The FPT consist of three phases: (1) titration of medicine (0-3 months), (2) participation in the FPT at a health care center or call center (3 months), and (3) follow-up with rehabilitation in everyday life (6 months). The three phases of the telerehabilitation program are illustrated in Figure 1.

The HeartPortal is a digital toolbox that functions as an interactive learning module; the setup is illustrated in Figure 2, and screen captures are shown in Figure 3. The portal has been developed using a participatory design approach [19,21,22]. The HeartPortal consists of four elements:

As part of the telerehabilitation program, there is continuous monitoring of the patient’s vital signs using technology-enabled devices/equipment. A list of the equipment used is provided below:

Figure 2 shows how data are presented in the HeartPortal. All the measured values are stored on a secure database at Aalborg University.

The patients receive instructions on evaluating their own values and how to react if these values are abnormal. A telerehabilitation coordinator observes all the measured values as well, and during phase 1, the coordinator also acts as a coach, helping patients learn to observe and act on any abnormal readings or worsening of symptoms to avoid possible hospitalization. During phase II, health care professionals also view the data. All health care professionals from the hospital and health care centers have access to the data.

There is a strong focus on data and network security in the Future Patient project due to the project’s complexity and the environment where data are generated from both the health system (pulled data) and the patients (pushed data). Based on different endpoints in the setup (patient’s home, workplace, health care center, etc), the technology used, and the classification of information, the formation of metadata will be addressed via reliable integration encryption and infrastructure reliability.

The target group in this study consists of patients diagnosed with heart failure according to the New York Heart Association (NYHA) class I-IV [27]. Patients with heart failure are recruited from the Cardiology Wards at hospitals in Skive, Viborg, Silkeborg, and Randers, all situated in the central region of Jutland, Denmark. The inclusion criteria are that the patient must have had a heart failure–related hospitalization within the past 2 weeks; have a stable internet connection at home; and have the ability to sign an informed consent form. Patients may have a pacemaker, and a maximum of 20% of the target group may belong to NYHA class I. Exclusion criteria are a revascularization or open-heart surgery within 3 months prior to inclusion; presence of previous neurologic, musculoskeletal, or cognitive disability or active psychiatric history (as noted in the medical record) other than depression or anxiety related to cardiac or other chronic illness; and absence of basic Danish language skills. All data from the recruitment process and patient withdrawal or dropout will be documented in a consort diagram.

The FPT will be tested in a multicenter randomized controlled trial (RCT) using a mixed method [28] approach. The intervention group will participate in the FPT, while the control group will follow a traditional rehabilitation program at the health care center, which is structured according to the national cardiac rehabilitation guidelines [29]. Enrollment of patients began in December 2016, and the RCT will end in September 2019.

The sample size was defined based on a power calculation. The aim of this project is to increase the quality of life (QoL) using telerehabilitation compared to usual rehabilitation. Based on the official KCCQ guidelines [25], the power calculation estimated the number of participants needed to show an increase in QoL on a scale of 0-100. When using the KCCQ, a “moderate” improvement is equal to a 10-point increase. The target group in the study includes patients with heart failure of a severe degree. As such, a 10-point improvement in the KCCQ will denote a change in the quality of life for the individual patient. Assuming normal distributions and a power of 80%, the number of participants in each group should be 63. However, accounting for 10% dropout, the total number of patients needed for this study is 140. Hence, we plan to enroll 70 patients with heart failure in the intervention group and 70 patients with heart failure in the control group.

The FTP program was designed using the self-determination theory (SDT) as a theoretical starting point. With the goal of increasing the effectiveness of behavior change interventions [30], SDT offers a framework for understanding the role of individualization in motivation. In particular, SDT highlights how fulfillment of basic needs such as autonomy, competency, and relatedness are necessary for initiating and maintaining changes in lifestyle and health behavior over time [31]. If motivation is sustained only by external factors, the patient will not have sufficient incentive to self-regulate and independently maintain a healthy lifestyle. However, if the patient finds that their health behavior goals are in accordance with their own internal values and beliefs (autonomy), the patient will feel that he/she possesses sufficient knowledge and skills to achieve this health behavior successfully (competency) and will feel supported by others (relatedness). The patient will, in turn, experience intrinsic motivation, that is, motivation based on internal factors enabling the patient to maintain the required lifestyle and health behavior over time [31].

The Future Patient project has been approved by the Regional Ethics Committee (N-20160055) and the Danish Data Protection Agency. The study is listed in listed in ClinicalTrials.gov (NCT03388918). The study is being carried out in accordance with the Helsinki Declaration, and all participants have signed an informed consent form prior to enrollment in the study.

