work. Dr Golbus is funded with salary support by an American Heart Association grant (grant number 20SFRN35370008). The authors are solely responsible for the design and conduct of this study, all study analyses, the drafting and editing of the paper and its final contents.

Figure 1.

Study design. Flowchart of randomized-controlled trial study of participants referred to cardiac rehabilitation. CR, cardiac rehabilitation; ED, emergency department.

The study has 3 primary objectives. The first objective, using a rigorous randomized controlled design, is to assess whether participation in a mobile device-facilitated program can augment and extend the benefits of cardiac rehabilitation by improving functional capacity, as defined by 6-minute walk distance and step count. The second objective, through use of a microrandomized design, is to learn which intervention components work best for whom and in what contexts, to maximize effectiveness and minimize burden. The third and final objective, is to collect baseline and follow-up clinical, functional, quality-of-life, and sensor data from low- and moderate-risk patients with CVD who enroll in cardiac rehabilitation to better understand digital phenotypes in this population. By promoting adherence to best practices for physical activity for patients participating in cardiac rehabilitation, this intervention has the potential to reduce morbidity and mortality from CVD.

Study procedures

Screening, eligibility, and consent

The VALENTINE study launched on October 19, 2020 and enrolls patients at an academic health system and large community healthcare system, both in the state of Michigan. Recruitment occurs through a combination of targeted emails and phone calls, as well as through flyers at cardiac rehabilitation. Patients are considered eligible if they (1) are low- or moderate-risk, guided by the American Association of Cardiovascular and Pulmonary Rehabilitation criteria¹⁷; (2) complete at least 2 cardiac rehabilitation sessions based on a qualifying diagnosis; (3) and own an eligible smartphone. Patients who are unable to safely exercise without supervision and who carry a high-risk condition are excluded from the study. Full inclusion and exclusion criteria are available in Table I.

Table I.

Inclusion and exclusion criteria

Inclusion criteria:

Enrolled in cardiac rehabilitation at a study site Age ≥18 and <75 y Owns an Android or iPhone with study supported operating software. Understands English to enable informed consent, completion of study-related surveys, and compliance with study notifications. Enrolled in cardiac rehabilitation within the past 8-wk based on one of the following indications and has completed at least 2 cardiac rehabilitation sessions: Following percutaneous coronary intervention or coronary artery bypass grafting Valve repair or replacement (either surgical or percutaneous) Coronary artery disease/acute coronary syndrome not requiring revascularization

Exclusion criteria:

Orthopedic or neurological condition limiting ability to exercise Greater than mild cognitive impairment Wrist too large to wear a smartwatch comfortably. Enrolled in hospice care Severe valvular stenosis or regurgitation Unrevascularized, obstructive left main or proximal LAD coronary artery disease Exercise-induced ventricular tachycardia Cardiac arrest within the prior 6 mo New York Heart Association class III or IV heart failure Pulmonary arterial hypertension treated with inhaled or intravenous pulmonary hypertension-specific therapy Ejection fraction <40% Unsafe for participation as determined by a clinical nurse or investigative team.

LAD, left anterior descending.

Informed consent occurs by phone with the consent form signed within the mobile study application MyDataHelps—an application by CareEvolution for conducting health-science studies. Participants have the option to use videoconferencing or screen sharing applications on an as needed basis. Following consent, participants are scheduled for an enrollment appointment and a compatible smartwatch is mailed to their homes (Fitbit Versa 2 or Apple watch series 4). Participants are then randomized to the intervention or control arms of the study, stratified by smartwatch and with block randomization with variable block sizes of 2 to 6 (Figure 1).

Enrollment

After receiving the smartwatch, enrollment appointments are performed by phone. During these appointments, participants are assisted in pairing their smartwatch with their mobile device, configuring their smartwatch to suppress notifications from other health and wellness applications, and saving the “VALENTINE Cardiac Team” as a contact within their phones as notifications are received as text messages. Additionally, participants in the intervention arm set time preferences for receiving notifications by text message in the study application and are orientated to the mobile application, including how to track and review activity data, set goals, and review prior notifications. At the end of the enrollment appointment, all participants are assigned disease-specific and general quality-of-life surveys, provided instructions on how to perform the remote 6-minute walk test, and reminded on expectations for watch wear time. Participants are requested to complete the 6-minute walk test and surveys within 7-days of enrollment.

