Data Mining to Identify the Right Interventions for the Right Patient for Heart Failure: A Real-World Study

Article information

Healthc Inform Res. 2025;31(1):66-87

Publication date (electronic) : 2025 January 31

doi : https://doi.org/10.4258/hir.2025.31.1.66

Keni Lee ¹

, Ramzi Argoubi ²

, Halley Costantino ³^,^*

¹Sanofi, Reading, UK

²Oracle Life Sciences, Paris, France

³Oracle Life Sciences, Austin, TX, USA

Corresponding Author: Keni Lee, Sanofi, 410 Thames Valley Park Dr, Earley, Reading RG6 1PT, UK., Tel: +44 7740935256, E-mail: keni.lee@sanofi.com (https://orcid.org/0009-0006-3991-2113)

*Current affiliation: BluePath Solutions, Los Angeles, CA, USA

Received 2024 July 4; Revised 2024 November 6; Accepted 2024 December 8.

Abstract

Objectives

To identify the right interventions for the right heart failure (HF) patients in the real-world setting using machine learning (ML) trained on individual-level clinical data linked with social determinants of health (SDOH) data.

Methods

In this retrospective cohort study, point-of-care claims data from Komodo Health and SDOH data from the National Health and Wellness Survey (NHWS), from January 2014–December 2020, were linked. Data mining was conducted using K-means clustering, an ML tool. Komodo Health data were used to access longitudinal data for the selected patient cohorts and cross-sectional data from NHWS for additional patient information. The primary outcome was HF-related hospitalizations; secondary outcomes, all-cause hospitalization and all-cause mortality. Use of digital healthcare (DHC)/non-DHC interventions and related outcomes were also assessed.

Results

The study population included 353 HF patients (mean age, 63.5 years; 57.2% women). The use of non-DHC (75.9%–81.9%) and DHC (4.0%–9.1%) interventions increased from baseline to follow-up. Overall, 17.0% of patients had HF-related hospitalizations (DHC, 6.9%; non-DHC, 16.5%) and 45.0% had all-cause hospitalization (DHC, 75.0%; non-DHC, 50.9%). Two archetypes with distinct patient profiles were identified. Archetype 1 (vs. 2) characterised by older age, greater disease severity, more comorbidities, more medication use, took steps to prevent heart attack/problems, had better lifestyle, higher HF-related hospitalizations (18.3% vs. 16.3%) and lower all-cause hospitalizations (42.9% vs. 46.3%). The trends remained the same regardless of the intervention type.

Conclusions

Identification of patient archetypes with distinct profiles can be useful to understand underlying disease subtypes, identify specific interventions, predict clinical outcomes, and define the right intervention for the right patient.

Keywords: Digital Health; Heart Failure; Machine Learning; Social Determinants of Health; Data Mining

I. Introduction

Digital healthcare (DHC) interventions have the potential to improve disease control and management, population health outcomes, and healthcare quality [1–3]. Current DHC solutions include telehealth, digital and virtual disease management platforms, modifiable risk factor technologies, dietary counselling, psychological assistance, and personalized short messaging. DHC tools are inexpensive, convenient, easy to navigate, provide accessible/concise information and secure data management leading to higher acceptability [4].

A combination of medical (point-of-care) data from clinical sources (e.g., electronic medical records, registries, insurance claims) and social determinants of health (SDOH, between-care) data from devices (e.g., smartphones and apps) can provide insights into patients’ behaviors, medication responses, lifestyle choices, and a holistic view of their healthcare journey [5]. These combined data can be used to train a machine learning (ML) model to predict responses to interventions.

DHC solutions can be safe alternatives to conventional healthcare services to manage patients with cardiovascular conditions [6]. Results from a randomized trial [7] on 765 heart failure (HF) patients suggested that remote patient management may reduce unplanned hospitalizations, morbidity, and mortality [8]. Data mining techniques can be used to discover patterns and associations in medical data to uncover solutions to existing gaps and has been used in HF studies [9]. Thus, in context of HF, this study aimed to identify the right interventions for the right patient in the real-world setting using data mining by (1) describing DHC/non-DHC interventions used by HF patients; (2) identifying HF patient archetypes according to socio-demographic, clinical characteristics, procedures, laboratory tests, patient-reported outcomes (PROs), and comorbidities; and (3) describing hospitalizations/re-hospitalizations, mortality, costs, and use of DHC/non-DHC interventions in all HF patients and archetypes.