Baseline data on demographics, clinical status, primary and secondary diagnosis, NYHA class, and actual prescribed medicines are being collected for both the intervention and control groups (Table 1). Questions on smoking, alcohol consumption, physical activity, IT competences, and rehabilitation will be asked in both groups in the RCT as well.

Primary and secondary outcome measures as well as the dates when they the data will be collected are shown in Table 1. The data collection process is described below.

QoL is the primary outcome, which is measured using the KCCQ [25]. KCCQ is a 23-item self-administered questionnaire and includes different clinical domains such as physical limitations, symptoms, self-efficacy, social interaction, and quality of life. The score is calculated by assigning an ordinal value to each response, beginning with 1, and then adding up to make up a scaled score for each domain. Missing responses are assigned a value corresponding to an average of the answered items within the domain. Scale scores are transformed into a 0-100 range. In the intervention group, the QoL is measured at baseline and then every second week throughout the duration of the program. In the control group, QoL is measured at baseline, 6 months, and 12 months.

All measured clinical data including weight, blood pressure, pulse (day/night), steps, sleep, and respiration in the intervention group will be collected and analyzed.

Changes in illness perceptions are being evaluated for both groups using the Brief Illness Perception Questionnaire (Brief IPQ) [32]. The Brief IPQ is a nine-item scale questionnaire designed to rapidly assess the cognitive and emotional representations of illness. Five of the items assess cognitive illness representation, two items assess emotional representation, and one item assesses illness comprehension. The last item is a causal question that asks the patient to list the three most causal factors in their illness. All items except one, the causal question, are scored on a scale from 0 to 10. The measure of the Brief IPQ is compared between groups at baseline, 6 months, and 12 months.

Changes in the type of motivation are measured using the Health Climate Change Questionnaire (HCCQ) [33]. The HCCQ is a 15-item questionnaire that assesses patients’ perceptions of the degree to which their health care providers are supportive of their autonomy. Respondents rate each statement on a seven-point Likert scale (from “Not at all true” to “Very true”). The total score is calculated by averaging the values of each item. The HCCQ can be used in its full version (15-items) or short version (6-items). In this study, the full version is used. The scores from HCCQ are compared between groups at baseline, 6 months, and 12 months.

Symptoms of anxiety and depression are measured using the Hospital Anxiety and Depression Scale (HADS) questionnaire [34]. HADS is a 14-item self-reported questionnaire consisting of two 7-item subscales measuring anxiety and depressive symptoms (ie, it does not assess somatic symptoms). Each item is scored on a four-point Likert Scale from 0 to 3. Scores are tabulated for each subscale, resulting in a score of 0-21, with higher scores indicating higher levels of anxiety or depressive symptoms. The measures from HADS are compared between groups at baseline, 6 months, and 12 months.

Patients’ health and electronic health (eHealth) literacy skills are measured using a Danish validated version of the eHealth literacy assessment toolkit (eHLA) [35] and the eHealth Literacy Questionnaire (eHLQ) [36]. The eHLA is a toolkit used to assess health literacy and digital literacy. The toolkits consist of seven tools, four health related and three digitally related, that are either self-reported, such as questionnaires, or a performance test of relevant skills [35]. The eHLQ is a 35-item 7-scale questionnaire used to evaluate and understand how patients interact with digital health services [36]. The measurement of health and eHealth literacy skills is compared between groups at baseline, 6 months, and 12 months.

Qualitative exploration of patients’, their relatives’, and health care professionals’ experiences using the HeartPortal will be conducted. Semistructured interviews inspired by Brinkman and Kvale [37] will be conducted after 6 months and 12 months. To analyze which parts of the HeartPortal are being used and for how long, time log files for login/logout of patients and relatives will be analyzed. The patients and relatives will be asked for their consent to extract their log files from the database for analysis.

A cost-effectiveness analysis will be based on the guidelines for economic evaluation by Drummond et al [38]. Estimates of the mean costs per patient will be made with a broad societal perspective, including use of resources for patients, hospital, and municipality. Estimation of costs will be based on data at the patient level for all patients, both at the intervention and control groups.

All demographic data are being stored in the Future Patient database (FPD). Data from questionnaires are stored in Research Electronic Data Capture software [computer software] (Nashville, TN: Vanderbilt University) and used for quantitative analysis.

