key: cord-0020842-p8guppl5
authors: Gordon, Kayleigh; Dainty, Katie N; Steele Gray, Carolyn; DeLacy, Jane; Shah, Amika; Resnick, Myles; Seto, Emily
title: Experiences of Complex Patients With Telemonitoring in a Nurse-Led Model of Care: Multimethod Feasibility Study
date: 2020-09-29
journal: JMIR Nurs
DOI: 10.2196/22118
sha: 8a451c454b75b0c0fdd3444d8c45aeae98d68aac
doc_id: 20842
cord_uid: p8guppl5

BACKGROUND: Telemonitoring (TM) interventions have been designed to support care delivery and engage patients in their care at home, but little research exists on TM of complex chronic conditions (CCCs). Given the growing prevalence of complex patients, an evaluation of multi-condition TM is needed to expand TM interventions and tailor opportunities to manage complex chronic care needs. OBJECTIVE: This study aims to evaluate the feasibility and patients’ perceived usefulness of a multi-condition TM platform in a nurse-led model of care. METHODS: A pragmatic, multimethod feasibility study was conducted with patients with heart failure (HF), hypertension (HTN), and/or diabetes. Patients were asked to take physiological readings at home via a smartphone-based TM app for 6 months. The recommended frequency of taking readings was dependent on the condition, and adherence data were obtained through the TM system database. Patient questionnaires were administered, and patient interviews were conducted at the end of the study. An inductive analysis was performed, and codes were then mapped to the normalization process theory and Implementation Outcomes constructs by Proctor. RESULTS: In total, 26 participants were recruited, 17 of whom used the TM app for 6 months. Qualitative interviews were conducted with 14 patients, and 8 patients were interviewed with their informal caregiver present. Patient adherence was high, with patients with HF taking readings on average 76.6% (141/184) of the days they were asked to use the system and patients with diabetes taking readings on average 72% (19/26) of the days. The HTN adherence rate was 55% (29/52) of the days they were asked to use the system. The qualitative findings of the patient experience can be grouped into 4 main themes and 13 subthemes. The main themes were (1) making sense of the purpose of TM, (2) engaging and investing in TM, (3) implementing and adopting TM, and (4) perceived usefulness and the perceived benefits of TM in CCCs. CONCLUSIONS: Multi-condition TM in nurse-led care was found to be feasible and was perceived as useful. Patients accepted and adopted the technology by demonstrating a moderate to high level of adherence across conditions. These results demonstrate how TM can address the needs of patients with CCCs through virtual TM assessments in a nurse-led care model by supporting patient self-care and keeping patients connected to their clinical team.

Multimorbidity, defined as the presence of more than one chronic condition in an individual [1] , has been increasing, particularly in patients with ≥4 to 5 conditions [2] . The associated risks of patients with chronic conditions include frequent hospitalizations, which can lead to increased risks of hospital-acquired infections and longer length of stay [3] [4] [5] . Previous qualitative research highlights numerous struggles patients experience in trying to manage their conditions [6] , including a high treatment burden in self-managing their care, attending multiple appointments, polypharmacy, and adhering to complex care regimens [7, 8] . As defined, multimorbidity does not account for common dispositions, such as frailty or aging, which may contribute to ineffective management and unanticipated outcomes. Patients with complex chronic conditions (CCCs) include both those with multimorbidity who also face complex social challenges such as socioeconomic vulnerability [9] . The layers of physical, mental, and social conditions cause extensive clinical variability in this patient population.

Several studies have found that nurse-led models enable patients to spend more time with clinicians and coordinate care under a more comprehensive and holistic approach, rather than just a single-disease focus [10] [11] [12] . Previous studies have also concluded that nurse-led models of care are feasible to deliver comprehensive chronic disease management [10, [12] [13] [14] [15] due in part to the broad scope of nursing practice, holistic approach, and interprofessional team of providers among other factors. A recent study focusing on patients with CCCs demonstrated that nurse practitioner (NP)-led care models provide sufficient quality and competency in diabetes and multimorbidity care [16] . Other studies have also found that nurse-led models of care are a potential mechanism to serve chronic populations, particularly when technology can complement patient self-care at home [15, 17] .

