Heart failure is a complex health problem characterized by debilitating symptoms, which adversely affect morbidity [1,2]. Despite advances in medical, pharmacological, and surgical treatments, the global prevalence of the condition continues to rise, and hospital admissions remain high [3,4]. Heart failure significantly impacts patients’ quality of life (QoL) and contributes to increasing health care expenditure [5,6]. Evidence suggests that individual behavior change and community support can help reduce hospitalizations related to heart failure [7-9]. Self-management support has become a critical strategy to empower patients with heart failure to actively engage in and take responsibility for their health and care [2,8,10,11]. Patients are encouraged to adopt healthier lifestyles and strengthen self-management behaviors, including symptom monitoring, physical activities, fluid management, and treatment adherence [8,12].

Mobile health (mHealth) technologies, such as smartphone apps, internet-based tools, wearable devices, and other patient monitoring systems, are increasingly being used to support the self-management of patients with chronic cardiac diseases [13,14]. Early evidence from experimental studies and randomized controlled trials (RCTs) involving patients with heart failure suggests that mHealth interventions can improve disease knowledge, self-care, and QoL [15-19]. However, most of these RCTs have been primarily feasibility studies or small-scale trials, and the clinical, health, and behavioral outcomes reported have been inconsistent, highlighting the need for further investigation [18,19].

Research on the design of mHealth technologies for heart failure self-management is also evolving [16,18,20]. Existing mHealth apps vary widely in the features they offer, with some lacking key components recommended by clinical guidelines and patient empowerment strategies, such as symptom monitoring, physical activity support, and patient education [18,19]. Digital interventions based on mHealth technology need to be adapted to the local context and involve both patients and clinicians in the design process [21]. To enhance effectiveness and promote patient empowerment, these interventions need to be more tailored and to incorporate features associated with improved clinical and behavioral outcomes [18,22-24].

Patient engagement with mHealth interventions remains a challenge [23]. Understanding patients’ usage behavior, motivation, and preferences in response to the interventions is crucial for comprehending how patients with heart failure interact with the interventions and for assessing their level of acceptance [22,24,25]. Measures of engagement vary across studies of mHealth [20]. Combining system usage data with qualitative insights on engagement patterns and quantitative outcome assessments may provide a more comprehensive understanding of acceptability and impact [18,20]. Given the importance of patient preferences and sustained engagement, evaluation studies need to examine not only health outcomes but also influencing factors for engagement [22].

An mHealth platform was developed through a collaboration between the Commonwealth Scientific and Industrial Research Organization (CSIRO) and The Prince Charles Hospital (TPCH) in Australia to support patients with heart failure. The mHealth platform, MoTER-HF, was developed from a research prototype called MoTER (Mobile Technology–Enabled Rehabilitation), which was originally designed for home-based monitoring in cardiac rehabilitation [14]. The platform incorporated a smartphone app and a web-based clinical portal (Figure 1). The clinical portal enabled clinicians to create patient accounts and configure personalized health measurements and goals. Patient-reported health measurements, symptoms, and exercise data collected through the app were automatically uploaded to the portal, allowing clinicians to monitor the data. The app stored data locally and synchronized it with the portal whenever a data connection was available.

The initial MoTER-HF prototype was refined after being tested by patients and clinicians from TPCH to tailor the features to the needs of patients living with heart failure. During the user testing, 9 patients with a clinical diagnosis of heart failure installed and tested the app for 2 weeks, while 8 clinicians (2 cardiologists, 1 nurse practitioner, 1 clinical nurse consultant, 2 cardiac nurses, 1 physiotherapist, and 1 pharmacist) reviewed the app and tested the clinical portal in design-oriented workshop sessions. On the basis of their feedback regarding functionality and clinical relevance, the study team identified areas for improvement, addressed technical issues, and refined the app features. Key app refinements resulting from user testing included the development of the daily self-monitoring questions by clinicians in accordance with clinical guidelines [2], the addition of a text message on the Symptoms page prompting patients to seek medical care if symptoms deteriorated, revisions to the symptom list and exercise list for patients to check, and an updated set of web-based educational resources.

The final version of the MoTER-HF app (available on iOS and Android) included the following features (Textbox 1; see also Multimedia Appendix 1).

The platform was evaluated in a feasibility study involving a cohort of participants diagnosed with heart failure. The study aimed to assess the feasibility of the app among these patients and explore potential health outcomes associated with the app use.

