Cardiovascular diseases are the leading cause of death worldwide, accounting for approximately 17 million deaths annually [1]. Improvements in cardiac treatment over the past decades have increased the number of people surviving cardiac events and living with cardiac disease. Consequently, cardiac disease management and tertiary prevention of adverse outcomes such as recurrent cardiac events or mortality have become increasingly important in clinical practice and patients’ lives [2]. Cardiac rehabilitation (CR) programs play an important role in this risk factor management and have been demonstrated to improve patient outcomes effectively [3].

The World Health Organization defines CR as “the sum of activities required to influence favorably the underlying cause of the disease, as well as to provide the best possible physical, mental, and social conditions, so that the patients may, by their own efforts, preserve or resume when lost as normal a place as possible in the community” [4]. CR is a guideline-recommended [5], multidisciplinary, multimodal tertiary prevention program consisting of modules that include exercise training, optimization of medical treatment and risk factor management, smoking cessation, heart-healthy dietary habits, risk factor education, and psychosocial counseling [6,7]. Phase 1 is the inpatient care that is characterized by the implementation of modern acute coronary revascularization techniques as well as the application of effective acute and long-term pharmacotherapy aiming at early mobilization of the patient after a cardiac event or surgery. This phase is followed by phase-2 CR, which typically addresses patient’s personal risk factor modification by adopting a healthy lifestyle. This represents the most important challenge, often described as the intention-behavior gap. Changes achieved during phase 2 are aimed at being consolidated in phase-3 CR. Long-term maintenance of health behavior changes after CR (ie, during phase-3 CR) varies substantially in the level of support offered to patients [8]. In most countries, the availability of phase-3 CR support is limited because of reimbursement policies. When center-based support is offered, only a small subgroup of patients attends these sessions, often due to a lack of personalized guidance and logistical challenges such as time constraints and travel burden [9,10]. These patient- and health system–related barriers contribute to low adherence to and uptake of long-term rehabilitation and cardiovascular health maintenance programs despite their potential health benefits [11,12]. Therefore, it is of utmost importance to explore delivery modes for long-term CR maintenance that better align with the needs of both patients and healthcare providers, while addressing existing barriers.

Web-based care modalities for CR might be the way forward to address the challenges associated with “traditional” in-person CR. These modalities offer the opportunity to provide personalized care or support to patients at their convenience, while also being scalable and cost-effective [13-15]. Such an approach should include both a continuous risk analysis by monitoring multiple disease-related prognostic parameters as well as regular support and guidance for a cardiac healthy lifestyle. A recent meta-analysis by Heimer et al [16] demonstrated that eHealth-based, phase-3 CR is effective in increasing exercise capacity, improving quality of life, and decreasing blood pressure (BP). Another meta-analysis found that remote CR is associated with better behavioral, psychosocial, and health service use outcomes [13,17]. These findings suggest that technological innovations can help develop a continuum of care after initial center-based CR to sustain its benefits and improve health outcomes. Therefore, recent studies have focused on the development of behavioral interventions to support long-term CR maintenance [18-20].

Developing and maintaining a cardiac-healthy lifestyle is of utmost importance to improve prognosis and reduce mortality risk in people with cardiac disease. It has been demonstrated by multiple studies [21,22] and advocated by others [23,24] that health behaviors, and in particular physical activity, should be addressed to improve cardiac health. Changing health behavior is a challenging endeavor, and only few people succeed [25]. Particularly, patients who face complicated disease regimen or adjustment problems (psychological or psychosocial) due to their illness (such as coronary artery disease [CAD]) may experience health behavior change as burdensome. This is also reflected by the typically temporary nature of the changes in health behavior that do not persist in the long term [26]. To promote sustainable behavior change, the application of behavior change techniques (BCTs) has been advocated [27]. The current literature indicates that a variety of BCTs may be implemented in CR programs to enhance their effectiveness. The TIMELY consortium conducted 2 studies that guided the development of the TIMELY intervention, 1 focusing on the patient perspective (qualitative) [28] and 1 evaluating existing literature through a systematic review [29]. Triangulating the findings from the studies revealed that specific BCTs can be successfully implemented to improve health behavior and clinical outcomes when they align with patients’ preferences and are optimally matched with target behaviors.

To overcome current barriers related to phase-3 CR programs, the TIMELY trial aims to evaluate the cost-effectiveness of a personalized eHealth platform and health behavior change app for patients with CAD. This platform and app will provide support for health behavior change (eg, artificial intelligence [AI]–supported personalized exercise prescription [EP]) and facilitate disease-related self-management based on current clinical guidelines and in accordance with patients’ individual needs. The TIMELY intervention will integrate (through the TIMELY platform) demographic, clinical, and psychosocial data to assess cardiac risk and support patients in maintaining and improving health behaviors via app-based feedback and prompts. Therefore, the TIMELY intervention is designed to preserve the benefits of CR and reduce cardiac risk in patients with CAD. At the same time, this approach reduces the burden on health care providers and the health care system. The TIMELY intervention is the first intervention for patients with CAD to support its users with physical activity and stress management through a chatbot function, and to offer a personalized weekly EP that is adapted based on exercise parameters collected throughout the week [30]. Previous interventions using increasing exercise goals based the goals on single parameters such as baseline activity levels or heart rate reserve, whereas the EP in the TIMELY intervention is based on several medical, demographic, and behavioral parameters [31,32]. Figure 1 shows a graphic outline of the TIMELY approach. It is hypothesized that participants who follow the TIMELY intervention, as compared to care as usual, will have a lower 10-year cardiac risk mortality and improved physical fitness over time. In addition, they are expected to have better physical and mental well-being and quality of life.