Study intervention

The study is designed around increasing functional capacity for patients enrolled in cardiac rehabilitation using a JITAI. This involves delivery of microrandomized notifications to participants’ phones and smartwatches to promote physical activity and exercise. Notifications were designed using conceptual behavioral health theories, including goal setting and implementation intentions.^18–20 Adapting from this framework, we use contextual information to provide tailored notifications that encourage participants to engage in physical activity and set exercise goals. Notifications are 1 of 2 types: activity notifications and exercise planning notifications.

Activity notifications are designed to encourage participants to be active though at lower than their target heart rate zones. Notifications are tailored based on 4 dimensions of contexts: weather, day of week, time of day, and phase of cardiac rehabilitation. Participants have a 25% probability of receiving a notification at each of 4 time points (morning, lunchtime, midafternoon, evening), averaging 1 notification per day over the duration of the study (Figure 2). Activity notifications were written by the study team and adapted from prior studies.²¹

Figure 2.

Activity tracking and goal setting through the mobile study intervention. Participants receive contextually tailored notifications on their smartwatches which can be subsequently reviewed in the mobile study application. The application additionally allows participants to set and complete activity goals and to review their recent activity (step count, exercise minutes).

In contrast, exercise planning notifications are designed to encourage participants to be active within their target heart rate zones. Notifications remind participants to plan their exercise for the subsequent day and suggest new or modified activities to increase their exercise repertoires. Notifications are tailored based on 2 dimensions of context: season and phase of cardiac rehabilitation. Participants have a 50% probability of receiving a notification each evening, averaging 3.5 notifications per week over the study duration. Exercise planning notifications were written by cardiac rehabilitation exercise physiologists and modified by study staff. Eleven exercise physiologists and 1 exercise physiology intern participated in a generative session after receiving a 10-minute presentation on study goals and a handout on notification structure. Exercise planning and activity notifications could include personalization with a participant’s preferred name, loss- or gain framing, and inclusion of an emoji or hyperlink to the study dashboard.

In addition to receiving notifications, intervention-arm participants have access to a mobile study application which is paired with their smartwatch. The mobile application allows for self-monitoring of activity data (step count, exercise minutes) and allows participants to set and complete activity goals, which can be adjusted based on performance (Figure 2). Finally, weekly activity summaries are provided to participants in the intervention group via email (Figure 3). Participant-facing emails include encouraging messages as well as a summary of their activity during the prior week, comparing it to earlier phases of the study. This data is also provided to their exercise physiologists while they are enrolled in cardiac rehabilitation as both weekly emails and through a web-based, HIPAA compliant dashboard. This dashboard allows exercise physiologists to review participants’ recent activity levels and to set custom target heart rate zones based on their medical history, which will then appear in the mobile study application (Figure 2).

Figure 3.

Sample weekly email. Weekly emails were designed based on behavioral health theory and change over the course of the study as participants’ progress through cardiac rehabilitation.

Participants in the control group of the study are provided with a Fitbit Versa or an Apple Watch series 4 for outcome assessments and either the Apple Health application or the Fitbit application, respectively. The study application is downloaded on their phones to enable data collection though provides no additional functionality. Participants in the control arm of the study do not have access to self-monitoring through the mobile application, tailored notifications, or weekly participant emails nor is their mHealth data provided to their exercise physiologists in cardiac rehabilitation. Participants in the control arm of the study continue to receive usual care, including cardiac rehabilitation.

Study follow-up

Participants are provided with $25 at baseline and at 6-months for completing study tasks and are eligible to keep their smartwatch after 3-months if adherent with all tasks and with general recommendations for watch wear time. Participants who are nonadherent with recommendations for watch wear time receive reminders through an automated text algorithm with participants receiving up to 2 text messages separated by 48-hours and then a subsequent phone call from a study team member for continued nonadherence. Participants are additionally contacted for nonadherence with study tasks. A study technical support line is available for participants who have study hardware of software problems. Technical issues are addressed by telephone though participants have the option of using screen sharing and videoconferencing software to remotely troubleshoot issues.

