Determinants of global cardiac implantable electrical device remote monitoring utilization – Results from an international survey

Background Despite near-global availability of remote monitoring (RM) in patients with cardiac implantable electronic devices (CIED), there is a high geographical variability in the uptake and use of RM. The underlying reasons for this geographic disparity remain largely unknown. Objectives To study the determinants of worldwide RM utilization and identify locoregional barriers of RM uptake. Methods An international survey was administered to all CIED clinic personnel using the Heart Rhythm Society global network collecting demographic information, as well as information on the use of RM, the organization of the CIED clinic, and details on local reimbursement and clinic funding. The most complete response from each center was included in the current analysis. Stepwise forward multivariate linear regression was performed to identify determinants of the percentage of patients with a CIED on RM. Results A total of 302 responses from 47 different countries were included, 61.3% by physicians and 62.3% from hospital-based CIED clinics. The median percentage of CIED patients on RM was 80% (interquartile range, 40–90). Predictors of RM use were gross national income per capita (0.76% per US$1000, 95% CI 0.72–1.00, P < .001), office-based clinics (7.48%, 95% CI 1.53–13.44, P = .014), and presence of clinic funding (per-patient payment model 7.90% [95% CI 0.63–15.17, P = .033); global budget 3.56% (95% CI -6.14 to 13.25, P = .471]). Conclusion The high variability in RM utilization can partly be explained by economic and structural barriers that may warrant specific efforts by all stakeholders to increase RM utilization.


Introduction
In patients with cardiac implantable electronic devices (CIED), remote monitoring (RM) has shown to improve survival, quality of life, and health care utilization. 1,2As such, RM has been recommended as part of the standard of care in patients with CIEDs. 2,3Despite the near-global availability of RM, there is a high geographical variability in the uptake and use of RM. 4,5 For example, in the prospective ADVANCE cardiac resynchronization therapy (CRT) registry approximately 60% of patients enrolled in North and South America were on RM, while this was ,6% in Asia. 5 The underlying reasons for this geographic disparity remain largely unknown, but have been hypothesized to depend on reimbursement, personnel resources, and infrastructure. 1,2Therefore, the recent collaborative multiprofessional society expert consensus statement on RM called for novel data to identify local barriers that may limit the utilization of RM in order to develop strategies to optimize RM usage. 2 An aim of this international survey included gathering global data on the use of RM and reimbursement to study the determinants of worldwide RM utilization and identify locoregional barriers of RM uptake.

Study design and participants
The intended target population of the survey included all CIED clinic personnel, including physicians, advanced practice providers, nurses, technicians, and others.An online survey was administered using the SurveyMonkey platform and distributed among the Heart Rhythm Society global network on December 8, 2022.All participants provided their individual informed consent electronically before proceeding with the survey.The questionnaire could be completed anonymously, but information on participants' location and the name of their center were requested.All responses submitted by December 31, 2022, were analyzed.This study was a qualitative improvement initiative of the Heart Rhythm Society Digital Health Committee.The research reported in this paper adhered to the Helsinki Declaration as revised in 2013.

Survey design
The survey was conducted in English and consisted of 40 questions in total.First, demographic information was collected, including the participant's CIED clinic role, size of CIED clinic, and location (country, city, and name of center).Subsequently, information on the use of RM and organization of the CIED clinic was collected, including the percentage of patients enrolled in RM and frequency of remote CIED checks for each type of CIED.Lastly, information about local reimbursement and clinic funding was collected.The survey also included qualitative questions, which are beyond the scope of the current manuscript.In total, the survey required approximately 20 minutes to complete.

Socioeconomic and geographical data
To study the impact of global socioeconomic and geographic differences on RM utilization, the following information was researched and collected for each country: -Population density, expressed as people per square kilometer in 2021, as per data from the United Nations Statistics Division. 6-Gross national income (GNI) per capita, expressed in US dollars (World Bank Atlas method 7 ).Countries were divided into high-and low-income countries, as per data from 2021 by the World Bank. 8The high-income threshold for 2021 was set at US$13,205 on July 1, 2022. 8

Statistical analysis
For this analysis, incomplete responses and responses by industry and third-party RM services were excluded.If multiple responses from the same center were submitted, the most complete response was included in this analysis.Continuous variables are presented as median and interquartile range (IQR), as all variables showed a non-normal distri-

Respondent characteristics
A total of 548 responses were received, of which 344 responses (62.8%) were complete (Figure 1).After excluding multiple responses from the same center (n 5 37), as well as third-party services (n 5 3) and responses from industry (n 5 2), a total of 302 responses (55.1%) were included in this analysis.The characteristics and origin of the available responses are presented in Figure 2 and Table 1.In total, responses from 47 different countries were received.Most responses were provided by physicians (61.3%), both adult and pediatric electrophysiologists, followed by nurses (16.2%) and nurse practitioners (7.3%).The majority of responses were from a hospital-based CIED clinic (62.3%) that operates only during business hours (76.1%).The median number of implanters per center was 4 (IQR 3-6) and a median of 2000 patients (IQR 700-4000) were followed in the CIED clinics of respondents, of which 80% (IQR 40-90) were on RM.The median patient to staff ratio was 750 (IQR 466-1100) and 58 centers (19.7%) used third-party RM services.