Several systematic reviews have shown that the use of telemonitoring (TM) can lead to improved health outcomes [18] [19] [20] [21] [22] [23] and reduced hospital readmissions [18, [24] [25] [26] . TM has also been shown to reduce all-cause mortality from heart failure (HF) [19, 22, 23, 27] , improve hemoglobin A 1c in patients with diabetes mellitus (DM) [28] [29] [30] [31] , improve blood pressure (BP) for patients with hypertension (HTN) [28, 32, 33] , and reduce respiratory exacerbations in patients with chronic obstructive pulmonary disease [24] . Studies also indicate that TM can improve shared decision making [34, 35] and patient experience with care [15, 36] . However, several large TM trials have also reported mixed results [23, 31, [37] [38] [39] [40] . It is possible that inconsistent findings are not only because of the technology itself but also because of the specific chronic conditions, combination of conditions, or lack of conditions targeted in the research. The model of care delivery in which the technology is implemented may also influence overall adherence. In particular, TM for patients with CCCs has not been widely studied in nurse-led models of care.

To address this gap, we evaluated the feasibility and perceived usefulness of the Medly TM system in a nurse-led care model for patients with CCCs. Feasibility is defined as the extent to which a new innovation can be successfully used or carried out within a given agency or setting [41] . The results of the implementation from the perspectives of the care team will be presented elsewhere. To the best of our knowledge, this is the first implementation of a TM system specifically targeting patients with CCCs within a nurse-led model of care. Our study was guided by the following central research question: What is the feasibility and perceived usefulness of a multi-condition TM platform in an integrated nurse-led care model?

A pragmatic, multimethod 6-month feasibility study was conducted for patients with CCCs in a nurse-led care clinic in Southern Ontario. The needs of patients and their families were identified in a previous qualitative study that informed the clinic's ongoing optimization [15] . Referral criteria to the NP-led CCC clinic included patients with multimorbidity; at least one hospitalization or more than two emergency visits within the last 6 months; and a length of stay, acuity of admission, comorbidities, and emergency department visits score >5 of a total of 14 [42] . The TM feasibility study commenced approximately 12 months after the clinic launch. Patients with HF, HTN, and/or DM involved in the study were asked to take frequent physiological readings at home via Medly, as per 

A pragmatic worldview was taken to evaluate the social construction of feasibility in multi-condition TM using the normalization process theory (NPT) by May [43] [44] [45] [46] and the Framework of Implementation Outcomes (IOs) by Proctor et al [41] . The relationship between the 4 constructs of NPT (coherence, cognitive participation, collective action, and reflective monitoring) was evaluated using a multimethods approach in the context of the patient's experiences of using the TM platform in the nurse-led care model. NPT was used to map the implementation process of multi-condition TM in patients with CCCs from the patient's perspective (Textbox 1), whereas the IOs were used to determine the success or failure of the implementation using the constructs of acceptability, appropriateness, adoption, and fidelity (Textbox 2). The outcome constructs of cost, penetration, and sustainability were not specifically assessed as this study focused on the evaluation of feasibility through the patient experience and not on the model of care delivery (ie, service outcomes) or health outcomes (ie, client outcomes). Figure 1 ). 

Between May and July 2019, patients were recruited through convenience sampling during their routine visit to the nurse-led clinic. Researchers did not specifically identify patients with a specific combination of conditions, as the intent was an evaluation of feasibility. A sample of 15 to 25 participants based on specific inclusion criteria was deemed to provide sufficient data to explore the feasibility of TM in a nurse-led clinic, including reaching saturation for the patient interview data [50, 51] .

Patients were eligible to participate if they (1) were aged at least 18 years; (2) were able to take home measurements as intended;

(3) were diagnosed with HF, HTN, or DM and one additional chronic condition; and (4) could understand English. Patients were excluded if they had visual or cognitive impairments that would prevent them from using the system or if they were already enrolled in a TM program. The NP identified which of the 3 conditions (HF, HTN, and/or DM) or a combination of conditions should be monitored for each participant. On written consent, the patient was provided with the necessary equipment and given an in-person training session on how to use the TM equipment. The NP set the specific algorithm ranges for each patient through a clinical dashboard. Patients were followed up with the Medly system for 6 months.

Patient adherence to taking readings was quantitatively assessed by analyzing the server logs of the TM system. Adherence was used as a measure to indicate TM's acceptability in this context and as an indicator of the perceived usefulness of TM. In this case, adherence was used as a pertinent implementation factor to determine feasibility [41] and was evaluated by measuring the frequency of TM readings for each condition (Table 1 ). For HF, as patients were instructed to take daily readings, adherence was measured as the percentage of days readings were taken over 6 months. For HTN, as per the algorithm, adherence was measured as the percentage of time patients took 2 consecutive BP readings at least once every 2 weeks. For DM, adherence was defined by taking readings at least once per week during the 6-month study.