End-of-study procedures

At 6-months, participants are asked to complete an additional 6-minute walk test remotely using their smartwatch and the mobile study application. They are also assigned surveys which collect information on medical history, general and disease-specific quality-of-life, and usability of the mobile study application (Figure 1). At the end of the study, participants receive instructions on removing the study application from their phones to protect their privacy and cease passive data collection.²²

Adverse events description and ascertainment

Anticipated adverse events include anxiety from answering questionnaires, skin irritation from the watches, symptoms with exercise, and those related to participants’ underlying cardiovascular conditions. Participants that experience serious adverse events during the first week of the study or prior to completion of the baseline 6-minute walk test (even if after 7-days) are withdrawn. In such situations, participants are given the option of reenrolling in the study if 2 additional cardiac rehabilitation sessions are safely completed. If a serious adverse event occurs beyond this initial period, data is provided on a case report form to a blinded review who makes recommendations regarding the safety of ongoing study participation. Serious events that make physical activity unsafe lead to withdrawal from the study, although data will still be collected on clinical endpoints at 6- and 12-months from the electronic health record for subsequent analyses.

Digital endpoints

Data for all study endpoints are collected remotely. The primary outcome is change in 6-minute walk distance over 6-months as measured by remote 6-minute walk tests. 6-minute walk distance was selected as it can be measured remotely and has been shown to have prognostic significance in diverse CVD populations and to improve in response to exercise-based intervention such as cardiac rehabilitation.^23–26 We selected change in 6-minute walk distance between baseline and 6-months as the primary outcome as the intervention was designed to both augment and extend the benefits of cardiac rehabilitation. The secondary outcome is change in average daily step count during the first and final weeks of the study, accounting only for days participants wore their watches for at least 8 hours. Exploratory outcomes are listed in Table II and include all-cause and cardiovascular hospitalization and mortality, patient reported outcomes on general and disease-specific quality-of-life questionnaires, and subgroup analyses for the primary and secondary endpoints. This includes change in 6-minute walk distance between baseline and 3-months, the time at which most participants are expected to graduate cardiac rehabilitation, to determine the impact of the intervention incremental to standard cardiac rehabilitation.

Table II.

Study endpoints

Primary endpoints	Change in 6-min walk distance on remote 6-min walk test
Secondary endpoints	Change in average daily step count
Exploratory endpoints	Subgroup analysis of the primary and secondary endpoints Change in 6-min walk distance between baseline and 3-mo Change in general- and disease-specific quality-of-life between baseline and 6-mo (PHQ-8, KCCQ, SAQ, SF-12, EQ-5D-5L) All-cause and cardiovascular mortality at 6- and 12-mo* Hospitalization for cardiovascular causes at 6- and 12-mo* Time to first hospitalization*
Proximal outcomes: Activity notifications	Average step count within 60-min of a notification
Proximal outcomes: Exercise planning notifications	Exercise minutes the day after a notification (primary) Minutes in target heart rate zone the day after an exercise notification (secondary)

Primary, secondary, and exploratory study endpoints refer to outcomes from the randomized control trial. Proximal outcomes for activity and exercise planning notifications refer to outcomes in the microrandomized trial and pertain only to the intervention group.

^*Only performed for Michigan Medicine patients.

EQ-5D-5L, EuroQol 5 level version; KCCQ, Kansas City Cardiomyopathy Questionnaires; PHQ-8, Patient Health Questionnaire, 8-items; SAQ, Seattle Angina Questionnaire; SF-12, Short Form health survey, 12 items.

For the microrandomized notifications, effectiveness is assessed via proximal outcomes, which refer to repeated short-term measurements collected passively by the smartwatch (Table II).^{14 , 16} For activity notifications, the primary proximal outcome is step count 60-minutes after a notification is sent. For exercise planning notifications, the primary proximal outcome is exercise minutes the day after a notification is sent. Exercise minutes were defined using criteria set by the Apple watch and as minutes “fairly active” or “very active” by the Fitbit watch.

Statistical considerations

Planned analyses

All analyses will be performed on an intention-to-treat basis. Baseline clinical characteristics will be described as means and standard deviations for continuous symmetric variables and median with interquartile range for skewed continuous variables. Categorical variables will be presented as counts and percentages. Chi-Square test and Fisher’s exact test will be used for categorical variables and the 2-sample t test for continuous variables for all between-group analyses. To account for potential measurement error between devices, for primary and secondary analyses, we will perform regression analysis to test the null hypothesis $H_0:{\beta }_0^{\left(F\right)}={\beta }_0^{\left(A\right)}=0$ for change between baseline and 6 months where (F) and (A) refer to Fitbit and Apple watches respectively. If the null hypothesis is rejected than a secondary analysis will be performed to determine whether to reject the individual null hypotheses. In all cases, we will adjust for study site as a covariate. For all statistical analyses, the level of significance will be set at P < .05.