Determinants of global RM utilization
The results of univariable and multivariable linear regression analysis for determinants of the percentage of patients with CIEDs on RM are presented in Table 2.The GNI per capita, clinic type, and clinic funding were independently associated with the percentage of patients with CIEDs on RM.For each US$1000 GNI per capita, the percentage of patients on RM was on average 0.86% higher (95% CI 0.72%-1.00%).For example, the 2021 US GNI per capita was US$70,930, while this was US$8220 for the Dominican Republic.Office-based clinics were associated with a 7.48% higher percentage of patients on RM (95% CI 1.53%-13.44%).Lastly, when compared to CIED clinics without any clinic funding, clinics funded by a per-patient payment model were associated with a 7.90% (95% CI 0.63%-15.17%)higher percentage of patients on RM while clinics funded by a global budget model were associated with a nonsignificant 3.56% (95% CI -6.14% to 13.25%) higher percentage of patients on RM.

GNI per capita
For further analysis, countries were categorized into highand low-income countries (Figure 2) based on the GNI per capita high income threshold for 2021, which was set at US$13,205. 8Table 3 provides a comparison of CIED clinics between high-and low-income countries.Responses from low-income countries were more often provided by physicians, most likely owing to the smaller CIED clinic team as the median patient-to-staff ratio was equal to the median number of CIED patients followed.CIED clinics in lowincome countries were significantly smaller and had a very low utilization of RM when compared to CIED clinics in high-income countries (10.0%vs 85.0%, P , .001).CIED follow-up routine was not significantly different between high-income and low-income countries, but the funding for CIED clinic follow-up differed significantly, where 57.1% of CIED clinics in low-income countries did not have any funding while this was only 24.0% for CIED clinics in high-income countries (P , .001).

Office-based vs hospital-based CIED clinics
A comparison by CIED clinic type is presented in Table 4.There was no difference in the median number of patients followed between office-based and hospital-based CIED clinics (2000 vs 2000; P 5 .0.662), but the percentage of patients on RM was significantly higher in office-based clinics (90.0%vs 70.0%, P , .001).The patient-to-staff ratio was nonsignificantly higher in office-based clinics (median 1000 vs 600; P 5 .057).The CIED follow-up routine was significantly more frequent for all CIED types and for heart failure diagnostics in office-based clinics, corresponding to a lower percentage of RM checks that are alerts in office-based clinics (17.5% vs 20.0%, P 5 .013).Most office-based clinics were funded by a per-patient payment model (75.2%), while this was only about half for the hospitalbased clinics (51.7%, P , .001).Lastly, office-based clinics more often used third-party RM services (31.5% vs 12.5%; P , .001).

Clinic funding
A comparison by clinic funding model is presented in Table 5.Overall, there was a large heterogeneity in CIED clinic funding models, as presented in Figure 3.In 19 countries, more than 1 different CIED clinic funding model was reported by the respondents.Besides the previously reported differences, the median number of CIED patients followed differed significantly between the different CIED clinic funding models (median: no funding 1200 vs per-patient payment 2000 vs global budget 3050; P 5 .010).The frequency of remote CIED checks and heart failure diagnostics was significantly higher in CIED clinics with a per-patient payment model when compared to no clinic funding and funding by global budget (Table 5).The use of third-party RM services was significantly higher in CIED clinics with per-patient payment models (no funding 5% vs per-patient payment 27.0% vs global budget 6.1%; P , .001).

