This qualitative implementation study examined the deployment of a multidevice home telemonitoring system for older adults with heart failure across three regional health organizations in Quebec. It aimed to explore feasibility, acceptability, and integration into nursing practice, as well as barriers, enablers, and contextual adaptation strategies related to workflow, digital tools, and long-term sustainability, primarily from the perspective of home care nurses, with complementary input from managers. Although the broader initiative included quantitative and qualitative components, this article focuses on the lived experience of implementation and does not report formal key performance indicators.

This study was conducted as part of the pilot project, implemented across three CISSS in Quebec. Sites were selected by the Quebec MSSS to reflect diverse home care delivery contexts (eg, rural vs urban). A multisite governance structure coordinated implementation, which was staggered but unified by shared protocols, training materials, and technical platforms. The pilot was mandated by the Direction générale des aînés et des proches aidants of the Quebec MSSS to generate real-world evidence on multidevice home telemonitoring with a view to a potential province-wide scale-up; in iCHECK-DH terms, the implementation is at the Piloting and Evidence Generation stage.

The intervention supported older adults with heart failure through remote monitoring of physiological and symptomatic data, caregiver engagement, and clinical follow-up by 21 home care nurses.

Managers from the three CISSS were involved in project development and local implementation planning. Each site determined how telemonitoring would be incorporated into nursing practice: two required nurse participation, whereas one used a voluntary model. Managers also identified eligible patients, defined as French-speaking adults aged 65 years or older receiving home care for heart failure and without neurocognitive disorder. Patients and caregivers who agreed to participate were then contacted by a research professional, who provided study information and obtained consent.

Finally, a total of 67 patients were initially recruited for the intervention. Each morning, patients received an alert to complete a symptom questionnaire, which they could complete by touch or voice. They also weighed themselves daily using the Bluetooth scale. The xPill adherence system tracked medication intake via pharmacy connectivity.

The telemonitoring solution relied on proprietary, closed-source platforms developed by Virtuose Technologies and DOmedic, which remained vendor-owned throughout the pilot. The two platforms were selected through a co-development partnership formalized at project inception—the founder of Virtuose Technologies was a project co-instigator alongside the academic lead—and equipment was acquired via direct contracts between each CISSS and the technology partners under the ministerial research mandate, without a public tender. At acquisition, the smartwatch was Health Canada-approved; the connected scale was not. The Virtuose dashboard operated independently from existing clinical information systems used in Quebec, including i-CLSC and SyMO, with no Fast Healthcare Interoperability Resources (FHIR)-based interoperability standard to support data exchange with other health information systems. Physiological data were transmitted via patients’ home networks to the Virtuose cloud; alerts were generated automatically based on preconfigured thresholds. Community pharmacists accessed xPill adherence data only through DOmedic’s proprietary interface. A privacy impact assessment and cybersecurity review were completed before deployment.

Nurses played a central role in operationalizing the telemonitoring model. They were responsible for:

Prior to deployment, nurses completed two training components: a self-paced clinical module on geriatric cardiac assessment (approximately 4 h across 10 videos), and a 2-hour hands-on session covering the digital devices and the Virtuose dashboard. Additional support was provided by trainers from Virtuose Technologies (dashboard) and DOmedic (xPill adherence system). Several nurses also acted as super-users or local champions during implementation.

This study draws on two categories of participants: (1) Home care nurses (n=19): recruited from the three participating sites, all nurses participated in post-implementation semi-structured interviews focused on their experience with the telemonitoring intervention; and (2) Managers and coordinators (n=11): interviews were conducted post-implementation to capture organizational perspectives on deployment outcomes, perceived impact on care delivery, and lessons learned for future scale-up. The average duration of interviews varied by participant type: between 20 and 30 minutes for home care nurses, and 60 and 75 minutes for managers and coordinators. All interviews were audio-recorded with consent and transcribed verbatim.

A directed thematic content analysis was conducted on the qualitative data obtained from the semi-structured interviews [23]. Interview recordings were transcribed with the aid of Whisper (OpenAI, local deployment), then checked against the audio by the research team before analysis. An initial coding framework, aligned with the study’s primary objective, focused on identifying barriers, enablers, and contextual strategies. The analysis was guided by sensitizing concepts from implementation science (eg, acceptability, feasibility, integration, and sustainability) and guided by the iCHECK-DH reporting framework [24]. The coding process was iterative and included constant comparison across participants, with attention to ensuring that insights reflected perspectives across all participating sites. Given the near-complete inclusion of eligible participants, data saturation was deemed to have been achieved when no new themes emerged in the final interviews across participant groups. This determination was confirmed through iterative team discussions. Codes relating to barriers, enablers, and contextual adaptation strategies were iteratively compared across participant groups and sites, then grouped into higher-order implementation themes through team discussion. Illustrative examples of the coding-to-theme process are provided in Multimedia Appendix 1.