Questionnaires were administered at the start and end of the study. Questionnaires included 3 sections: (1) the 36-Item Short Form Survey (SF-36) was used as a baseline indicator to evaluate the overall health status [52] of our participant group, (2) a series of technology questions were utilized to better understand the demographics of whom multi-condition TM is feasible within a nurse-led model, and (3) the participants' level of comfort with technology. Determining statistical significance in a small feasibility study was not the goal of including the SF-36 tool but rather to contextualize patient health status data in this care model. A chart review was conducted to collect basic demographic information (ie, age, sex, marital status, education level, etc). Sociodemographic data and clinical histories for all participants were summarized using a descriptive analysis.

All participants were invited to participate in a poststudy interview, each lasting between 30 and 60 min. Interviews were conducted onsite or over the telephone upon study completion. A semistructured interview guide was developed, informed by the 4 constructs of NPT (Multimedia Appendix 1) [43, 44, 53] .

Participants were asked to describe their experiences and perspectives on using the TM system. Caregivers were not asked any specific questions during the interview but were able to participate in the discussion. All interviews were audio taped and transcribed verbatim for analysis.

An interpretative descriptive approach was used to guide the qualitative analysis [54, 55] , as this approach examines a clinical phenomenon with the goal of identifying themes and patterns among subjective perspectives while also accounting for variations between individuals [56] . The process of inductive thematic analysis followed the method outlined by Thorne et al [54] , starting with an initial reading of each transcript. Subsequent readings included coding salient ideas and inductively deriving conceptual themes, first within and subsequently across transcripts. Detailed notes and codes were grouped according to emerging themes. The inductive process was conducted independently by 2 researchers (KG and AS) who came together to discuss emerging codes. Disagreements were discussed together and with the larger research team to reach consensus. Together, inductive codes were then compared with the implementation process constructs (NPT [43, 44, 46] ) and the outcomes of patient experience (IOs by Proctor et al [41] ) to structure the evaluation of the feasibility of this care model [41] . Saturation occurred at a point at which the codebook stabilized, a consensus was reached between the 2 researchers, and no additional codes or themes were identified in the data [57, 58] .

In total, 26 participants identified by the team agreed to participate. Of the 26 participants, 18 were put on HF monitoring, 1 on DM, 1 on HTN, 5 on HF+DM, and 1 on HTN+DM. The remaining 8 patients were removed from the TM system for several reasons, including winter relocation (n=1), rehabilitation admission (n=1), unanticipated language barrier (n=1), inability to use a smartphone touchscreen (n=4), and death (n=1). Table 2 presents the characteristics of the participants who completed the study. Overall, 65% (17/26) of the participants used the system for the full 6 months. The age of the participants ranged from 44 to 91 years, with a mean age of 73.8 (SD 13.3) years. All the study participants had more than 3 chronic conditions, with some participants having more than 8 conditions in total. The total number of medications ranged from 8 to 19 per participant. This feasibility study was conducted in a geographic area that had a high percentage of recent immigrants, with more than 42% of the city's residents being born outside Canada [59] . Approximately 60% of the participants were identified as non-white. The level of education also varied widely across participants, with 35% having not completed high school and 35% having completed a college or university degree. With regard to technology experience, 11 of the 17 (65%) participants who completed the questionnaire stated that they owned a cellphone (n=4) or smartphone (n=8), whereas 3 participants did not specify which kind of device (Table 3 ).

Only one participant had previously used a TM system, although it was not smartphone-based TM or multi-condition-based TM. Despite more than half of the participants owning a cellphone or smartphone, the comfort level across the participants was mixed. In the end, the response rate for the prestudy questionnaire was fairly high (16/17, 94%) but lower for the poststudy (10/17, 59%) questionnaires. One participant did not complete a prestudy questionnaire. An evaluation of the SF-36 data did not find any meaningful results, likely because of the low poststudy response rate. 

Of the 17 patients who completed the study, 13 patients used the HF module, 1 patient used the HTN module, 2 patients used the HF+DM modules, and 1 patient used the HTN+DM modules. One participant (MCCP008) was initially monitoring HF and DM but was later offboarded from the DM component because of a change in health status.

Adherence to each condition was evaluated independently, irrespective of the combination of conditions. The adherence for each patient with HF is displayed in Table 4 . The evaluation of adherence to HF was divided into 2 categories: the number of days patients took just the physiological readings (BP/HR/WT) and the number of days patients took all required physiological readings and completed symptom questions (full set). Although individual usage patterns varied across the participants, patients took physiological readings (BP/HR/WT), on average, over 77.2% (142/184) of days of the expected reported days. Of the patients with HF, 56% (9/16) were adherent to physiological readings over 80% of the days on TM, and 31% (5/16) were adherent over 90% of the days on TM. One patient (MCCP0017) took a full set of readings over 99% (183/184) of the days on TM. When evaluating the percentage of days that patients took all 3 physiological readings and symptom questions (ie, full set), the percentage dropped to 69.0% (127/184).