II. Methods

1. Study Design and Data Sources

This retrospective cohort study conducted between January 2014–December 2020 in the United States (US) linked claims data from Komodo Health database [10] (Supplement A) and data from the National Health and Wellness Survey (NHWS) [11] based on probabilistic matching of first and last name, date of birth, sex, address, and Zone Improvement Plan code. The linked dataset from the Komodo database was de-identified in compliance with the Health Insurance Portability and Accountability Act; therefore, no Institutional Review Board approval was required [12,13]. The 2015–2017 NHWS database has been granted Institutional Review Board exemption status by the Pearl Institutional Review Board.

Patients identified with HF diagnosis in the Komodo database were enrolled between January 2015–December 2019. Index date was defined as the date of the first HF diagnosis during enrolment period. A baseline period of ≥1 year was defined before index date to assess baseline characteristics and exclude patients with previous HF diagnosis. A follow-up period of ≥1 year was defined after index date to assess use of interventions and outcomes (Figure 1).

Figure 1

Study design. NHWS: National Health and Wellness Survey, HF: heart failure.

2. Patient Population

Adult HF patients aged ≥18 years at index date, having ≥1 year baseline and follow-up linked data from Komodo Health and NHWS, and ≥1 diagnosis code—International Classification of Diseases 9th edition Clinical Modification (ICD-9-CM) and 10th edition (ICD-10-CM)—for HF in Komodo data during enrolment were included. Patients having ≥1 diagnosis code for HF in Komodo data during baseline or having completed NHWS ≥2 years before or after index date were excluded.

3. Study Variables

1) Primary outcome

Proportion of patients with HF-related hospitalization (≥1 record of any hospitalization for HF) at 1-year follow-up period.

2) Secondary outcomes

All-cause hospitalization and all-cause death (≥1 record of any hospitalization/death) at 1-year follow-up period; and overall healthcare costs (sum of costs for all claims) at 1-year baseline and follow-up period.

Komodo data were used to assess primary and secondary outcomes.

3) Other variables

Demographics, lifestyle, and clinical characteristics: ≤2 years before index date; NHWS data.
PROs: ≤2 years before index date; NHWS data using the Medical Outcomes Study Short Form 36 (SF-36v2 Health Survey; 36 items, physical and mental component summaries; score 0–100; higher score indicating better health outcome) [14] and the EuroQol 5-Dimension 5-Level Health Questionnaire (EQ-5D-5L; hereafter EQ-5D) [15].
Comorbidities (baseline period), interventions (DHC/non-DHC; HF medications), procedures and laboratory tests (stratified by baseline and follow-up period) using Komodo data (Supplements B–E).

4. Statistical Analysis

No formal calculation of sample size/power was performed, as hypothesis testing was not involved. Descriptive statistics were used to summarize all variables and outcomes; continuous variables as mean ± standard deviation and median (interquartile range); and categorical variables as number (%).

K-means clustering, an ML tool, was used to identify HF patient archetypes according to socio-demographic and clinical characteristics, procedures, laboratory tests, PROs, and comorbidities (Supplements F–G). Eighty-seven variables were available for analysis. Factor analysis of mixed data, a dimensionality reduction technique, was used to derive a new set of uncorrelated variables (dimensions) that reduced the number of features while preserving most information in the dataset (Supplement H). The 87 variables described eigenvalues and contribution of each dimension to the total variance and were used to characterize main archetypes identified during clustering analysis. For each archetype × variable, prevalence within the archetype and cohort, and lift score = (Prevalence within archetype)/(Prevalence within cohort) was calculated. Variables with lift score >1 (i.e., enriched densely within archetype than cohort) were used to describe patient archetypes (Supplement I). Then, standard descriptive statistics were used to analyse outcomes within each archetype, including hospitalizations/re-hospitalizations and mortality, and overall healthcare costs. All outcomes were analysed at 1-year follow-up.

All analyses were performed on non-missing and valid values. Missing data were not imputed and reported as number (%) in a separate category. Statistical analyses were conducted using R software (version 3.6.3; https://www.r-project.org/).

III. Results

1. Baseline Characteristics

Of 6,915 HF patients with Komodo and NHWS linked data, 1,472 patients (age of 18+ years) with two closed claims (2014–2020), including ≥1 HF claim (2015–2019), with ≥1 year of continuous enrolment before index date were identified. Of these, 353 patients with no HF visit during baseline period, with ≥1 year follow-up, who completed NHWS ≤2 years before index date, were included for analysis (Figure 2).

Figure 2

Patient selection flowchart. HF: heart failure, NHWS: National Health and Wellness Survey.