The TIMELY study is an international (Germany, Spain, and the Netherlands), multicenter, randomized controlled trial. Patients randomly assigned to the intervention condition will receive a 6-month, app-based, personalized behavior change intervention. Assessments will be performed at baseline and at 3, 6, and 12 months.

The TIMELY trial has been registered at ClinicalTrials.gov (NCT05955625) and has been approved by the Medical Ethics Committee (METC-Brabant NL82723.028.22/P2251 Tilburg, protocol V9 24-05-2023, the Netherlands). Local approvals from the participating centers (Germany: Z-125/2023; Spain: CEIm-G 2023/119) have been obtained. Data collection was initiated after receiving a signed informed consent from the participants. Participants have the option to opt out at any time—before, during, or after data collection—in which case their data will be deleted. Data will be pseudonymized, and only the research team from each center will have access to the key list for their center. A data sharing agreement between the participating centers has been signed for sharing pseudonymized data for data analysis purposes. Informed consent covers secondary analysis without additional consent. Participants were not reimbursed for participation except for study-related expenses (eg, parking costs).

Tilburg University has been appointed as the trial sponsor and will perform regular monitoring of the trial. This includes 2 visits (midterm and closing-out) to the clinical centers in Germany and Spain. During this visit, the investigator site file will be reviewed, and random checks of 20% of the participants regarding clinical data and questionnaires will be performed. Informed consent forms will be checked for all participants. In the Netherlands, the monitoring visits will be performed by an independent researcher from Tilburg University who is not affiliated with the TIMELY trial.

Throughout the study period, monitoring of serious adverse events will be performed per center. If serious adverse events are reported by patients or discovered in the patient record, this will be reported to the sponsor, who will register the events within 1 week at the Dutch Central Committee on Research Involving Human Subjects. The events will also be registered at the German Federal Institute for Drugs and Medical Devices.

The results of the TIMELY trial will be disseminated in academic and health care settings through scientific publications, symposia, and conference visits. Study participants will be informed about the results via a newsletter containing the highlights of the study outcomes. To reach a broader audience, local newspapers will be contacted to disseminate the results among the general public.

Women and men diagnosed with CAD will be included in the study. The following eligibility criteria have been defined:

A total of 360 patients will be enrolled across 3 participating centers: Hospital Universitario de Santiago de Compostela, Spain (120/360, 33.3%); Clinic Königsfeld of the German Pension Fund at the University of Witten/Herdecke, Ennepetal, Germany (120/360, 33.3%); and Elisabeth-TweeSteden Hospital, Tilburg, the Netherlands (120/360, 33.3%). At each center, patients will be randomly assigned, with 60 (50%) out of 120 receiving the intervention in addition to care as usual, and 60 (50%) receiving care as usual. Of 360 patients, 180 (50%) will receive the intervention, and 180 (50%) patients will receive care as usual.

On the basis of participation rates in previous randomized controlled trials [33], it is expected that approximately 33% of patients will decline participation. It is expected that a total of 480 patients will have to be approached to include 360 participants across the 3 study sites. Sample size calculations were based on medium-large effect sizes and statistical power analyses were conducted using G*Power (version 3.1.9.4; Heinrich-Heine-Universität Düsseldorf) [34]. A sample of 120 patients per center will enable detection of a medium effect size. The f² for the between-within group interaction is 0.15 in a mixed model ANOVA, with a power of 0.80 and a two-sided α level of 0.025 (0.05/2, as two primary outcomes will be evaluated; details are provided in the Study Objective section). In brief, the set of variables for this analysis was based on a 2 (between-group: intervention vs control)×2 (repeated measures at baseline and 6 months) mixed model ANOVA, assuming sphericity and a correlation of 0.3 between the repeated measures.

Participants will be randomly assigned (1:1) to either the intervention group or the care-as-usual group. A computer-generated randomization sequence will be delivered, in an Excel (Microsoft Corp) file, to each recruitment site and will be concealed by one of the researchers by covering the allocation numbers with “black paint.” Randomization will be conducted and dated, per patient, by the researcher after the participant has signed the informed consent and baseline assessments have been performed.

Given the nature of the study, blinding of the researchers or participants during the behavioral intervention trial is not possible. However, the analysis of blood samples and all other outcome measures will be conducted by research team members who are blinded to the participants’ group assignment.

Approximately 2 weeks before completing CR, patients who have started the program will be approached by a research assistant for participation after eligibility (inclusion and exclusion criteria) has been checked by the local primary researcher or clinician. Patients who are eligible for CR but decide not to take part in the program will be approached after the research assistant is notified of patients’ participation status. If a patient expresses interest in participating, study information will be provided both verbally and in writing (patient information letter including the informed consent form).