Overall, adherence in the HTN module was 55% (29/52) on average. HTN adherence was defined as taking at least one reading every 2 weeks. However, participants using the HTN component of the system, on average, took readings more frequently than required by the algorithm (Figure 2 ). DM adherence was on average 72% (19/26) defined as taking at least one reading per week ( Figure 3 ). 

The qualitative findings of the patient experience can be represented by 4 main themes and 12 subthemes. The main themes were (1) making sense of the purpose of TM, (2) engaging and investing in TM, (3) implementing and adopting TM, and (4) perceived usefulness and perceived benefits of TM in CCCs. The themes and subthemes were mapped together with the constructs within NPT and IOs as shown in Table 5 . "We were keeping a log book and it wasn't consistent but we were still trying to do that...The NP would have me email her the sugar readings. So I would have to transpose all that information into an email... they would review it when we go visit them." (MCCP0019) Acceptability Cognitive participation Engaging and investing in TM "We did] nothing. If we felt sick we'll go to the doctor. We took blood pressure now and again, but not on a regular basis, not like this." (MCCP0025)

Comparing the old ways of working to use TM "I think it was we actually really liked that it was connected, because it gave me a peace of mind. At least I know, okay, if something goes wrong, there's somebody there to call her. Someone is kind of checking on me." (MCCP0026) "You don't know whether it's good or bad, but with Medley they give that guideline to live with -within" (MCCP002) Improvement to ongoing self-care practices "I know if the reading comes up in orange-that's the orange. If it comes up in orange, check. If it gets worse during the course of the day check the clinic. And I know that feels good. I feel good. Yeah, I feel good, so that is very helpful, that is very, very helpful." (MCCP002)

Enabling immediate action on abnormal readings and trends "Yeah. I like knowing someone's keeping an eye on me. Certainly no one else will keep an eye on me now. [After I you mean?] Yes, there's no one now...I just liked it, that's all, knowing, they were there looking out for me." (MC-CP0024)

Concerns moving forward without TM "There were times [when it was] kind of like a grey area...She's not always feeling great and you enter the information based on the prompts and some of those prompts alerted the nurse at the hospital and we would get a phone call. And sometimes to me they were kind of unwarranted... Was she dizzy? Well she feels dizzy a lot of times. She's usually OK." (MCCP0019's caregiver) 

The first theme synthesizes participants' accounts of how they made sense of TM, from the initial introduction to the technology to understand its purpose to enable a more structured method of physiological monitoring to connect patient information to the clinic. Participants described the process of measuring physiological readings such as BP or blood sugar at home to track their conditions for their care team. Participants were familiar with the process of keeping a logbook to track readings taken between appointments:

We were keeping a log book and it wasn't consistent, but we were still trying to do that... One participant described how the ability to go back and look at previous readings made sense in terms of alleviating the challenge of maintaining up-to-date logs at home (ie, appropriateness). Often, manual tracking did not work because of inconsistent patient-provider communication or the inability to recall context-specific readings.

By comparing the patients' current care practices at home (ie, logs) and envisioning new work of taking digital remote readings, patients began to engage in using TM. Instead of calling in readings, by taking readings through the app, they were sent automatically to the clinic. This supported the relational work necessary to gain buy-in from patients to accept the technology and to invest time in a remote way of care.

Before the introduction of the TM intervention, there was significant variability in the frequency of participants taking physiological readings at home routinely. Almost all participants described taking physiological readings on occasion at home, in a doctor's office, or at a local pharmacy. One participant described their monitoring processes before beginning TM as irregular: 

Having the devices connected directly to the clinic supported buy-in from patients through increased patient-clinician communication, thereby building their capacity at home (ie, saving time, providing new information to manage their care, and connecting with their personal clinicians). The ability to visualize live data and provide reassurance that someone was looking out for their health generated recurring engagement with TM and an investment to continue using the system at home:

We were keeping a log book and it wasn'tconsistent but we were still trying to do that and we're going to have to go back to that unfortunately after this...

Several participants were hesitant about the process of learning TM. Initially, several participants described the smartphone as challenging. Some participants seemed to rely more on help from family members or other informal caregivers to use TM than others: 

This theme describes the process of enabling the patient's work to implement TM at home and how these practices led to a decision to adopt the intervention seamlessly into their normal routine.