With respect to the microrandomized notifications, the primary hypothesis is that administration of activity and exercise planning notifications, as compared to no notification, will increase step count in the next 60-minutes and exercise minutes the following day, respectively. Following each microrandomization, we will compute the appropriate proximal outcome (eg, 60-minute step count or daily exercise minutes). To analyze the data, we will use a generalization of regression analysis that ensures unbiased estimation of causal effects of time-varying intervention prompts in mHealth settings. The analysis will be performed separately for each watch type and include study site as a covariate and control covariates that are most highly correlated with physical activity to reduce noise and increase our power to detect a causal treatment effect. To test the primary hypothesis, we will include 3 covariates: step count in prior 60-minutes (or prior day’s exercise minutes), an indicator of the randomized treatment condition (notification vs no notification), and day-in-study. We will test secondary hypotheses concerning treatment effect moderation. Specifically, we will assess if phase of cardiac rehabilitation, time of day, and/or day of the week (weekday vs weekend) moderate effectiveness. Other moderators will be considered in exploratory analyses. To account for missing data in the outcomes, we will investigate which observed variables can be used to explain the missingness. These variables will be included in analyses to adjust for missingness.

Sample size considerations

Sample size calculations were based on the distal outcome of 6-minute walk distance, which reflects the impact of the intervention package. We assume that the intervention group will experience a 50-meter improvement in distance walked on a 6-minute walk test at 6-months. A change in 6-minute walk distance of more than 50 meters is clinically significant in most disease states although prior research has suggested that an improvement of as little as 25 meters may be clinically significant.^{27 , 28} Assuming a standard deviation of 125m, we would need to enroll 200 participants to have 80% power. Given a 10% drop-out rate, we plan to enroll 220 participants total – 110 in each arm.

Data flow and security

Data sources include the smartwatch, smartphone, and electronic medical record. At the end of the study, participants receive instructions on how to stop data sharing through the mobile application MyDataHelps, which would otherwise continue through third parties even in the absence of ongoing data collection by the study team. Importantly, this is consistent with evolving best practices for data collection for mHealth research.²²

Data are stored on secure cloud systems with privacy and security protections in place. All participants are assigned a study ID when recruited, and all identifiable information is stored securely on password-protected computers. All analytic datasets are deidentified.

Study launch and enrollment

The VALENTINE study launched on October 19, 2020. As of November 19, 2021, 145 participants have been consented and randomized, of whom 130 (89.7%) are from the academic medical center (Figure 4). Sixty-four (64) participants have completed 6-months in the study; 56 (38.6%) participants are 65 or older (Table III). The majority of participants enrolled in cardiac rehabilitation after undergoing coronary revascularization [93 (64.1%)] and, secondarily, after valve repair or replacement [32 (22.1%)]. Participants have a high burden of comorbid conditions with 118 (81.4%) participants having coronary artery disease, 46 (31.7%) heart failure, and 44 (30.3%) diabetes mellitus. The majority of participants [90 (62.1%)] have an iPhone and were provided with an Apple watch.

Figure 4.

Enrollment timeline. A total of 145 participants have enrolled in the VALENTINE study as of November 19, 2021.

Table III.

Baseline clinical characteristics for 145 participants enrolled as of November 19, 2021

	Overall (n = 145)
	Mean (SD) or N (%)
Demographics
Age, y	59.6 (10.5)
Sex
Male	100 (69.0)
Female	45 (31.0)
Race
Asian	6 (4.1)
Black	9 (6.2)
White	124 (85.5)
Other	4 (2.8)
Ethnicity
Hispanic	3 (2.1)
Non-Hispanic	139 (95.9)
Phone type
iPhone (Apple Watch)	90 (62.1)
Android phone (Fitbit Versa 2)	55 (37.9)
BMI classification
Underweight	1 (0.7)
Normal weight	16 (11.0)
Overweight	48 (33.1)
Obese	66 (45.5)
Indications for cardiac rehabilitation
PCI or CABG	93 (64.1)
Valve repair or replacement	32 (22.1)
Valve procedure + PCI or CABG	4 (2.8)
CAD or ACS, not revascularized	16 (11.0)
Clinical Diagnoses
Atrial fibrillation or atrial flutter	39 (26.9)
Chronic kidney disease	28 (19.3)
Congestive heart failure	46 (31.7)
Coronary artery disease	118 (81.4)
Diabetes mellitus	44 (30.3)
Hypertension	103 (71.0)

ACS, acute coronary syndrome; CAD, coronary artery disease; CABG, coronary artery bypass grafting; PCI, percutaneous coronary intervention.