All adverse events, including deaths, dropouts, or withdrawals from the study will be recorded and documented. If the patients no longer want to participate in the study, they can withdraw their consent at any time, and the reason for withdrawal will be documented. In this case, the project team will collect the equipment upon request. Patients who do not participate actively in the intervention will still be included in the study and analyzed according to the intention-to-treat approach. These patients will be allowed to retain use of the project equipment for as long as they want. Technical problems with the equipment will be recorded and documented.

At baseline, descriptive statistics [39] will be reported as median with 25th-75th percentiles in the case of skewed distribution or mean (SD) for normally distributed continuous variables. The nonparametric Kolmogorov-Smirnov test will be used to investigate normality of the distribution. An intention-to-treat analysis on all the randomized subjects will be conducted to provide unbiased comparisons between the intervention and control groups in order to avoid the effects of dropout.

The analysis of change in QoL, anxiety and depression, self-determination, illness perception, type of motivation, and health and eHealth literacy tests will be performed using data acquired through questionnaires. Comparison of incidence rates will be used to investigate the differences between the intervention and control groups. Two-sided tests and a significance level of 0.05 will be used.

All events from the day after randomization to patient exit will be included, while any other evident outcomes will be measured as changes from baseline to all assessment time points. Changes in the other evident outcomes will be tested using linear mixed models. Linear mixed models allow repeated measures to be collected in a longitudinal design and are superior when dealing with dropouts compared to other methods used for repeated measures. Hence, it will not be necessary to use imputation techniques on missing data. Statistical analyses will be performed using SPSS [computer software] (version 25.0. Armonk, NJ: IBM Corp), and values of P<.05 will be considered significant for all tests.

All interviews will be transcribed into text files by a research assistant. The data will be coded in NVivo [computer software] (version 12.0. Melbourne, Victoria, Australia: QSR International; 2018), with inspiration from the SDT theoretical framework and based on methods developed by Brinkman and Kvale [37]. Two researchers will conduct the interviews and analysis of the data. The date will be presented in themes and findings.

The aim of the Future Patient study is to test, implement, and evaluate a telerehabilitation program for patients with heart failure. The RCT was divided into three phases in the intervention group: a posthospitalization phase, a rehabilitation phase, and daily life adjustment 6 months after rehabilitation. A more holistic approach to patient rehabilitation is introduced on the basis of user-driven innovation, to accommodate the needs of the end user. This was just the first step toward engaging the patient in self-management. The uniqueness of this study lies in two areas: the use of a variety of self-monitoring methods and the use of SDT within a telerehabilitation context. The measurements include activity, pulse, and sleep trackers, which automatically transmit the monitored data to the HeartPortal. The HeartPortal has the potential to facilitate and educate patients in self-managing their disease. In this way, patients can become more engaged in their disease progression. The patient has the option of sharing their data, thus providing the patient’s relatives the opportunity to monitor the progress of the patient through the HeartPortal, which, in turn, enables the relatives to motivate and support the patient according to their progression. The PRO data is a new tool that we will be tested, and we have not identified other studies that are working with PRO data for patients with heart failure.

The expected benefit of the health economy evaluation for the telerehabilitation group compared with the control group is increased quality of life, reduced contacts to outpatient clinics at the hospital, and a reduction in admissions to hospitals.

The HeartPortal is also designed to enable the patients to improve their eHealth literacy through a combined information platform. In addition, a communication platform has been incorporated into the HeartPortal, from which the patient and his/her relative can communicate with health care professionals, psychologists, and technical personnel. As such, the relative is an important member in this study, as studies have found that support from family and friends benefits patients with heart failure by improving patients’ quality of life and reducing rehospitalizations [40]. Additionally, support from family and friends may promote self-management behaviors in the patient with heart failure, facilitating increased medication and dietary adherence [40].

This study had three limitations that should be considered. First, a subject group consisting of patients with heart failure is highly sensitive to death, as this group is dominated by the elderly [40]. Second, the intervention period extends to 1 year and is therefore highly dependent on the sustained motivation of the subjects. Third, as the tracking devices used in the study are commercial products purchased for the purpose of automatically transmitting data to the HeartPortal, the study is highly dependent on the data transmission and access hereof. If the regulations for the devices change during the study period, it would be necessary to replace them with other devices, which may lead to inconsistency in biases.

The expected outcomes are increased quality of life, increased motivation and illness perception, reduced anxiety and depressions, improved eHealth literacy, and health economics benefits. We expect the study to have a clinical impact for future telerehabilitation of patients with heart failure.