This study had a cross-sectional, descriptive, mixed methods design. All patients with HF and an ICD with RPM from one region in Sweden were invited to participate in the study. In addition, all ICD nurses working at all the ICD clinics in Sweden were invited to complete a survey.

The region of Sörmland has a land area of 6103 km² and about 300,000 inhabitants. There are three hospitals with a total of 270 hospital beds on medical wards, but only one hospital has an in-hospital device clinic. The travel distance for patients with an implanted device could be up to 100 km.

Over 1 year, from October 2015 until October 2016, RPM was introduced and started for all ICD recipients during their scheduled visits at the device clinic, and all new ICD recipients received RPM directly after the implant. Due to various reasons, 4 patients declined RPM, so at the end of 2018 there were 310 ICD recipients on RPM in Sörmland County.

Before implementing RPM, ICD recipients had scheduled office-based visits every 3 to 6 months. After the implementation of RPM, patients routinely visit the clinic once a year and, in between, a scheduled transmission is performed. The ICD recipients can call the ICD nurse on weekdays when needed.

All adult ICD recipients having a verified HF diagnosis according to the European Society of Cardiology guidelines [1] (N=177) were invited to participate in the study during their yearly follow-up visit at the in-hospital device clinic, from January to December 2018. Exclusion criteria were being less than 18 years old and not being able to understand Swedish.

Patients interested in participating provided written informed consent. They were thereafter given additional written information and questionnaire packets to complete at home and return by mail in prepaid envelopes.

The ICD nurses were identified by contacting the National Swedish Pacemaker and ICD Registry, which provided names and email addresses for all ICD nurses (N=50) working at an ICD clinic in Sweden at the time. An electronic survey was sent out and two reminders were posted.

Descriptive statistics were used to present sample characteristics for all study variables. Demographic and clinical variables were compared using chi-square statistics or the Student t test.

Regarding experiences of remote monitoring, for both patients and nurses, the items were divided into two subscales based on an exploratory factor analysis: one with negative experiences (items 2-4) and one with positive experiences (items 5-7). The items in each subscale were summed to a total score. For the negative experiences (score range 3-12), lower scores reflected more negative experiences with RPM. For the positive experiences (score range 3-12), higher scores reflected higher levels of more positive experiences with RPM. Item 1 was considered neutral and was calculated separately for each group.

The level of statistical significance was set to P<.05. The statistical analyses were conducted using SPSS Statistics for Windows, version 25.0 (IBM Corp).

The final open-ended question was analyzed with manifest qualitative content analysis [31]. The three authors who developed the survey and analyzed the qualitative data are researchers with years of experience as clinical nurses, two working in HF care (ML and AS) and one in ICD care (IT).

The study was approved by the Regional Ethics Committees for Human Research in Linköping, Sweden (ref 2017/441-31), and was conducted in accordance with the World Medical Association Declaration of Helsinki and the Code of Ethics for Nurses [32,33].

The sample consisted of 175 patients (response rate 98.9%) and 30 nurses (response rate 60%). The patients’ mean age was 69.9 years (SD 9.7); 138 out of 175 patients (78.9%) were males, 144 were retired (82.3%), and 109 were married (62.3%). A total of 128 out of 175 participants (86.0%) reported no symptoms of anxiety or depression, and 129 out of 175 patients (73.7%) scored a low level of ICD-related concerns (see Table 2).

The nurses’ mean age was 52.7 years (SD 8.7) and 26 out of 30 (86%) were females. They had been working as nurses for a mean of 26 years (range 5-47) and as ICD nurses for 7.6 years (range 1-14). They spent a mean of 7.5 hours/week (range 1-30) working with RPM (see Table 2).

The majority of patients (154/175, 88.0%) as well as nurses (23/30, 77%) experienced RPM in general as very good. The mean scores of the negative experiences subscale were 11.5 (SD 1.1) for the patients and 10.7 (SD 0.9) for the nurses (P=.08), with a trend for the nurses being more negative than for the patients, although this was not statistically significant.

A total of 16 out of 175 patients (9.1%) and 4 out of 30 nurses (13%) found RPM unnecessary or partly unnecessary (see Figure 1). A total of 17 out of 30 nurses (57%) found the technical equipment somewhat difficult for the patients to handle; in contrast, 149 of the 175 patients (85.1%) answered that they did not experience any technical difficulties in handling RPM (P=.04). Only 11 out of 175 patients (6.3%) were worried or anxious about what the RPM entailed, while 15 out of 30 nurses (50%) felt distressed by the responsibility that accompanied their work with RPM (P=.04).

The mean scores of the positive experiences subscale were 11.1 (SD 1.6) for the patients and 8.5 (SD 1.9) for the nurses (P=.03), meaning that patients were more positive toward RPM than were the nurses.

The majority of the patients found that RPM increased their own (173/175, 98.9%) and their relatives’ (169/175, 96.6%) security. All nurses (30/30, 100%) answered that RPM increased patient security and safety, and 28 out of 30 nurses (93%) found it necessary from a patient safety perspective (see Figure 2). Also, 15 out of 30 nurses (50%) answered that the responsibility that accompanied working with RPM increases their security, since they knew there would be an alert in case of malfunction or arrhythmias.