Discussion

mHealth interventions have the potential to revolutionize healthcare delivery by improving quality of care, clinical outcomes, and reducing costs. Given their widespread adoption, there is hope that mHealth technologies can deliver personalized, dynamic interventions directly to patients in real-time to assist with management of chronic diseases. Yet mHealth technologies have failed to achieve their potential to date as our understanding of how best to integrate these devices into patients’ daily lives and into the current healthcare infrastructure remains immature for most conditions.^{12 , 29 , 30} The VALENTINE study addresses many of the limitations of prior digital health interventions in several ways.

First, the study utilizes microrandomization, serially randomizing participants in the intervention arm of the study to receive or not receive contextually aligned notifications promoting low-level physical activity and exercise.¹⁶ Messages are delivered at prespecified thresholds designed to minimize user burden and are thus analogous to the frequency with which they would be delivered in practice as part of an optimized intervention. By measuring the response to these notifications using the wearable device, we will be able to determine which intervention components work best for whom and in what contexts.^{14 , 21} Second, we are remotely administering all aspects of the study, including endpoint assessment, which has enabled us to conduct the trial even through early waves of the COVID-19 pandemic. Third, we are using an inclusive study design. Many digital health studies have required that participants own a particular smartwatch or smartphone. We are enrolling patients who own either an Android or an Apple phone and are providing them with a compatible smartwatch. While utilizing 2 study devices presents challenges with respect to the analysis given differences in measurement error between devices, we believe it allows for a more inclusive enrollment strategy. To accommodate this design, we have created an analytic strategy that accounts for these potential device-related differences without significant loss of statistical power. Finally, we follow participants for 6-months, extending study participation for 3-months on average beyond participants’ graduation from cardiac rehabilitation. Thus, the VALENTINE study was designed to promote sustained behavioral change which we will be formally studied through outcome assessments at 3-months and 6-months.

Limitations

While the VALENTINE study is a multicenter trial, the majority of participants have been recruited from a large academic medical center, which may limit generalizability. Second, we are enrolling low- and moderate-risk patients younger than 75 years of age. Thus, we will not know whether this intervention is effective for patients 75 year of age or older or for those with high-risk conditions. Third, all participants are enrolled in center-based or hybrid cardiac rehabilitation programs. As such, the efficacy of the intervention within the framework of home-based cardiac rehabilitation settings cannot be ascertained from this study. Fourth, the study duration is limited to 6-months. While this time frame was selected given the potential for early behavioral recidivism after graduation from cardiac rehabilitation, additional studies are needed with extended follow-up for functional and quality-of-life endpoints over longer periods of time. For participants from the academic medical center, however, we will assess clinical endpoints through 12-months. Fifth, we did not provide participants with a smartphone for use in the study, which has the potential to exacerbate health inequities and could limit the generalizability and scalability of this intervention. Smartphone ownership is high, however, with 85% of adults now owning a smartphone, and home broadband use is growing steadily.^{9 , 31} Finally, all study outcomes are collected digitally which presents unique challenges. While a remote 6-minute walk test has previously been shown to be valid,³² performance characteristics may vary across populations and by device type, and we have had to address varying technical challenges. Our first secondary outcome is average daily step count during the first and final weeks of the studies. Obtaining this data, however, depends on participants to wear their watches and to use the mobile study application to sync their data. As participants in the control group have a lesser need to use the mobile application and may wear their watches less, there is the potential that differences between study groups may not be due to the intervention itself.

Summary

The VALENTINE study leverages innovative techniques in behavioral and CVD research and will make a significant contribution to our understanding of how to extend the benefits of cardiac rehabilitation and treat CVD more effectively. Precision discovery requires a granular understanding of which patient populations respond to which health interventions. By delivering microrandomized notifications within an overarching randomized controlled study design, we can determine how participants’ prior behavior and current context influence the efficacy of individual intervention components over time. Such an approach will allow us to deliver precise digital interventions tailored to the individual participant. By better understanding which participant characteristics influence response to treatment, we can then apply this knowledge more broadly to those with other forms of CVD. Further research is thus necessary to enhance our understanding of behavioral health theory and to develop patient-centered interventions which can be used to augment current modes of healthcare delivery.

Abbreviations:

CVD

Cardiovascular disease

HIPAA

Health Insurance Portability and Accountability Act

JITAI

Just-in-time adaptive intervention

mHealth

Mobile health