The feasibility study used a single-group pretest-posttest mixed methods design. The study examined the feasibility of the app, including patients’ adherence, acceptability, and perceived usability, identified factors influencing patient engagement and acceptability, and explored potential health outcomes. Feasibility outcomes were assessed descriptively without predefined benchmarks or progression criteria and were treated as the primary outcomes of the study. The investigation of health and behavioral outcomes was exploratory and not an effectiveness evaluation of the intervention. The study followed the International Council for Harmonization’s Good Clinical Practice guidelines to ensure the protection of participants’ rights and provide assurance for data integrity.

Feasibility study participants were recruited from the TPCH Heart Failure Service between April 2024 and November 2024. Eligibility criteria included (1) adults (aged >18 y) with a primary or secondary diagnosis of heart failure or outpatients who had been hospitalized for heart failure in the last 12 months, (2) proficient in English, (3) had access to a smartphone compatible with the MoTER-HF app, (4) the ability to operate a smartphone for the purposes of the study (eg, no vision, hearing, and cognitive or dexterity impairments), and (5) access to mobile internet and/or Wi-Fi at home. Patients who were unable to attend TPCH Heart Failure Service or were involved in another intervention study were excluded.

A clinical nurse coordinator assisted participants with installing the MoTER-HF app, configuring the health measures to be tracked, and discussing suitable goals following consultation with the treating clinician during the onboarding visit. Participants were provided with 2 Bluetooth-enabled measurement devices (a blood pressure [BP] monitor and a digital weight scale) and used the app and devices over a 12-week study period. During this time, they were required to log Health Measures and Daily Diary responses via the app on a daily basis, log Symptoms and Exercises when applicable, and use the Education and Goals when needed. Training on using the app was provided by the nurse coordinator during participants’ onboarding visits. Participants also received a printed user manual outlining the app’s function, step-by-step instructions, and contact details (phone and email) of the hospital’s cardiac clinical research center for technical support or study-related questions.

A nurse practitioner, who was also an investigator of the study, discussed the app data with participants during in-person clinic consultations as part of routine care. Although clinicians had access to participant data at any time, continuous monitoring was not performed in the feasibility study due to workload constraints. Instead, the clinical nurse coordinator reviewed the data twice weekly. Thresholds for each measure were set up by the nurse coordinator in the clinical portal. Alerts for out-of-range values were flagged in the portal, and email notifications were automatically sent to the TPCH Heart Failure Service for the nurse coordinator to follow up by phone. The nurse coordinator also monitored participants’ adherence to app use and provided technical support to participants who experienced difficulties using the app.

The primary outcomes of this study were feasibility measures, which were assessed as adherence, usability, and acceptability (Table 1).

The feasibility measures were investigated through a satisfaction questionnaire, semistructured interviews, and MoTER-HF app usage logs. The satisfaction questionnaire (Multimedia Appendix 2) was developed by the research team based on the Diffusion of Innovation Theory [26] and our previous studies [27,28]. It includes engagement questions related to frequency of use, perceptions of each app function, and compatibility with participants’ daily routines. It also included questions to understand their empowerment experience, its observability, and their overall acceptance of the app. The questionnaire also included the System Usability Scale [29] to assess the subjective usability of the app and questions related to the frequency of use. It was administered after participants completed the 12-week study, within a 2-week poststudy window. App usage logs were exported from the platform server after study completion to assess adherence.

A purposively selected sample of participants, based on their demographic characteristics (eg, sex and age) and level of engagement with the app (eg, frequent or infrequent use according to app usage log), was invited to participate in the qualitative arm for an interview conducted upon their study completion and scheduled to coincide with a clinic visit. The interviews aimed to explore participants’ experiences and perceptions of using the MoTER-HF app. The interviews were conducted by 2 researchers (JL and AAB) via Microsoft Teams video conferencing. Participants attended the interviews from a private office during their end-of-study visits to the hospital, with video conferencing facilities arranged by the nurse coordinator. The 2 researchers conducted debrief discussions after each interview and collectively determined the data saturation. Each interview lasted approximately 30 minutes and was recorded using Microsoft Teams.

Health and behavioral outcomes were measured at baseline and at the 12-week time point, as an exploratory investigation. The following questionnaires were included:

Individual participants were clinically assessed by study clinicians at baseline and at the 12-week time point. Assessments included clinical examinations of height, weight, lying and standing BP, heart rate (HR), respiratory rate, and oxygen saturation, and related physical examinations. A demographic survey was completed by participants at baseline. Major Adverse Cardiovascular Event data were collected from hospital information systems by the clinical nurse coordinator at the 12-week time point.  