Discussion
This international survey on CIED clinic organization for RM in patients with CIEDs aimed to improve the understanding of global RM utilization in the post-COVID era.There was a high variability in the use of RM across the globe, which was associated with the CIED clinic type, the country's GNI per capita, and the funding model of the CIED clinic.
While the number of CIED manufacturers is limited and all have a near-global availability, RM utilization is highly variable across the globe.Lau and colleagues 4 reported a high variability in the use of RM in the Asia-Pacific region in 2013.While the RM utilization was very low in Southeast Asia, India, and Hong Kong, the RM utilization in Australia, New Zealand, and Japan was around 50% in patients with implantable cardioverter-defibrillator (ICD) and CRT devices. 4More recent data from 2022 by Varma and colleagues 5 showed that this discrepancy between Asia and  America persists, albeit they could not identify the underlying reason.The results of this survey reflect the post-COVID era.A recent European survey showed a significant increase in RM utilization for pacemakers and implantable loop recorders, while there was no difference for ICD and CRT devices. 10In this European survey, RM utilization  Countries were categorized as high and low income based on the gross national income per capita.The high-income threshold for 2021 was set on July 1, 2022. 8ontinuous variables are presented as median and interquartile range (Q1-Q3).Categorical variables are presented as numbers and percentages.
was highest in patients with ICDs, at 69.6%, which is still lower compared to the observed median RM utilization in our survey.This emphasizes the geographical discrepancies discussed above and the need for more detailed insights in global trends in RM use.Further, even within countries the uptake of RM varies significantly. 11A socioeconomic analysis in the United States showed that lack of health insurance, unemployment, lower median income, and living in an urban neighborhood were associated with less RM utilization. 11To the best of our knowledge, this international survey is the first to collect global data on the use of RM and to identify determinants of RM utilization on a global basis.
In our survey, there was no significant difference in the frequency of routine remote CIED checks and heart failure diagnostics between high-income and low-income countries, but the frequency of remote CIED checks and heart failure diagnostics was significantly higher in officebased CIED clinics and in clinics with a per-patient payment funding model.Furthermore, third-party RM services were more frequently used by CIED clinics with a perpatient payment funding model.This may result from a financial incentive for CIED clinics where RM, besides providing the recommended clinical service, may be used as a business model.Conversely, the lack of such a financial incentive might prevent the provision of this recommended clinical service.This economic and financial barrier has been identified before.In 2015, a European survey identified the lack of reimbursement as the most important barrier for RM implementation in Europe, followed by the increased workload and lack of infrastructure. 2,12Similar findings were reported by a serial analysis of 2 surveys in Italy over a 5-year period, where 86.7% of centers reported that the lack of reimbursement was the main barrier to implement RM. 13 Identifying the socioeconomic and structural barriers raises the question of how to improve care for these patients.Given the proven benefits of RM and that it is considered standard of care, the overall aim would be to increase RM utilization at a global level. 2,14This includes providing the CIED patients in low-income countries with the same benefits associated with RM as CIED patients in high-income countries, but also decreasing these disparities within a single country.This will require identifying local barriers and then specific efforts by health care providers, CIED manufacturers, third-party RM services, and patients to overcome    these barriers and improve RM utilization. 2Future research should aim to identify specific hurdles that can be addressed, such as mobile network and Wi-Fi access, but also nonfinancial determinants such as details on culture and education programs.RM utilization (provision of RM to the patient) is insufficient, and attention must be paid to increasing RM adherence (RM continuous connectivity by the patient) as well. 2,11,15

Limitations
The results from this survey should be interpreted along with the inherent limitations.First, as the survey was distributed by the Heart Rhythm Society Digital Health Committee, the number of invitations sent will be several orders of magnitude larger than the number of responses received, resulting in a risk for selection bias that may affect the results.Second, more than half of the included responses were from the United States and several other countries had a low number of responses.A sensitivity analysis was performed in which the number of responses from the United States was reduced by random sampling down to a third of available responses, and the results of the multivariate linear regression remained significant.Third, no data on market shares for CIED vendors were available, nor did the survey address whether there was a price difference for RMenabled devices or services for the patient.Further, data on mobile network access and access to Wi-Fi were not available.Fourth, reported numbers and details on the CIED clinic may represent rough estimates; however, in surveys one must assume that respondents report data to the best of their knowledge.Finally, several nonfinancial variables of interest were not collected in the survey, for example cultural differences and details on literacy and education programs.

Conclusion
Worldwide, there is a high variability in RM utilization, which was associated with the gross national income per capita, office-vs hospital-based clinics, and the CIED clinic funding model.The frequency of routine RM CIED checks and heart failure diagnostics was significantly higher in CIED clinics funded by per-patient payment models, suggesting that financial incentive can increase RM utilization.Identification of these economic and structural barriers should be followed by implementation of appropriate reimbursement models adapted to the local health care systems.This could not only increase RM utilization, but also optimize CIED clinic workflows to cope with the associated data load when increasing RM uptake.

Figure 1
Figure 1 Data flow from the survey to the current analysis.

Figure 2
Figure 2 World map illustrating available responses with comparison of high-and low-income countries.Countries were categorized as high-and low-income based on the gross national income per capita.The high-income threshold for 2021 was set on July 1, 2022.8

CIED 5
cardiac implantable electronic device; HF 5 heart failure; RM 5 remote monitoring.Continuous variables are presented as median and interquartile range (Q1-Q3).Categorical variables are presented as numbers and percentages.

Figure 3
Figure3World map illustrating the global heterogeneity in cardiac implantable electronic device clinic funding models.
bution using the Kolmogorov-Smirnov test.Categorical variables are presented as numbers and percentages.Continuous parameters were compared between groups using the nonparametric Mann-Whitney U test or Kruskal-Wallis test.Categorical parameters were compared using c 2 tests.

Table 2
Multivariable linear regression for percentage of patients with cardiac implantable electronic devices on remote monitoring †Forward selection with entry if P value for addition to the model , .050.Constant of the final model 5 4.03 (95% CI -5.59 to 13.65, P 5 .411).

Table 3
Comparison of responses from high-and low-income countries CIED 5 cardiac implantable electronic device; HF 5 heart failure; RM 5 remote monitoring.

Table 4
Comparison of office-based and hospital-based cardiac implantable electronic device clinics

Table 5
Comparison of clinic funding models