This study received ethics approval from the Research Ethics Board of the CISSS de Chaudière-Appalaches, the lead institution for ethics review (MP-23-2023-1037), on July 11, 2023. The two additional participating institutions approved the project through institutional agreements recognizing the lead site’s review. All participants provided written or verbal informed consent, including permission for audio recording, anonymized data use, and secure data storage. Transcripts were deidentified, and data were stored on encrypted servers accessible only to the research team. Participants were assured that their decision to participate would not affect their employment or professional standing. No financial compensation was provided to participants.

The study was conducted in accordance with applicable Quebec provincial data protection requirements: Act respecting health and social services information, CQLR c R-22.1 [25]; Act respecting Access to documents held by public bodies and the Protection of personal information, CQLR c A-2.1 [26]; and Act to modernize legislative provisions as regards the protection of personal information, SQ 2021, c 25) [27]. A privacy impact assessment was completed to authorize limited data access for technology partners, and all digital tools underwent cybersecurity testing by institutional information security teams. External access to patient information, such as by community pharmacists, was permitted only with explicit patient consent. These safeguards ensured participant autonomy, data security, and institutional accountability throughout the project lifecycle.

Nurses held mixed views on the feasibility of telemonitoring. Some were enthusiastic, seeing it as aligned with evolving home care practices and useful for earlier detection and targeted interventions.

Others voiced reservations about added workload, the reliability of patient-device interaction, and their own comfort with digital tools.

In some cases, participation was perceived as encouraged institutionally but not entirely voluntary. These initial attitudes shifted over time, as many nurses reported that direct use of the system increased their confidence and perceived clinical value.

Nurses received both clinical and technical training prior to deployment, including a self-paced module and a hands-on session delivered by the technology partners. While this format was broadly appreciated, particularly the chance to interact directly with devices, many nurses found overlap between formats. Hands-on exposure was consistently seen as more effective than theoretical content.

In one organization, a supplemental virtual session on the xPill adherence system was well received and helped clarify a tool initially viewed as complex. Training aligned with early deployment allowed just-in-time troubleshooting, while delays led some nurses to forget key elements, highlighting the need for refresher resources.

Managers emphasized that beyond training format or timing, the deeper issue was limited digital readiness and cultural resistance. Many nurses struggled with basic tools like email or Teams, underscoring the need for foundational digital support alongside technical onboarding.

Nurses described varied routines for checking the Virtuose dashboard, depending on their workload, comfort with technology, and local workflows. In some sites, alerts were handled by a designated colleague; in others, each nurse monitored their own patients. This difference shaped how the system fit into daily practice. In centralized models, some nurses—especially those with only one enrolled patient—relied on others to manage alerts. While some found the system easy to use, others described it as just one more platform to manage alongside existing tools. They faced challenges moving between systems, repeating information, and trying to reconcile data that did not always match. For some, this added complexity led to overload and reduced confidence in their ability to track patient information accurately.

Alert management was a major operational challenge. Nurses received alerts for clinical thresholds (eg, weight, oxygen saturation), non-adherence, and connectivity issues. Clinical alerts were generally seen as more useful, but the high volume—especially for minor deviations or unfamiliar patients—often caused fatigue and disengagement.

Over time, troubleshooting was increasingly delegated to users or vendor support. Still, unresolved connectivity issues in rural areas often led to unnecessary home visits.

Nurses broadly agreed that telemonitoring enhanced their ability to detect subtle clinical changes and intervene earlier when necessary.

At the same time, many emphasized that raw data alone, particularly when detached from a personal clinical relationship, was insufficient for confident decision-making. Familiarity with the patient remained essential for interpreting alerts.

Although the intervention was framed as a potential efficiency tool, few nurses reported any reduction in home visit frequency. Most patients continued to need in-person care for other conditions, and existing visit schedules were already minimal. Several questioned the added value of intensive remote monitoring for clinically stable users.

Concerns about long-term sustainability were also prominent. Participants stressed the need for institutional ownership, clear governance, and integration into core workflows. Many viewed the ongoing reliance on vendors and ad hoc coordination as misaligned with the demands of routine public care delivery.

Professionals adapted pragmatically but pointed to structural limitations that must be addressed for telemonitoring to scale in public home care.

Several factors supported the successful onboarding of nursing staff. Training that combined self-paced video modules with in-person, hands-on sessions was generally effective, especially when timed with early patient deployments. Nurses valued real-time support during installations, often finding it more useful than theoretical preparation. This aligns with research emphasizing the importance of active facilitation and practical guidance in digital health adoption [29].

