This is an open-access article distributed under the terms of the Creative Commons Attribution License (https://creativecommons.org/licenses/by/4.0/), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work, first published in JMIR Research Protocols, is properly cited. The complete bibliographic information, a link to the original publication on https://www.researchprotocols.org, as well as this copyright and license information must be included.
Participation in ambulatory cardiac rehabilitation remains low, especially among older adults. Although mobile health cardiac rehabilitation (mHealth-CR) provides a novel opportunity to deliver care, age-specific impairments may limit older adults’ uptake, and efficacy data are currently lacking.
This study aims to describe the design of the rehabilitation using mobile health for older adults with ischemic heart disease in the home setting (RESILIENT) trial.
RESILIENT is a multicenter randomized clinical trial that is enrolling patients aged ≥65 years with ischemic heart disease in a 3:1 ratio to either an intervention (mHealth-CR) or control (usual care) arm, with a target sample size of 400 participants. mHealth-CR consists of a commercially available mobile health software platform coupled with weekly exercise therapist sessions to review progress and set new activity goals. The primary outcome is a change in functional mobility (6-minute walk distance), which is measured at baseline and 3 months. Secondary outcomes are health status, goal attainment, hospital readmission, and mortality. Among intervention participants, engagement with the mHealth-CR platform will be analyzed to understand the characteristics that determine different patterns of use (eg, persistent high engagement and declining engagement).
As of December 2021, the RESILIENT trial had enrolled 116 participants. Enrollment is projected to continue until October 2023. The trial results are expected to be reported in 2024.
The RESILIENT trial will generate important evidence about the efficacy of mHealth-CR among older adults in multiple domains and characteristics that determine the sustained use of mHealth-CR. These findings will help design future precision medicine approaches to mobile health implementation in older adults. This knowledge is especially important in light of the COVID-19 pandemic that has shifted much of health care to a remote, internet-based setting.
ClinicalTrials.gov NCT03978130; https://clinicaltrials.gov/ct2/show/NCT03978130
DERR1-10.2196/32163
Among older adults with ischemic heart disease (IHD), participation in ambulatory cardiac rehabilitation (CR) remains low despite decades of evidence about its benefits. Recent estimates suggest that fewer than two-thirds of eligible patients are referred, and even among those referred, only half attend the first session [
Mobile health CR (mHealth-CR) for IHD, which involves the delivery of rehabilitation via portable electronic devices, has proliferated rapidly in recent years [
The rehabilitation using mHealth for older adults with ischemic heart disease in the home setting (RESILIENT) trial (NCT03978130) is recruiting 400 participants with a hospital visit for IHD at 3 academic medical centers: the original 2 sites were New York University (NYU) Langone Health (New York, New York) and Yale New Haven Health (New Haven, Connecticut). The first participant was enrolled on September 1, 2020. A third site, University of Massachusetts (Worcester, Massachusetts), was added in March 2021 to enhance recruitment. For NYU Langone Health, participants are being enrolled at both the NYU Langone Medical Center (New York, New York) and the NYU Langone Hospital–Long Island Hospital (Mineola, New York). NYU Langone Health serves as the coordinating center for both study administration and data management. The primary objective of RESILIENT is to test whether mHealth-CR improves functional capacity, as measured by the 6-minute walk test (6MWT), compared with usual care. We hypothesize that 6MWD (6-minute walk distance) will show significant improvement among participants receiving the study intervention, compared with participants in the usual care arm. RESILIENT was designed using pragmatic trial principles [
The phenotype of interest for RESILIENT is IHD, which is operationalized as a hospital visit for either acute myocardial infarction (AMI) or coronary revascularization (percutaneous coronary intervention [PCI] or coronary artery bypass graft). We chose the hospital visit as the time of enrollment for two reasons: first, previous research has demonstrated that a serious medical illness or procedure can serve as a
Age ≥65 years
Hospital visit for either acute myocardial infarction or coronary revascularization (percutaneous coronary intervention [PCI] or coronary artery bypass graft)
Nonambulatory or regular use of walker for ambulation
Moderate or severe cognitive impairment—defined as cognitive impairment that interferes with daily function
Unable or unwilling to consent
PCI-related groin hematoma that precludes brisk walking
Incarcerated
Unable to use mobile health software in English or Spanish
Severe osteoarthritis or joint replacement within the last 3 months
Parkinson disease or other progressive movement disorder
Projected life expectancy of <3 months
Clinical judgment concerning other safety or nonadherence issues
Adverse event during the screening 6-minute walk test (drop in systolic blood pressure ≥15 mm Hg, chest pain, and ventricular arrhythmia)
Participants are identified through daily electronic health record (EHR) screening of hospital lists with the index condition of AMI, elective PCI, or coronary artery bypass graft. Eligible cases are reviewed by a study physician (JAD, KM, SIC, LK) to ensure they meet the study criteria and are not a false-positive screen (eg, takotsubo cardiomyopathy may be screened by biomarkers, but this does not meet the phenotype of IHD). Potential participants are initially approached while still in hospital whenever possible or, if this is not feasible owing to off-hours procedures, within 48 hours of discharge. The baseline assessment is scheduled within 2 weeks of hospital discharge.