Overall, mean age was 63.5 ± 12.18 years; 57.2% were women. Most patients (76.8%) were White (Caucasian); without university education (72.0%); 50.1% were retired, 28.7% had <$25,000 income. Patients with HF diagnosis at index date in inpatient and non-inpatient setting had similar characteristics. Most patients were overweight/obese (78.3%), had not exercised in the previous month (56.4%), were former smokers/non-smokers (78.4%), and consumed alcohol once a month/abstained (63.7%); 42.5% reported taking steps to prevent heart attack/problems (Table 1).

Table 1

Baseline socio-demographic characteristics (closed claims)

2. Patient-Reported Outcomes

Mean physical and mental component scores on SF-36v2 were 41.99 and 48.80, respectively (Table 2). Using EQ-5D instrument, mean utility index value in the total population was 0.752 (median = 0.79). Overall, 61.8% patients reported difficulties in walking and 57.1% faced challenges in usual activities.

Table 2

Baseline quality of life measures

3. Comorbidities

Both cardiac and non-cardiac comorbidities were observed. The most common (>20%) were obesity (51%), systemic hypertension (47.9%), diabetes mellitus (DM; 39.4%), and coronary heart disease (CHD; 21.8%). Others (10.5%–16.4%) included chronic kidney failure (CKF), chronic obstructive pulmonary disease (COPD), sleep apnea, and anemia (Supplement J).

4. DHC and Non-DHC Interventions

Proportion of patients using DHC (all telehealth interventions) increased from 4% at baseline to 9.1% during follow-up. The proportion of those using non-DHC interventions at baseline (75.9%) increased to 81.9% during post-index period. At baseline, the most frequently used medications were low-density lipoprotein (LDL)-cholesterol-lowering therapy (45.3%), diuretics (38.2%), beta-blockers (37.7%), and angiotensin-converting enzyme inhibitors (28.3%) (Table 3).

Table 3

Interventions of interest by archetype

5. Association between Hospitalization, Costs, and Use of Different Interventions

Overall, 45.0% patients had all-cause hospitalization/re-hospitalizations and 17.0% had HF-related hospitalizations. A lower rate (6.9%) of HF-related hospitalizations were reported following any DHC intervention post-index date than in overall population and other treatment groups (non-DHC interventions, 16.5%; medications, 15.2%). However, the subgroup with DHC interventions had considerably higher rate (75.0%) of all-cause hospitalisations than other treatment groups (non-DHC interventions, 50.9%; medications, 44.9%) (Figure 3).

Figure 3

Hospitalization and re-hospitalizations in patients with HF and by subgroup. HF: heart failure, DHC: digital healthcare.

Seven patients (2%) died at 1-year post-HF diagnosis (Figure 4). However, due to the small sample size, these estimates are likely unreliable and were not subjected to further analyses.

Figure 4

Mortality in patients with heart failure and subgroups of interest. DHC: digital healthcare.

At baseline, mean total cost was $22,240; outpatient costs accounted for 50% of total costs ($10,886), followed by pharmacy ($4,824) and inpatient costs ($3,569). The mean total cost significantly increased to $45,702 at 1-year follow-up (change from baseline, 105%), mainly comprising inpatient ($16,170), followed by outpatient ($16,002) and pharmacy costs ($5,483) (Table 4).

Table 4

Costs during 1-year baseline and follow-up period

1) Descriptive analyses of archetypes

Two archetypes of HF patients with distinct baseline characteristics were identified in Supplement H. Archetype 1 (n = 126; vs. archetype 2 [n = 227]) was characterised by higher proportion of older patients (aged >70 years; 46.8% vs. 30.8%); diagnosis in non-inpatient setting (74.6% vs. 64.8%), income ranging $25,000–$75,000 ($25,000–$50,000/$50,000–$75,000, 28.6%/19.8% vs. 23.3%/16.7%), former smokers (46.8% vs. 28.2%), those who exercised regularly (46.8% vs. 41.9%), took steps to prevent heart attacks/problems, including consuming low-fat diet (55.6% vs. 0%), stress management (39.7% vs. 0.4%), had generally higher prevalence of comorbidities (both cardiovascular and non-cardiovascular), and use of laboratory tests (electrocardiogram, creatinine, etc.); also, when compared with total population (Table 5).

Table 5

Characterization of each archetype according to clustering variables — lift values

Medication use was more frequent in archetype 1 than archetype 2 at baseline (73.0% vs. 68.7%; p = 0.398); a significant difference in use of LDL-cholesterol lowering therapy (54.8% vs. 40.1%; p = 0.008) and angiotensin II receptor blockers (ARB; 25.4% vs. 15.4%; p = 0.022) was observed. During follow-up, medication use increased, especially in archetype 2, and became similar in both archetypes, except for a significant difference in ARB use (32.5% vs. 18.9%; p = 0.004) (Table 3).