The patient's work of using TM frequently became a part of their everyday routine. Along with getting up, going to the bathroom, and taking their medication, taking readings was described as a normal step in this everyday process: On the basis of the readings, patients would adjust the tasks they had done every day, for example, monitoring fluid intake or restricting salt intake.

Participants described that they perceived someone was looking out for them (ie, nurse) and taking care of them in the virtual 

Participants were trained to use the TM system individually before starting the study. Although a caregiver was not required for use, 11 of 17 (65%) participants had support at home. During the poststudy interviews, it was clear that caregivers heavily participated throughout the TM process, despite whether the support was necessary: 

In conclusion, all participants seemed to appraise the TM system as if it was already an embedded part of their daily routine in normal life. Participants described improvements in self-care knowledge, such as being able to identify when something was not right even if they felt no physical symptom differences. This enabled action on abnormal readings they may not have identified without Medly. Participants shared that clinicians would often quickly reach out to discuss abnormal findings. Participants described a strong desire to continue monitoring, even after the study ended. Despite high adherence rates, several technical issues presented minor challenges. By taking readings daily, patients described getting to know their target range (ie, reflexive monitoring). Participants described how the device data kept them informed, improving their knowledge of their condition, and how they could keep track of their health over time (ie, feasibility).

All participants referenced how Medly could identify when a reading or trend was not right. Many patients specifically noted using the color to identify the severity of an alert as well as the graphs within Medly to help identify abnormal readings:

When something is not right, your reading comes up a different color and you know something is wrong, but you can wait and retake it or you try it again on the next day.

[MCCP0020]

Although participants described differing preferences in how they identified abnormal readings, they described the actions they took to manage these situations:

. Although patients described being able to more clearly identify when something was not right while using TM, many patients still relied somewhat on the nurse-led team to identify abnormalities:

They did call because they were concerned and then we did come into the clinic later that day. So maybe it was a good thing to have Medly that particular time.

[MCCP0019]

When a reading measures outside of the individual tailored range, participants received algorithm-based self-care messages. However, several participants found that clinicians within the model, usually the nurse, would call the patients first:

Because they're usually pretty good. If there is a real critical [reading], they'll try to get us right away.

[MCCP0026]

There were times participants even described their intention to call the nurse in the clinic; several participants noted that the nurse followed up with them before they could reach out.

Participants expressed strong concerns about having to go back to managing their conditions at home without Medly, even to the point where they were willing to pay for the program out of pocket, or ask what store they could purchase the devices immediately. Participants described a sense of anxiety, knowing that no one would be monitoring their readings going forward: 

This study provides a detailed evaluation of the feasibility and perceived usefulness of a multi-condition TM platform using the experiences of patients with CCCs in an integrated nurse-led model of care. Study findings revealed that patients were highly adherent to self-monitoring using multi-condition TM, irrespective of which conditions were monitored at home. A virtual connection to the nurse-led team enabled patient acceptance of this new way of tracking readings at home and engaging in their care using multi-condition TM. By choosing to adopt TM into a daily routine, patients perceived that someone was looking out for them, reinforcing routine adherence and enabling patients to evaluate abnormal readings and trends. Participants perceived TM as useful, describing improved self-care knowledge and acting on information provided by TM in tandem with their nurse-led care team. Evidence of new normal practices was clear (ie, high adherence rates and the patient's detailed descriptions of perceived usefulness of TM), such that ending the study affected this new normal routine.

NPT and the IOs framework were used for the structural evaluation of feasibility based on the patient's experience using TM in nurse-led care (Table 5 ). Using the results, a conceptual framework was developed to visualize the patient experience in light of these constructs (Figure 4 ). 

The introduction of multi-condition TM for patients with CCCs in nurse-led care stimulated a simultaneous evaluation of acceptability (Does it make sense to the users?) and appropriateness (Does it fit?). Given these patients had frequent health system encounters that involve evaluating ongoing assessment of physiological indicators, TM of critical physiological measures aligned well with existing care plans (ie, coherence). The results indicated that patients did not always feel physical differences in their health when readings measured outside of their clinically indicated range, leading to often unanticipated condition exacerbation. Participants described the TM system as appropriate, suggesting that they understood the purpose of TM as part of their normal work, aligning with the NPT construct of cognitive participation that identifies how patients engaged with TM, accepting the technology in part as a replacement for traditional care practices. Patients accepted the idea of their new work in TM, in part because these devices were connected to the nurse. In many cases, an appointment was not necessary because of the virtual connections made through the TM to the clinic, creating further patient engagement and willingness to accept TM.