After informed consent and completion of the baseline assessment, including the 6MWT (performed in-person at the study site), enrolled participants are randomized through a function in REDCap (Research Electronic Data Capture; Vanderbilt University) [
Rehabilitation using mobile health for older adults with ischemic heart disease in the home setting (RESILIENT) study design overview. Participants will be randomized in a 3:1 manner to receive mobile health cardiac rehabilitation (mHealth-CR) versus usual care. A target of 400 participants will be enrolled to retain 320 with evaluable end points (accounting for 20% attrition between baseline and 3 months). AMI: acute myocardial infarction; CABG: coronary artery bypass graft; PCI: percutaneous coronary intervention.
The primary efficacy endpoint is the change in 6MWD, reflective of functional capacity, as measured by the 6MWT. The 6MWT is performed during the baseline visit and at the 3-month follow-up visit by a blinded clinical assessor (exercise therapist or research nurse). At baseline, blinding is maintained by randomization occurring after 6MWT completion in a separate space to ensure that the 6MWT assessor is not present. During follow-up, the randomization group is not disclosed to the assessor by the study staff, and the study documentation with this information is not accessible by the 6MWT assessor. The concept of walking to measure functional capacity was introduced by Balke [
There are 5 prespecified secondary efficacy end points.
Participant-reported health status will be measured using the 12-item Short Form Health Survey (general health status) questionnaire [
Changes in activities of daily living (ADLs) are defined as any improvement or worsening in basic ADLs (BADLs) or instrumental ADLs (IADLs) over 3 months. BADLs are basic self-care behaviors, including feeding, toileting, bathing, dressing, transferring, and ambulating [
We explicitly designed RESILIENT to enable the study of participant engagement with mHealth (an implementation end point) in addition to the efficacy end points. Our main measure of engagement is the weekly percent completion of the mHealth-CR program. Completion of mHealth-CR analyzed at weekly intervals allows us to determine distinct engagement trajectories throughout the 3-month study period. Weekly engagement is measured as the fraction of the following 11 elements completed each week: (1-7) daily entry of exercise data and relative perceived exertion (RPE); (8) completed weekly phone calls with exercise therapists; (9) at least one electronic communication with an exercise therapist; (10) watching an educational video (which varies by week); and (11) at least one home blood pressure (BP) measurement.
Study participants randomized to the intervention (mHealth-CR) arm receive three components: (1) mHealth-CR software, (2) communication with an exercise therapist (in-hospital assessment or counseling followed by regular communication postdischarge), and (3) a wearable activity monitoring device. These components are designed to work in concert.
We have partnered with Moving Analytics, which has developed a commercial software platform to deliver mHealth-CR on portable electronic devices (
Study intervention. The intervention includes commercially available mobile health cardiac rehabilitation software from Moving Analytics where participants record their physical activity and self-rate difficulty of their exercise session. This is coupled with passive activity monitoring and weekly phone calls with an exercise therapist for a duration of 3 months.
The intervention arm participants meet the therapist immediately after the baseline interview. This visit includes education on cardiac risk factor management, assessment of baseline functional status, and an introduction to the mHealth-CR software. A personalized exercise program has been designed and includes alternating aerobic exercise (walking and stair climbing) and low-level isometric resistance training (upper body strength exercises using elastic bands). Participants are recommended to exercise for at least 5 out of 7 days per week, with an ideal goal of 150 minutes per week of moderate-intensity exercise [
The intervention follows the United States’
At least 150 minutes of moderate-intensity exercise per week, or 75 minutes of vigorous-intensity aerobic physical activity, or an equivalent combination. Activity should be spread throughout the week (eg, 5 sessions of 30 minutes each).
If 150 minutes of moderate-intensity activity is reached, participants will be encouraged to increase to 300 minutes of activity.
Muscle-strengthening activities take place at least 2 days per week. For purposes of the RESILIENT trial, participants are provided with elastic resistance bands and trained on their use at the time of the baseline visit. They are trained to perform upper and lower body exercises using these bands. Each participant is provided with 3 levels of resistance bands to allow for progression and to be able to adjust resistance for the various exercises.