2) Association between study outcomes and use of different interventions — by archetype

Patients in archetype 1 tended to have more HF-related hospitalizations than those in archetype 2 (18.3% vs. 16.3%); trend remained same regardless of intervention. HF-related hospitalizations were particularly higher among those who had inpatient index diagnosis (28.1% vs. 23.8% in archetype 2) (Figure 5A).

Figure 5

(A) Heart failure-related hospitalizations and (b) all-cause hospitalizations by archetype. DHC: digital healthcare.

Overall, lower proportion of patients in archetype 1 had all-cause hospitalizations than in archetype 2 (42.9% vs. 46.3%). Archetype 1 was characterized by less all-cause hospitalizations than archetype 2 in DHC (73.3% vs. 76.5%), non-DHC (46.7% vs. 53.3%), medications (40.8% vs. 47.4%) subgroups, and in those with inpatient HF diagnosis at index date (56.3% vs. 65.0%) (Figure 5B).

3) Overall healthcare costs by archetype

Archetype 1 was characterized by higher total costs than archetype 2 ($28,049 vs. $18,932; p = 0.066) (Table 4) and total population during baseline and follow-up periods; difference was significantly higher during follow-up ($62,023 vs. $36,409; p = 0.022). Inpatient costs were higher in archetype 1 than archetype 2 ($24,802 vs. $11,256; p = 0.050); this large difference was borderline significant due to high variability in overall data (standard deviation = 60,109); findings were similar for pharmacy costs.

IV. Discussion

The key challenge in using retrospective databases to explore the potential of DHC interventions in the real world is the poor availability of between-care and point-of-care data that co-exist at an individual-level longitudinal detail. Unlike most studies that only included data on clinical, laboratory, and healthcare use (except one that included QoL variables) [16], our study also considered social determinants of health as they can impact chronic disease management and enable a more targeted intervention. This combined dataset provides a more holistic view of the patient and clues to their behaviors between points of care, which is necessary to identify the patient profile most likely to benefit from chronic care interventions.

Our study population included 353 relatively young HF patients (63.5 years), with a higher proportion of women compared to the general HF population. Horiuchi et al. [17] estimated an average age of 73 years and a proportion of 65% men in their cohort of HF patients. A systematic review of methods to identify HF patients in general practice reported a weighted mean age of 75 years [18]. Another systematic review of cost-of-illness studies in adults with HF in the United States reported mean age 59–84 years, with most studies estimating an average age of ≥70 years [19]. The inclusion of relatively young patients was likely due to lower participation of older patients in the NHWS.

The SF-36v2 scores at baseline suggested moderate deterioration in the physical health status and mild impairment in the mental HRQoL. The EQ-5D instrument and mean utility index value in the total population corresponded to mild HF severity at baseline [20]. Further analysis showed that the mild HRQoL deterioration was driven by challenges in mobility and usual activities. Most patients had co-existing obesity, systemic hypertension, DM, CHD, COPD, sleep apnea, anemia, iron deficiency, atrial fibrillation and flutter, CKF, valvular heart disease, thyroid disorders, anxiety, and depression, which is consistent with the most frequent HF-associated comorbidities [21–23].

According to a recent report of the United States Department of Health and Human Services, the COVID-19 pandemic led to increased DHC utilisation [24], also observed in our study (pre- vs. post-index date, 4.0% vs. 9.1%). This increase was expected in the index period, as patients were more likely to utilise DHC following disease progression or increased severity or on development of comorbidities, complications, or mobility problems. DHC interventions were associated with lower HF-related hospitalizations than in the total population and other subgroups, but with higher all-cause hospitalizations than with non-DHC interventions, possibly due to prevalence of non-cardiovascular comorbidities. The higher rate should be interpreted with caution and is likely overestimated as patients can have multiple types of interventions.