For patients who consent to participate in the study, the baseline assessment includes physical exercise and functional tests, body composition analysis, blood draws (Multimedia Appendix 1), and completion of validated (web-based) questionnaires. Clinical data will be provided by the patient through an inquiry of selected data from the patient’s medical record. At 3 months, all participants will receive an online link to fill in a set of validated questionnaires to evaluate potential early changes that occur during the intervention. At 6 months, participants will be invited to the hospital to return the study devices (refer to the Study Procedure section) and undergo the second physical exercise testing, blood sampling, and questionnaire assessment (online). At 12 months, patients will be invited for the final hospital visit during which the same assessments as obtained at the 6-month follow-up will be performed. For patients who do not fill in the questionnaires within a given period of 10 days, up to 3 reminder phone calls will be made by the researcher.

For descriptive statistics, categorical variables will be compared with the χ² test (Fisher exact test when appropriate). Continuous variables will be compared using independent sample t test or a nonparametric test where appropriate.

To address the main hypothesis, repeated measures ANOVA using generalized linear mixed modeling analysis will be used to investigate the effect of the intervention on changes from baseline to 6 months for the 2 primary end points (Coropredict score and a 6-minute walk test) using a 2 (intervention vs control)×2 (baseline to 6 months) design, accounting for potential baseline differences and correlations between baseline and 6 months assessments. Data for the primary outcomes will be analyzed according to the intention-to-treat principle.

For the secondary outcome measures, a similar approach will be used. To explore long-term effects at 12 months, 2 (group)×3 (repeated measure) mixed models will be used. In addition, univariable and multivariable regression analyses (both linear and logistic regression) will be used to examine cross-sectional and longitudinal associations between variables, with the type of regression analysis depending on the distribution characteristics of the variables.

Exploratory analyses will also include completer analyses and evaluation of trial adherence as related to the outcome measures. As this study includes multiple secondary outcome measures without correction for statistical type 1 error, the obtained results will be interpreted with caution and will primarily provide estimates of effect sizes related to the TIMELY intervention. Data will be analyzed using SPSS statistics (IBM Corp) [48] or equivalent.

Lifestyle behaviors are a recognized risk factor and account for a significant proportion of the disease burden among people with CAD [2,21-23]. Eliminating or attenuating of these behaviors is warranted, as up to 80% of cardiovascular events can be prevented by adopting a healthy lifestyle [49,50]. However, changing and maintaining healthy behaviors is challenging, and only a subgroup of people tends to adhere to the guideline recommendations and succeed in long-term behavior change [51]. CR programs (provided online or face-to-face) are successful in changing health behaviors; however, maintaining the gains accomplished during the initial phases of CR remains challenging and requires additional support [26]. To maintain the benefits of CR, eHealth has been identified as a potentially effective approach. However, there are still gaps in the evidence supporting its use [52]. This mode of delivery provides personalized yet scalable care to those who need it. Hence, the TIMELY project aims to evaluate the effectiveness of the described patient-centered eHealth-based maintenance program, composed of an integrative care platform and an app-based, long-term health behavior change program. It is expected that supporting patients with CAD using technological innovations and applying patient-tailored and behavior-specific change techniques or domain-specific BCTs may result in significant and relevant improvements compared to care as usual. The TIMELY project goes beyond the current state-of-the-art by (1) using a highly personalized approach supported by AI technology; (2) providing a fully eHealth supported program; (3) including effective BCTs per target behavior; (4) focusing on both mental and physical factors; and (5) including the perspectives of all relevant stakeholders. In addition, evaluating cost-effectiveness is crucial for determining the economic value of the TIMELY project’s eHealth-based maintenance program. Integrating a cost-effectiveness analysis will measure not only the costs associated with implementing and using the integrative care platform and the app but also the economic benefits derived from improved health and reduced future events. This analysis will help justify the investment in these technologies by demonstrating that the proposed solutions are not only effective in terms of health outcomes but also economically efficient.

Given the involvement of 3 different recruitment centers with varying phase-2 CR procedures (inpatient vs outpatient CR), patient enrollment procedures might be slightly different between centers. While these differences are not expected to affect the results of the study, it is important to note that patients may be approached at different time points during or following their center-based CR. Furthermore, participants will receive several monitoring devices and extensive instructions on navigating the TIMELY app. This could be perceived by some patients as overwhelming and perhaps result in refusal or early termination of participation. Finally, as the TIMELY intervention will provide personalized advice based on incoming patient data, the intervention may be more effective for those providing data frequently.

The results of the TIMELY intervention will provide insights into the effectiveness of a fully web-based, personalized program to support long-term cardiac healthy behavior and reduce cardiac risk. The results will also provide further understanding of relevant factors contributing to successful health behavior change over time and those that help sustain these behaviors. In addition, the TIMELY project can provide new insights into which BCTs are associated with successful behavior change, helping to enhance the effectiveness of future interventions.