For those unable to achieve at least 150 minutes of moderate-intensity exercise (eg, owing to functional limitations), lower exercise targets are adapted as endorsed by the guidelines. The principle of move more, sit less will is also recommended.
Balance training is also incorporated into the treatment plan, as recommended by the guidelines, as they pertain to older adults.
Exercise therapists for the trial have at least a master’s level training in exercise therapy. As clinically trained professionals, they may adapt the intervention based on individual study participants’ physical or sensory limitations or specific rehabilitation needs. This concept is similar to traditional rehabilitation, in which the intervention is individualized.
Participants are provided with a Fitbit Inspire or Fitbit Inspire 2 wearable wrist device (Fitbit Inc). This is a commercially available product that measures physical activity based on the number of steps per day. Activity is categorized (based on step count) as sedentary or mildly, moderately, or vigorously active. Heart rate information is also collected by the Inspire 2 model, which was adopted after the first 26 participants. Data are automatically uploaded daily to Moving Analytics and are viewable by both the study participant and the exercise therapist. Weekly phone calls with the exercise therapist include a review of activity data, including the percentage of time spent for each category and total daily step count.
In accordance with current guidelines [
All participants undergo a baseline visit and a 3-month ambulatory visit, led by a research coordinator who measures the elements listed in
Timeline for study participants.
Study arm | Baseline (in hospital) | Home activities | 3 months (ambulatory) |
Intervention and control armsa |
In-person assessment Demographics Height, weight, blood pressure 6MWTb Health status (SF-12c and SAQ-7d) Activities of daily living Cognition (MOCAe) Goal attainment scaling (GASf) Depression (PHQ-9g) Frailty elementsh Chart abstraction Comorbidities, medications, and laboratory values |
Monthly activities of daily living assessment |
In-person assessment Weight and blood pressure 6-6MWT Health status (SF-12 and SAQ-7) Activities of daily living Goal attainment scaling (GAS) Depression (PHQ-9) Frailty elements Hospital readmissions Chart abstraction Hospital readmission (verification)i and attendance at traditional cardiac rehabilitation |
Intervention arm |
Exercise therapist assessment Education on cardiac risk factor management Ascertainment of home environment and mobility barriers Introduction to the mobile health–cardiac rehabilitation software platform Personalized exercise plan |
Daily therapist-directed activity (walking and upper extremity resistance training) Daily mHealth data entry Weekly therapist phone call (counseling or activity review) Weekly video education Weekly blood pressure Fitbit activity tracking and review |
System Usability Scale |
aIntervention and control participants will also receive referral to traditional (ambulatory) cardiac rehabilitation at hospital discharge but not mandated or facilitated attendance of first visit. Usual first ambulatory cardiac rehabilitation visit at New York University and Yale takes place within 4 weeks.
b6MWT: 6-minute walk test (this will be performed by a blinded research nurse).
cSF-12: 12-item Short Form Health Survey.
dSAQ-7: Seattle Angina Questionnaire 7.
eMOCA: Montreal Cognitive Assessment.
fGAS: goal attainment scale.
gPHQ-9: Patient Health Questionnaire 9.
hOn the basis of the 3/5 criteria: unintentional weight loss, weak grip strength (dynamometer), exhaustion, slow gait, and low physical activity.
iHospital readmission will also be ascertained at 6 and 12 months through electronic health record review.
Treatment fidelity is monitored based on principles outlined by the National Institutes of Health’s Behavior Change Consortium (
The Steering Committee consists of the principal investigator (JAD), a biostatistician (SA), and coinvestigator (AS). Moreover, 3 data safety monitoring board (DSMB) members (2 cardiologists and 1 biostatistician) have also been appointed by the National Institute on Aging. The DSMB meets biannually to review recruitment and monitor participant safety.
The safety end point includes (1) fall-related injury (operationalized as any fall requiring acute medical care); (2) hospitalization for acute coronary syndrome; and (3) hospitalization for unstable arrhythmia. Separately, among intervention participants, study staff monitor potential exercise-related adverse events on an ongoing basis. Points of contact include weekly phone calls with the exercise therapist and electronic communication via the mHealth app, which is checked daily. All adverse events are reported to the principal investigator and the DSMB. To reduce the likelihood of these events, participants complete the baseline 6MWT before randomization; if any adverse event occurs during the 6MWT (eg, a drop in systolic BP ≥15 mm Hg, chest pain, or ventricular arrhythmia), participants are deemed ineligible for the trial. Other exclusion criteria (severe osteoarthritis, recent joint replacement, and moderate or severe cognitive impairment) are also intended to minimize risk. Previous studies on home-based CR have reported that adverse events are uncommon [
Statistical comparisons will be performed after enrollment of the full study sample, using 2-sided significance tests and 2-sided CIs; no interim comparative analyses are planned. We will begin all analyses with descriptive summary statistics and graphical displays of all variables, with attention to assessing balance in these characteristics by study group assignment and by assessing the distribution of variables relevant to the choice of statistical tests.