Based on patients’ and nurses’ experiences of RPM, Table 3 presents weighted results with clinical implications and needed interventions for changed practice.

A main finding in this study was that both patients and nurses found several positive aspects with RPM, but the nurses noted more downsides than did the patients. There could be several explanations to this. Two survey studies exploring telemonitoring in HF in various countries found that health care professionals described several patient-related barriers due to physical or mental conditions. Further, the nurses have seen a range of different technical problems in different patients over time, while the individual patients only have their own experiences and, therefore, a more optimistic and positive view on the technology [34,35].

Patients expressed that they could live a more normal life after receiving RPM. This confirms results in previous research showing that patients were content with RPM and did not feel like patients as much anymore [16,36]. This is probably related to the fact that patients felt that RPM increased security and safety for both themselves and their relatives. Also, for the vast majority, it was considered as an advantage that RPM lead to fewer office-based visits. The advantage that RPM is less time-consuming than in-hospital follow-ups has also been seen in other studies [16]; however, both in this study and in previous research, the lack of direct face-to-face contact was missed by some patients [37].

Even as patients received information and education about RPM, some patients believed that RPM included live transmissions, both day and night, and that someone continuously watched their electrocardiograms as when being monitored with telemetry during a hospital stay. Ottenberg et al found that when patients were prescribed RPM for their ICD but then did not install the home monitor, it was largely attributed to not understanding the purpose of the RPM system or being unsure whether their system was correctly transmitting information [37]. This highlights the need for repeated patient education about RPM and its management using a person-centered approach and by applying, for example, teach-back methodology [38]. High-quality training has been found to improve patients’ understanding and comprehension and has been positively associated with anxiety and acceptance levels [23].

Downsides described by the nurses were often related to organizational issues; for example, a clear description of their own responsibilities handling alerts. Further, about half of the nurses also found the responsibility associated with managing RPM stressful due to different technologies, and they had concerns related to patient safety. A recent review describes the requirement of the referring nurse to be an expert in cardiac pacing and device follow-up, and a daily connection with a website should be performed to evaluate received alerts. To avoid stress and worries for the nurses, every center must define a written decision algorithm in order to standardize the handling of alerts [13].

It was also referred to as “invisible work” that is not recognized as time-consuming by managers and without proper reimbursement systems. It is important that health systems provide a supportive environment for RPM (ie, activities that do not involve direct patient interaction), since the most frequently reported barrier for not implementing RPM was found to be lack of reimbursement [17,39]. In the eHealth era, with increasing remote monitoring of various treatments, symptoms, and devices, adapting the organization of care is key and organizations need to adapt to best make use of remote monitoring.

Nurses also found handling the daily transmissions as burdensome and time-consuming since it was difficult to tailor the alarm limits for each patient. It is known, as shown previously, that many transmissions are patient initiated without any event. In a recent study by Ninni et al [40] that included 1423 transmissions, it was found that as many as 77% were initiated by the patients, and only about 3% of the transmissions per patient led to actions or interventions by the health care personnel. The authors stressed the need to optimize automatic transmissions and focus on patient education to reduce the workload at the device clinic.

In addition, technical concerns took a lot of time, with nurses trying to locate the problem and get in contact with the patient. This led to stress and an increased workload, and nurses also felt that it was a false safety for the patient who did not always realize that the transmission failed. A recent study highlights the encounter through telephone calls that took place related to home monitoring. Five types of clinical work were performed that may also refer to RPM: inclusion work, coordination work, diagnostic work, education work, and comfort work [41]. The authors found that telephone calls increased time spent in telemonitoring, and most telephone calls contained more than two issues [41].

Currently, patients do not receive information directly from their remote monitored device, which patients in this study pointed out as a downside. However, findings from a feasibility [42] and evaluation [20] study in the United States suggest that it is not only feasible to deliver data from remote monitoring directly to patients, but also that this data sharing does not adversely impact clinic workflow and that patients perceive a benefit from having access to their remote monitoring data. The same result was found by Mirro et al when evaluating the impact of sharing ICD data summaries through a patient portal. At the end of the study, two-thirds of patients were satisfied with the amount of information received through the electronic or paper ICD data summary. Further, providing patients with their device data did not increase ICD-specific clinical workload [43].

In the future, by using specific apps in smartphones, patients may have the ability to check the website with information about their own devices and to communicate or chat online with the health care personnel involved in the care of the patient.

Both patients and ICD nurses found RPM to be safe and to increase the sense of security for patients and caregivers. There was a discrepancy between nurses and patients with regard to the technical equipment, where very few patients, but every other nurse, stated that the technology was difficult for the patients to handle.

Few patients were worried or anxious about the RPM, while half of the nurses felt distressed by the responsibility that accompanied their work. Nurses also described it as time-consuming to contact patients in case of alerts. To improve RPM from the perspective of the nurses, the organizational routines, reimbursement systems, and the balance of responsibilities and workloads need to be reviewed.