All questionnaires were paper-based and collected by the clinical nurse coordinator, who also entered the responses into the secure web-based REDCap platform hosted by the CSIRO.

Descriptive statistical analysis was conducted for the satisfaction questionnaire, as well as for app log data, using Microsoft Excel. Changes in health and behavioral outcomes from baseline to 12 weeks were assessed using a 2-tailed paired-samples t test in the R environment.

Interview audio recordings were transcribed by a professional transcript service. A thematic analysis was conducted by 2 researchers (JL and AAB) using NVivo software (version 14; Lumivero). Initial codes were developed based on the categories of engagement experience and perceived impact. Coding was cross-checked and discussed between the 2 researchers prior to the second round of coding and results synthesis. Member checking was not performed due to logistical difficulties in reaching participants after the study. Triangulation was used to integrate quantitative usage data (to quantify behavioral interaction patterns), questionnaire responses (to capture perceived engagement and satisfaction), and qualitative interview data (to contextualize behaviors and identify underlying mechanisms).

Ethical approval was granted by the Metro North Hospital and Health Service Human Research Ethics Committee (project ID: HREC/15/QPCH/256), and reciprocal approval was obtained from the CSIRO Health and Medical Human Research Ethics Committee (project ID: 2021_059_RR). Written informed consents were obtained from all participants. All questionnaire responses and platform log data were stored on secure servers of CSIRO and were identified only through study IDs. Interview transcripts were deidentified before analysis, and the results were reported in aggregate. Participants did not receive compensation for participating in the study.

A total of 26 eligible candidates were referred to the study team, of whom 23 (88.5%) completed the baseline assessment and were onboarded (Figure 2). The remaining 3 (11.5%) did not complete onboarding due to inadequate technical skills or failure to attend the onboarding visit. Among the 23 participants who installed the MoTER-HF app, 1 (4.3%) withdrew after 3 weeks due to being ineligible (no longer receiving care from TPCH Heart Failure Service), 1 (4.3%) required technical assistance and was unable to use the app at home, and 2 (8.7%) discontinued due to competing personal commitments. Participants’ baseline and end-of-study visits were aligned with their hospital clinical appointments for their convenience. However, difficulties in reaching some patients and coordinating with their clinic schedules at times affected the timeliness of follow-ups and questionnaire data collection. In total, 19 participants completed the study and used the app, and their available data were included in the data analyses. A total of 6 (31.6%) participants took part in the qualitative arm and were interviewed.

The participant cohort (N=19) was primarily men (n=14, 73.7%), with most participants aged 50 to 60 years (n=9, 47.4%) or 40 to 50 years (n=4, 21.1%). The majority (n=15, 78.9%) lived with a partner or children. Most participants had been diagnosed with heart failure within the past year (11/17, 64.7%) and were enrolled in a hospital-based heart failure program (12/19, 63.2%). Table 2 summarizes the key characteristics of the participant cohort. Among the 6 interviewed participants, 4 were men, and 2 were women, with 3 aged 50 to 60 years, 2 aged 60 to 70 years, and 1 aged <50 years.

The usability perception measured by the System Usability Scale was positive (n=12), with an average score of 69.38 (SD 22.39), corresponding to a grade of “good.” Seven (58.33%) participants scored usability above 68, giving the app a Good or Excellent usability rating. 

Interview participants also reported positive experiences with the app’s usability and user interface. They found the app easy to use, especially its simple design and straightforward functionality. Most participants described the app as user-friendly, requiring minimal instructions, and easy to navigate without external assistance:

To further explore which app features participants valued most, an open-ended question was included: “What did you like best about the app?” A total of 8 (66.7%) participants provided responses, with comments highlighting the aspects of the graphical displays for reviewing BP and weight data (n=2), ease of use (n=2), and Bluetooth integration (n=2). Other mentioned aspects included having all data consolidated in one app (n=1), the ability to share progress (n=1), the simplicity of daily recording (n=1), and reminders (n=1). The majority of the interview participants indicated that the daily input and graphic display of BP and weight data were particularly useful, as they allowed participants to visually track health measurements over time:

While results of usage log satisfaction questionnaire showed less frequent use for some functions (eg, Symptoms, Exercises, and Education; Table 3 and Figure 3), factors that limited their engagement were discussed in the interviews. These influencing factors included perceptions of limited need and relevance, stable health condition, personal preferences for recording information and learning, and low digital literacy.