We will assess the difference in 6MWD by calculating difference scores for each participant and comparing the intervention and usual care groups with independent group
Analysis of secondary efficacy end points will proceed in a similar fashion to that of the primary end point.
For
Health status at 3 months will be assessed using linear regression with adjustment for baseline levels and a binary treatment indicator; if necessary, health status scores will be log-transformed to improve the approximation to normality. ADLs (BADLs and IADLs) will be assessed using longitudinal models for monthly scores, with indicators to incorporate time, a binary indicator of treatment group, and patient-level random effects to accommodate repeated assessments within individuals; if monthly scores are not approximately normally distributed, suitable transformations will be sought. Hospital readmission and death will be evaluated using Kaplan-Meier estimates, tested with log-rank statistics, and investigated using Cox proportional hazard models with adjustment for confounders if necessary. As with the models for the primary end point, in each model described, we will assess the need for adjustment for confounders using the change-in-estimate criterion.
Among participants offered mHealth-CR, we will conduct latent class analysis to identify engagement profiles and explore whether these factors indicate membership in a class; these models will use maximum likelihood estimation, implemented with the iterative expectation–maximization algorithm, to identify a latent class solution for the set of indicators. We will evaluate the model fit using the
Although the engagement analysis is largely exploratory (given the paucity of data on mHealth-CR engagement), guided by literature related to engagement in other technologies, we will test whether the trajectory classes differ based on the following characteristics: age (≥80 years), sex, race or ethnicity, comorbidity burden (≥2 chronic medical conditions), frailty, social support (based on living alone), and depressive symptoms (based on Patient Health Questionnaire 9).
We designed our sample size to detect a clinically meaningful difference between treatment arms in our primary efficacy end point, which is the change in the 6MWD from baseline to 3 months. A recent meta-analysis [
New York was the first epicenter in the United States of the COVID-19 outbreak, with the first case being reported on January 3, 2020. Several weeks later, RESILIENT was closed to new participant enrollment by a university-wide mandate applicable to all clinical trials not related to COVID-19. The NYU Langone Main campus opened to enrollment on June 1, 2020, whereas NYU Langone Hospital—Long Island and Yale New Haven Health opened later (July 27, 2020, and July 20, 2020, respectively). Despite reopening, recruitment remained slow during the remainder of 2020 owing to a combination of slow resumption of normal clinical activities (eg, elective PCI scheduling), lower than expected hospital admissions for AMI, and patient fears of returning to the medical center for study visits.
In response to the pandemic, we made several changes to the original protocol. First, as originally designed, much of the mHealth-CR platform relied on recommendations to walk to achieve physical activity targets. This walking typically occurred outdoors or in large indoor spaces (eg, shopping malls); however, during the pandemic, many participants expressed fear of COVID-19 infection through being in public. Accordingly, with DSMB approval, we provided participants with access to several home exercise videos developed by an exercise therapist affiliated with the study (Dr Patrice Hazan). These videos are assigned by study exercise therapists weekly and include warm-ups, three levels of aerobic workouts (beginner, intermediate, and advanced), and stretch routines. Second, participants are not offered cost transportation reimbursement for the baseline (if occurring within 2 weeks of discharge rather than while in hospital) and 3-month study visits, in addition to the regular reimbursement for participation, to alleviate fears of needing to take public transportation. Finally, the University of Massachusetts was added as a study site to accomplish our recruitment goal.
The study operates on a single Institutional Review Board mechanism (sIRB), and was approved by the NYU School of Medicine Institutional Review Board. The NYU Institutional Review Board study number is 18-02017 for RESILIENT.
As of December 2021, the RESILIENT trial has enrolled 116 participants. Enrollment is projected to continue until October 2023. The trial results are expected to be reported in 2024.
The RESILIENT study will evaluate whether mHealth-CR improves functional mobility in older adults with IHD and a range of secondary outcomes, including goal attainment, health status, and hospital readmission. To our knowledge, RESILIENT is the largest trial to date for mHealth-CR in an older adult population. We designed this study in light of an aging US population that faces many current impediments to attending traditional ambulatory CR, including transportation barriers and physical impairments, coupled with widespread dissemination of mobile technologies that enable the delivery of CR at home.