Participants described an ongoing cyclical process in which the collective actions required to adopt TM resulted in a cycle of evaluation and readoption (ie, construct of reflexive monitoring) over time. In some cases, implementing TM requires embedding new routines at home or adjusting existing ones. Patients described clinicians, particularly the nurse, calling to check-in on questionable readings. Given the strong support of caregivers in our participant group, NPT might suggest that caregivers played an integral role as contributors to the relational work required in adopting TM technology (ie, collective action). The collective actions, which demonstrated engagement in the care process, reinforced patients' reflections on the value of the technology to be positive. This led to further engagement, greater collective action, involvement of family caregivers, and overall feasibility, such that patients wanted to continue TM use after the study.

The results indicated an average adherence rate of 77.2% (142/184) in HF monitoring, 55% (29/52) in HTN monitoring, and 72% (19/26) in DM monitoring. Previous research has found varying adherence rates, typically between 40% and 90%, to TM in HF [47, 60, 61] . Another study found that the average adherence rate to BP monitoring and blood sugar monitoring was 59.7% and 50.2%, respectively [61] . In addition, other previous studies have found an initial drop in patient adherence during onboarding, followed by a steadier adherence rate over time [47, 48, 60] . In this case, an initial drop in adherence did not occur. A consistent overall adherence rate supports the feasibility of this multi-condition TM system. When comparing patients with HF who were adherent to physiological readings versus those who completed a full set including symptom questions, the average change in adherence was less than 10%. This suggests that the majority of patients in this cohort were adherent to both readings and symptom questions.

Although the adherence rates reported in this study are comparatively high, we anticipate that they are likely higher than reported. Similar to other Medly studies [47] , we did not have functionality, which would enable researchers to automatically account for periods when patients were unable to take readings for legitimate reasons (eg, admitted to hospital, traveling, device replacement, etc). Patients were asked to notify the team if they were going to miss a reading. In many cases, the clinical team was notified in advance, suggesting that adherence rates are higher than reported findings. It is also possible that a subset of the incomplete readings, such as missing symptom questions, was because of the Bluetooth connectivity issues. Although patients may have taken a set of readings, it is possible that a loss in network connectivity could have reported a missed reading on that day.

The fidelity of the intervention was considered to contribute to the overall evaluation of TM feasibility in nurse-led care. The degree to which the intervention was delivered as intended is defined as intervention fidelity [41] . In this study, patients did not always complete the symptom questions, contributing to a difference in the adherence between a physiological reading and a full set across participants with HF and therefore lower fidelity to the original intervention. For patients with HF, adherence to completing the symptom questions was lower than completing only the physiological measures. Symptom questions may not have been completed for 3 reasons: (1) questions were not reflective of significant changes on a day-to-day basis, (2) questions were not relevant to how they felt that day (ie, determination of self-management), or (3) symptoms that are perceived as important to patients with CCCs may not be reflected in single-disease protocols. Therefore, the patient experience in the context of the combination of conditions monitored is important in evaluating the intervention's fidelity in nurse-led care.

On the basis of the patient experience, a multi-condition TM platform is feasible for patients with CCCs in an integrated nurse-led care model. Patients accepted and adopted the technology as demonstrated by a high level of adherence. Historically, the adoption and use of technology have had greater benefits in younger populations compared with older adults who may be less familiar with new technologies [62] . However, the wide age range across participants of both genders suggests the ability to use Medly does not appear to be associated with age or technology experience in this case. Although patients described being able to more clearly identify when something was not right while using TM, our results found that patients still relied somewhat on the clinical team to identify abnormalities. This contributed to adherence and continued use. This reliance on a clinical connection to their clinical or nurse-led team has been demonstrated in other research [15, 63, 64] .

There are several implications for future multi-condition TM apps as well as scaling up existing programs that focus on populations with CCCs. Attempts to tailor specific symptom questions have already been initiated in other eHealth technologies such as the electronic patient-reported outcome tool [65] . Further research should be conducted to explore the content and frequency of symptom questions in CCCs. Given the lower level of adherence to symptom questions in this study, an evaluation may suggest subtle changes to the question content or the frequency of questions required by the algorithm. Future research could explore these adjustments within the Medly platform as well as other TM platforms.

More research is needed to explore the extent of caregiver support in TM as well as identify the criteria for suitable enrollment of certain patient subgroups that require caregiver support. It is possible that TM interventions for patients with CCCs could be expanded if the implications of caregiver support and the role of caregivers are more broadly understood.

Finally, given the historically high rates of health care utilization in complex populations and the need for physical distancing because of infectious diseases such as COVID-19, TM solutions that improve patient experience should be explored as viable solutions to avoid in-person appointments while continuing to monitor complex patients closely, manage care needs remotely, and mitigate unanticipated visits to the emergency department.