There were no significant differences in health and behavioral outcome measures compared to baseline, although the “after” results showed improved scores in Self-Care of Heart Failure Index (confidence, maintenance, management, and perception); Depression, Anxiety, and Stress Scale (anxiety and stress); Kansas City Cardiomyopathy Questionnaire (overall, physical limitation, QoL, and symptom frequency); and Self-efficacy for Managing Chronic Disease Scale. Given the small sample size and short study period limitations, these score changes were observations only.

Physical examination records (n=7) showed no significant differences in BP and weight compared to baseline, although the readings were slightly higher at the week 12 follow-up. No Major Adverse Cardiovascular Event associated with participants was reported in the hospital information systems during the study period.

Summary of paired t tests for these validated outcome measure questionnaires and selected physical examination measures is presented in Table 4.

Despite the growing use of mHealth technologies to support heart failure self-management, challenges remain in understanding patient engagement, the most appropriate features, and how best to evaluate their outcomes and benefits [16,18,20]. This study examined the feasibility of an mHealth intervention developed to support self-management in patients with heart failure. A key strength of this study is its mixed methods design, which provides a comprehensive understanding of patient engagement, user experiences with specific app features, factors impacting engagement, and preliminary potential health outcomes. By analyzing system usage logs, self-reported user experiences from questionnaires, perceptions from in-depth interviews, and health outcome measures, the study offers valuable insights into the app usage patterns, acceptability, facilitators of and barriers to engagement, and the potential impact on patients’ health and self-care behavior. Furthermore, the findings of this study contribute to mHealth design research by demonstrating that feature uptake is driven less by availability or usability and more by alignment with existing care practices, perceived need, and integration within clinical ecosystems.

The study examined engagement patterns associated with individual app features, factors influencing use, and patients’ overall acceptance of the mHealth technology. Research in the literature highlights the importance of unpacking engagement with individual components of mHealth intervention to avoid the “black box” issue of unclear user engagement and effectiveness elements [18]. Consistent with this perspective, our study findings showed that engagement was selective rather than uniform across app functions, offering important insights for the design of mHealth interventions in this patient population.

Participants reported satisfaction and perceived improved self-monitoring practices, with most finding the app easy to integrate into their daily routines and aligned with their preferred self-management practices. Engagement was notably high with core monitoring features, such as BP and weight tracking, due to the alignment with their existing monitoring practices, the Bluetooth-enabled automatic data entry, and the perceived value of visualizing historical data. These findings show that features that seamlessly extend established routines and provide monitoring value are more likely to be sustained over time.

Engagement with other features, such as symptom reporting, exercise tracking, goal checking, and education materials, was lower. Our interview findings suggest that this did not necessarily reflect poor usability but rather limited perceived relevance for participants. Participants described stable health conditions, long-standing self-management experience, or ongoing clinical contact, which reduced the perceived added value of symptom logging, goal review, or educational materials. Exercise tracking was viewed by some participants as redundant with existing rehabilitation programs, which were not digitally integrated with the app, while educational content competed with preferences for direct clinical advice or alternative learning modes.

Our findings suggest that acceptability and adherence to mHealth interventions are maximized when features are aligned with patients’ clinical needs, existing care structure, and personal preferences. Features such as education, exercise tracking, and goal setting may require greater personalization and closer integration with their existing care to enhance perceived value. Studies of mHealth adoption in patients with heart failure have reported person-related, technology-related, and contextual-related engagement barriers [24]. Our findings extend existing evidence and highlight the need for tailoring both content and functionality to the specific circumstances of patients.

From a design perspective, our results align with a growing body of research in mHealth development that emphasizes the value of tailored design and appropriate evidence-based information in a way that is relevant to patients, particularly for older adults and those who require self-management support [23,35]. The initial user testing of the prototype prior to the feasibility study exemplified these principles by incorporating clinician-developed Daily Diary questions and Symptoms monitoring list, as well as clinician-customized daily health measure items. The app provided flexible data entry options of both automatic Bluetooth integration and manual input for smooth use. Participants expressed preferences for varying daily reminder notifications. Some preferred alternative formats for materials in the Education feature of the app, such as audio-based content, which may be a consideration for improving the app design. These findings reinforce the importance of personalization in mHealth design [23,24,35]. Furthermore, technical issues were barriers for some participants, indicating the need for robust systems and user-friendly features to support older patients or those with limited digital experience [23,35].