Despite the promise of mHealth-CR, definitive evidence of its efficacy among older adults is lacking. Although a recent systematic review reported that mHealth-CR programs led to similar functional mobility and better adherence compared with traditional ambulatory CR programs, most trial participants were <65 years of age [
With these barriers in mind, we designed RESILIENT as a hybrid intervention in which the use of the mHealth app is supported by a baseline visit and weekly phone calls by an exercise therapist. These clinical encounters serve to establish rapport between the exercise therapist and study participant, evaluate functional limitations and home safety, and address person-specific barriers to technology use. Although the coupling of human intervention with mHealth technology may limit the scalability of the intervention (as the success of RESILIENT is partially dependent on the proficiency of exercise therapists working at the study sites) and may make it difficult to disentangle the effects of human interaction versus the direct benefit of the mHealth platform, we feel that this tradeoff is necessary to provide the intervention with the best chance of success in the context of the target population. We have built-in auditing of these encounters and direct feedback to ensure the fidelity of the study intervention.
Another decision we made in the design of the RESILIENT trial was to enable both intervention and usual care arm participants to receive traditional ambulatory CR. An alternative strategy, which has been adopted by some trials, would have been a head-to-head study of mHealth-CR versus ambulatory CR. However, we feel that denying older adults access to traditional ambulatory CR in light of limited information about mHealth-CR’s efficacy was not in accordance with the standard of care. In the context of our study, mHealth-CR may therefore serve to reinforce behaviors learned in ambulatory CR among those who attend or as a replacement for those who cannot. Accordingly, we will analyze the heterogeneity of the treatment effect among these 2 subgroups.
We have attempted to minimize bias in the RESILIENT trial through randomized treatment allocation, blinded assessment of the primary end point, and a multicenter design that includes a diverse population. However, we acknowledge that there are several potential sources of residual bias. First, as with any clinical trial, there is selection bias as people who agree to enroll are likely to be more motivated than the general patient population, typically with higher health literacy and a lower burden of chronic illness. Second, there is the potential for transfer bias whereby there may be a differential loss to follow-up in the intervention versus control arms. To minimize this possibility, participants in the control arm receive regular calls from the research coordinator to maintain a connection to the study, and the 3-month visit is scheduled on the same day as a clinical encounter whenever possible to minimize barriers to follow-up.
The generalizability of the RESILIENT trial may be limited by factors including a limited number of study sites, availability of the software only in English or Spanish, and use of a proprietary software from a single company. Furthermore, our intervention pairs the expertise of exercise therapists with an mHealth-CR platform, and positive findings should not be construed as the software platform being effective as a standalone product. We designed the intervention to couple in-person contact with the capabilities of mHealth to guide and reinforce healthy behaviors. In our opinion, this pairing of technology with human interaction—especially in an older adult population that may have limited technological proficiency—provides the best chance of success.
In summary, the RESILIENT trial will generate important evidence about the efficacy of mHealth-CR among older adults in domains including functional mobility, health status, and goal attainment. Moreover, patterns of engagement with mHealth-CR (eg, sustained engagement, declining engagement, and persistent low engagement) will be analyzed to understand the characteristics that predict different trajectories. These findings will help in designing future precision approaches to mHealth implementation and in understanding which patients are likely to engage. This knowledge is especially important in light of the COVID-19 pandemic, which has shifted much of health care to a remote, internet-based setting.
National Institutes of Health reporter data.
6-minute walk distance
6-minute walk test
activity of daily living
acute myocardial infarction
basic activity of daily living
blood pressure
cardiac rehabilitation
data safety monitoring board
electronic health record
goal attainment scale
instrumental activity of daily living
ischemic heart disease
mobile health
mobile health cardiac rehabilitation
New York University
percutaneous coronary intervention
Research Electronic Data Capture
rehabilitation using mobile health for older adults with ischemic heart disease in the home setting
relative perceived exertion
This rehabilitation using mobile health for older adults with ischemic heart disease in the home setting trial is supported by the National Institute on Aging (R01AG062520). REDCap (Research Electronic Data Capture) is supported by the National Center for Advancing Translational Sciences, National Institutes of Health, through grant UL1 TR00000. The content is solely the responsibility of the authors and does not necessarily represent the official views of the National Institutes of Health. TMG is supported by the Yale Claude D Pepper Older Americans Independence Center (P30AG021342).
None declared.