The strengths of this study include the depth of the interview data collected as well as the rigorous approach to analysis using 2 trained qualitative coders and 2 theoretical frameworks (NPT and IOs). Interviews were candid, and diverse participants appeared to be forthcoming in their experiences using TM and perspectives on how to move forward with TM in nurse-led care. Given the nature of evaluating the initial feasibility of TM in nurse-led care for this population, the use of NPT and IOs worked well. There is an opportunity to use other well-known theoretical frameworks in future research, such as the Unified Theory of Acceptance Use of Technology 2 that undertakes a deeper dive in constructs such as price value, hedonic motivation, and effort expectancy in a program implementation.

There are several limitations to this study. First, the heterogeneous sample size was small (n=26), with only few patients using TM to monitor more than one condition. In an effort to capture the broad spectrum of CCCs, we did not attempt to randomize our sample. Only a small number of participants monitored HTN or DM exclusively. It is possible that different adherence results may have been experienced with more multi-condition TM participants. Second, as participants were conveniently sampled based on the recruitment criteria and not randomly selected, a selection bias is possible within the study sample. Third, a defective phone battery in several phones during the study generated the illusion of nonadherence, but for legitimate reasons. Owing to delayed shipping, these participants were unable to synchronize readings from the devices to the phone, likely lowering the apparent adherence data. Due to the low response rate of the SF-36, we were unable to incorporate it into our analysis. Finally, participants were only followed up for 6 months; therefore, adherence after the study period remained unknown as well as an optimal duration of TM in this population.

Patients with CCCs perceived TM within a nurse-led care model to be feasible based on their experience using a multi-condition TM platform for 6 months. Overall, this study found promising adherence rates across the 3 conditions monitored by TM in this study. Patients monitoring HF demonstrated the highest rates of adherence at 77.2% (142/184) of the days in the study period. The qualitative results enabled an exploration of the feasibility of multi-condition TM, which could then be mapped to the constructs of NPT and IOs. Given the experiences of patients with CCCs, TM via multi-condition platforms in nurse-led care models should be considered to meet the growing need for virtual care interventions to support remote care of CCCs in the future.

None declared.

Aging with multimorbidity: A systematic review of the literature

The prevalence of multimorbidity and associations with lifestyle factors among middle-aged Canadians: an analysis of Canadian Longitudinal Study on Aging data

The association between multimorbidity and hospitalization is modified by individual demographics and physician continuity of care: a retrospective cohort study

Multimorbidity and healthcare utilization among home care clients with dementia in Ontario, Canada: A retrospective analysis of a population-based cohort

The effects of multiple chronic conditions on hospitalization costs and utilization for ambulatory care sensitive conditions in the United States: a nationally representative cross-sectional study

An 'endless struggle': a qualitative study of general practitioners' and practice nurses' experiences of managing multimorbidity in socio-economically deprived areas of Scotland

Multimorbidity patterns in relation to polypharmacy and dosage frequency: a nationwide, cross-sectional study in a Japanese population

| iss. 1 | e22118 | p

Understanding patients' experiences of treatment burden in chronic heart failure using normalization process theory

determinants and patterns of multimorbidity in primary care: a systematic review of observational studies

A multiple-case study in nurse practitioner-led clinics: an exploration of the quality of care for patients with multimorbidity

A nurse-led model of chronic disease management in general practice: Patients' perspectives

Nurse-Led Models of Care for Patients with Complex Chronic Conditions: A Scoping Review

Quality of Care for Patients With Diabetes and Mulitmorbidity Registered at Nurse Practitioner-Led Clinics

A formative evaluation of a nurse practitioner-led interprofessional geriatric outpatient clinic

Exploring an Innovative Care Model and Telemonitoring for the Management of Patients With Complex Chronic Needs: Qualitative Description Study

Nurse practitioner's perceptions of the impact of the nurse practitioner-led clinic model on the quality of care of complex patients. Prim Health Care Res Dev

Improving Patient Experience and Primary Care Quality for Patients With Complex Chronic Disease Using the Electronic Patient-Reported Outcomes Tool: Adopting Qualitative Methods Into a User-Centered Design Approach

Systematic Review of Home Telemonitoring for Chronic Diseases?: The Evidence Base Home Telemonitoring?: A Definition

Effects of home telemonitoring interventions on patients with chronic heart failure: an overview of systematic reviews

Effectiveness of mHealth interventions for patients with diabetes: an overview of systematic reviews