Engagement with the Virtuose dashboard varied, with some nurses deferring to centralized monitoring owing to low patient load or digital discomfort. Frustrations included redundant alerts, poor prioritization, and notification about unfamiliar patients, consistent with alert fatigue in clinical settings [30].

Technical challenges also limited integration. Frequent device disconnections, connectivity issues, and user adherence problems sometimes required in-person troubleshooting, undermining the model’s efficiency. Although nurses eventually redirected technical problems to vendor support, these disruptions eroded confidence in the system’s reliability, consistent with reported unintended consequences of telehealth deployment [31,32].

Telemonitoring was generally perceived as clinically useful, particularly for detecting weight fluctuations, respiratory symptoms, or oxygen desaturation associated with heart failure decompensation. Nurses appreciated the ability to intervene earlier and more selectively, especially for patients at higher risk. However, participants underscored that data alone was not sufficient for confident decision-making. When alerts involved unfamiliar patients or when values fell within chronic variability rather than true deterioration, nurses hesitated to act without contextual insight. In that sense, the technology served as a decision support tool rather than a decision-making system.

The intervention did not consistently reduce nurse home visits, as many patients required in-person care for other conditions and baseline schedules were already minimal. Nurses questioned the value of telemonitoring for clinically stable users, suggesting a need for clearer patient-selection criteria—findings consistent with other implementation studies [33].

Beyond workflow integration, the study revealed structural and governance-level issues affecting long-term sustainability. Nurses expressed concern about the system’s dependence on external vendors for technical support and updates, and the lack of integration between the telemonitoring dashboard and their primary clinical software, which fragmented workflows and increased cognitive load. These concerns reflect broader challenges in digital health sustainability, particularly in systems reliant on proprietary platforms without long-term procurement and governance strategies [34].

Budget planning and financial oversight were managed by the MSSS, which funded the project as part of a pilot initiative over an 18-month period. The budget covered device acquisition, software licensing, training, technical support, and coordination activities. Although detailed financial breakdowns were not shared with the research team, implementation partners indicated that vendor contracts and licensing fees represented a substantial portion of total costs. Equipment was acquired via direct contracts between each CISSS and the technology partners, and biomedical engineering departments in each CISSS handled device identification, calibration, maintenance, and end-of-pilot retrieval, reset, and secure storage for potential reuse.

The pilot was designed as an evidence-to-policy exercise rather than a vendor-led commercialization: findings and recommendations were submitted to the MSSS to inform a potential managed expansion across the Quebec network, with decisions on continuation, broader deployment, or hand-over remaining with the ministry at the time of writing.

Interoperability limitations further constrained implementation. The dashboard’s lack of integration with core clinical systems forced staff to navigate additional platforms, while the xPill adherence system required community pharmacists to subscribe to a proprietary interface, limiting participation and undermining the utility of adherence alerts. These challenges underscore the importance of interoperable design, alignment with existing infrastructure, and coordinated sector-wide implementation [34].

Taken together, several implementation lessons are broadly transferable to other public or community-based home care settings: the importance of hands-on training for variable digital readiness, alert-fatigue mitigation, and the continued centrality of relational clinical judgment. However, certain challenges—reliance on proprietary platforms without interoperability standards, pilot-based MSSS funding, and CISSS governance structures—are specific to Quebec’s public system. Successful scale-up elsewhere would require interoperable infrastructure, sustained institutional ownership beyond pilot phase, and alignment with local nursing workflows and capacity.

Trustworthiness was supported through inclusion of diverse perspectives across sites and roles, use of consistent interview guides, systematic and iterative coding, collaborative theme development, and grounding interpretations in participant quotations. Detailed contextual description supports transferability, and team-based analysis and reflexivity strengthened confirmability.

Despite the rigor of our approach, this study has certain limitations. First, it does not present cross-site comparisons. While this was a deliberate design choice, consistent with the collaborative and staggered nature of the implementation, and with the homogeneity of experiences reported across sites, it limits the ability to assess region-specific variations in implementation. Second, the analysis centers primarily on the experiences of home care nurses, who were the primary users of the telemonitoring tools. Although managerial and technical perspectives were included for context, this focus may exclude broader organizational dynamics. Third, the study captures early implementation experiences and does not evaluate long-term impacts on clinical outcomes or sustained integration into routine practice. Finally, given the qualitative design and limited number of participating regions and professionals, the findings are not statistically representative and should be interpreted as context-specific insights rather than generalizable conclusions.

The following conditions appear essential for health systems seeking to replicate this implementation (Table 2).