The implementation of mHealth in real-world settings requires more than just technical design. It also needs effective integration with ongoing clinician follow-ups and existing care programs [22,36]. Our study found that participants felt supported by the study clinicians who could access their data and valued the discussions about their data during face-to-face consultations. However, limited integration with rehabilitation programs and communication pathways reduced engagement with some app features. While some of these limitations may be attributed to the feasibility nature of the study, our findings echo previous research and suggest that incorporating the app data review in outpatient consultations, aligning with patients’ health care programs, and enhancing clinician integration could encourage more consistent use and better support for heart failure self-management [35,37]. Although clinician oversight was identified as a perceived benefit, its sustainability is contingent on balancing clinical value with workload demands when integrating mHealth solutions into clinical workflow [22,36]. Continuous monitoring was not performed in this study due to the limited availability of study nurses and the additional workload it would have had on their existing clinical practices. Technical training and troubleshooting also created additional tasks for study nurses. Furthermore, while workload was not formally measured, the process of reviewing app data and responding to alerts suggests a substantial time commitment that may increase with larger cohorts. Scaling up digital health solutions will require addressing barriers at individual, organizational, policy, and technological levels [36,38]. In larger populations, manual review of patient data and alerts may become unsustainable without automation, prioritization algorithms, and redistribution of responsibilities within care teams. Strengthening system functionality and interoperability will be crucial for transitioning to a scalable digital model of care. Effective implementation will also require strong leadership, supportive regulatory and reimbursement frameworks, and stakeholder cocreation [36,38]. These implementation challenges and strategies will need to be further explored in future studies.

The study also explored changes in health and behavioral outcomes using validated measures and qualitative data. While no statistically significant improvements were observed in self-care, QoL, heart failure knowledge, self-efficacy, and psychosocial measures, exploratory patterns of change consistent with improvement on several subdomains (eg, confidence, maintenance, management, and perception of self-care) were observed. Due to the small sample size and relatively short study period, these changes should be interpreted as observations for guiding future rather than evidence of efficacy. Participants described perceived benefits in managing their condition, as reflected in the satisfaction questionnaire and follow-up interviews, where participants valued improvements in self-monitoring, a better understanding of their health measures, and the sense of reassurance provided through study nurses’ monitoring and discussions, although these perceptions do not imply corresponding measurable changes in health and behavioral outcomes.

In line with recommendations from the broader mHealth literature, longer-term RCTs with larger and more diverse cohorts will be needed to determine whether mHealth platforms, such as MoTER-HF, can produce meaningful improvements in health and behavioral outcomes. Our study results can serve as an important foundation for future studies by providing useful insights into the selection of appropriate outcome measures and the sample size calculations for future RCTs.

The feasibility study had some limitations. The platform used was a research prototype, and troubleshooting during the feasibility study presented challenges. Technical issues, such as problems with Bluetooth device pairing and connectivity, affected participant engagement. Practical barriers, such as scheduling difficulties for participant onboarding and offboarding at their clinic visits and limited time available for the clinical nurse coordinator to follow up with participants, also contributed to missing questionnaire data at the 2 time points. Another limitation was the relatively small number of participants, with participants who were recruited from a single hospital, which might limit the generalizability of the findings.

Future work will focus on gaining a more in-depth understanding of clinicians’ workload and resource requirements, which were not captured in the feasibility study, as only one nurse practitioner and one clinical nurse coordinator who were part of the study team were directly involved in managing the portal and reviewing app data and alerts. Furthermore, following this feasibility study, a larger-scale study within routine clinical workflows will be necessary to evaluate the efficacy and cost-effectiveness of mHealth for heart failure management, as well as to assess the correlations between patient engagement and health outcomes. These efforts will inform the development of a scalable, evidence-based digital health program that can be seamlessly integrated into clinical practices, ensuring its effectiveness and sustainability in real-world settings.

Overall, the feasibility study demonstrated the potential of MoTER-HF to support self-management in patients with heart failure. Most participants actively engaged with the app for health monitoring and reported positive experiences of usefulness and usability, although no statistically significant improvements in health and behavior measures were observed. The study identified several factors influencing participant engagement and provided insights for the future technical design and implementation. Given that this was a feasibility study, further evidence-based research is needed to investigate its efficacy and to guide the development of an integrated mHealth program and a digital model of care.