Remote Care Technology: A Systematic Review of Reviews and Meta-Analyses

Comparative Effectiveness of Telemonitoring Versus Usual Care for Heart Failure: A Systematic Review and Meta-analysis

Home telemonitoring effectiveness in COPD: a systematic review

Remote home management for chronic kidney disease: A systematic review

Health-economic evaluation of home telemonitoring for COPD in Germany: evidence from a large population-based cohort

Mobile phone-based telemonitoring for heart failure management: a randomized controlled trial

Clinical effects of home telemonitoring in the context of diabetes, asthma, heart failure and hypertension: a systematic review

| iss. 1 | e22118 | p

A Tele-Behavioral Health Intervention to Reduce Depression, Anxiety, and Stress and Improve Diabetes Self-Management

Telemonitoring and Team-Based Management of Glycemic Control on People with Type 2 Diabetes: a Cluster-Randomized Controlled Trial

Comparative effectiveness of telemonitoring versus usual care for type 2 diabetes: A systematic review and meta-analysis

Long-term Outcomes of the Effects of Home Blood Pressure Telemonitoring and Pharmacist Management on Blood Pressure Among Adults With Uncontrolled Hypertension: Follow-up of a Cluster Randomized Clinical Trial

Effectiveness of home blood pressure telemonitoring: a systematic review and meta-analysis of randomised controlled studies

Effects of home telemonitoring to support improved care for chronic obstructive pulmonary diseases

Perceptions and experiences of heart failure patients and clinicians on the use of mobile phone-based telemonitoring

Patient expectations and experiences of remote monitoring for chronic diseases: Systematic review and thematic synthesis of qualitative studies

Better Effectiveness After Transition-Heart Failure (BEAT-HF) Research Group. Effectiveness of Remote Patient Monitoring After Discharge of Hospitalized Patients With Heart Failure: The Better Effectiveness After Transition --Heart Failure (BEAT-HF) Randomized Clinical Trial

Efficacy of telemedical interventional management in patients with heart failure (TIM-HF2): a randomised, controlled, parallel-group, unmasked trial

Telemonitoring to Manage Chronic Obstructive Pulmonary Disease: Systematic Literature Review

Effectiveness of telemonitoring versus usual care for chronic obstructive pulmonary disease: A systematic review and meta-analysis

Outcomes for implementation research: conceptual distinctions, measurement challenges, and research agenda

Derivation and validation of an index to predict early death or unplanned readmission after discharge from hospital to the community

Development of a theory of implementation and integration: Normalization Process Theory

Implementing, Embedding, and Integrating Practices: An Outline of Normalization Process Theory

Normalisation process theory: a framework for developing, evaluating and implementing complex interventions

Towards a general theory of implementation

Patient adherence to a mobile phone?based heart failure telemonitoring program: A longitudinal mixed-methods study

Mobile phone-based remote patient monitoring improves heart failure management and outcomes: a randomized controlled trial

| iss. 1 | e22118 | p

Developing healthcare rule-based expert systems: case study of a heart failure telemonitoring system

An audit of sample sizes for pilot and feasibility trials being undertaken in the United Kingdom registered in the United Kingdom Clinical Research Network database

Sample size of 12 per group rule of thumb for a pilot study

The MOS 36-ltem Short-Form Health Survey (SF-36)

Agency and implementation: Understanding the embedding of healthcare innovations in practice

Interpretive description: A noncategorical qualitative alternative for developing nursing knowledge

Interpretive description: a noncategorical qualitative alternative for developing nursing knowledge

Strengths and Challenges in the Use of Interpretive Description: Reflections Arising From a Study of the Moral Experience of Health Professionals in Humanitarian Work

Code Saturation Versus Meaning Saturation

Saturation in qualitative research: exploring its conceptualization and operationalization

Census Profile

Patient Compliance in Home-Based Self-Care Telehealth Projects

Patient Adherence to Scheduled Vital Sign Measurements During Home Telemonitoring: Analysis of the Intervention Arm in a Before and After Trial

Adoption of Smart Technology Among Older Adults: Challenges and Issues

What is important to people with multimorbidity and their caregivers? Identifying attributes of person centred primary health care from the user perspective

Processes of care desired by elderly patients with multimorbidities

The Electronic Patient Reported Outcome Tool: Testing Usability and Feasibility of a Mobile App and Portal to Support Care for Patients With Complex Chronic Disease and Disability in Primary Care Settings

Experiences of Complex Patients With Telemonitoring in a Nurse-Led Model of Care: Multimethod Feasibility Study JMIR

), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work, first published in the Journal of Medical Internet Research, is properly cited. The complete bibliographic information