The total healthcare costs increased significantly from baseline ($22,240) to follow-up ($45,702), mainly driven by inpatient costs. Overall, 45.0% patients had all-cause hospitalizations; and 17.0%, HF-related hospitalizations. In another analysis, the high costs (mean total costs, $62,615; HF-related costs, $35,329) in the year following HF worsening in patients with HF with reduced ejection fraction (HFrEF) were attributed to inpatient encounters [25]. Similarly, another study reported a significant (p < 0.001) change in the mean costs/person in the year after HF diagnosis ($34,372) than in the year before diagnosis ($8,219) [26]. However, HF-associated costs in similar published studies have varied widely [19]. In a systematic review of HF-associated costs in the US, HF-specific hospitalizations (median cost/patient, $15,879) accounted for the increase in annual median total costs for HF care ($24,383). Costs were largely driven by the length of stay and varied based on patient characteristics (e.g., comorbidities) [19]. A review synthesizing international cost estimates of cardiovascular events also reported lower costs [27]. The average cost of HF hospitalization across studies was $11,686 (median, $10,291). Costs from US claims analysis were high ($27,006); and follow-up costs through 1 year, $12,931 (median, $15,238) [27].

ML tools enable the use and analysis of large datasets to examine multiple clinical features and identify trends in disease progression and prognosis within a patient population [7]. Several clustering analyses have been conducted to identify HF phenotypes/subclasses and comorbidity patterns in HF patients [16,28]. The K-means clustering method is one of the most adopted methods of clustering in real-world evidence studies due to its simplicity and performance [29].

Although two archetypes were identified, there was a low separation between them due to a homogenous study population and relatively small sample size. Archetype 1 may comprise HFrEF patients as indicated by greater disease severity, more comorbidities, and significantly higher ARB prescription at baseline and follow up versus archetype 2. Furthermore, archetype 1 had slightly higher HF-related hospitalizations and lower all-cause hospitalizations. Better lifestyle and higher rate of heart disease prevention practices in archetype 1 may have contributed to patients’ general wellbeing, causing lower all-cause hospitalizations. Costs associated with archetype 1 were significantly higher than for archetype 2, possibly due to older age and higher comorbidities, use of laboratory tests, and medications. High variability between archetypes in terms of costs could be due to different follow-up durations, disease severity, or comorbidities. A more in-depth analysis of comorbidities, medication, New York Heart Association classes, and study outcomes stratified by HF subtypes is needed to investigate these hypotheses.

Similar studies have examined relationships between patient profiles in different HF archetypes and outcomes. In a Swedish registry study, of the four distinct HF patient clusters differing significantly in outcomes and therapeutic response, the two clusters with the lowest 1-year survival rates were characterized by older age, low body mass index, high blood pressure, prior strokes/transient ischemic attacks, more comorbidities, low β-blocker, angiotensin-converting-enzyme inhibitors and implanted devices, and high diuretics, nitrates, and digoxin uptake; patients were least likely to have a university degree and had the lowest income [30]. As methods to diagnose/identify HF patients may differ across countries, our results may not be generalizable. Additionally, HF management choices are influenced by local guidelines, therefore treatment patterns, interventions, and costs may also differ.

Our work may set a new framework for generating data driven ML approaches that link point-of-care data with SDOH data. This deterministic linking of datasets can help to obtain insights from a more complete historical data of the patient. Furthermore, the availability of such linked datasets in the future can be improved with big data technologies, which can help to obtain better insights from patient populations.

Our study had several limitations. First, the linkage of claims data with online survey data required active patient participation, which could have been challenging for severely ill/elderly patients. Therefore, the study population (younger, healthier, fewer comorbidities) may not be representative of the general US HF population. Second, US claims database enables inclusion of many patients, especially for a medical condition like HF, but sample size was reduced due to linkage with NHWS data. However, the linkage was necessary to obtain variables not available in the claims database for better patient characterization. Third, patients were selected based on any HF diagnosis, which may have led to inclusion of those with other conditions/comorbidities, impacting patients’ outcomes, including disease severity, HRQoL, healthcare resource utilization, and costs. To reduce the likelihood of including patients without a primary HF diagnosis, those with HF during the baseline period were excluded.

More clinically meaningful archetypes could have been identified using a larger sample. However, the nature of the first HF diagnosis (inpatient/non-inpatient) was considered as a proxy for disease severity in the analyses. Additionally, mortality data were available only for approximately 85% patients in the Komodo database, potentially leading to underestimation of the number of deaths. As for all claims data analysis, medication use was assessed based on prescriptions, but information on medication adherence by patient was unavailable. Finally, missing values for clinical measures in the NHWS data could be a limitation.

In conclusion, HF is associated with substantial clinical and financial burden and impacts patients’ QoL. Efforts to integrate DHC interventions as complementary to traditional face-to-face health services may improve patient outcomes, efficiency of healthcare delivery, and cost savings. Despite certain limitations, identification of two archetypes with distinct patient profiles and outcomes using K-means clustering algorithm can help to better understand underlying disease subtypes, predict clinical outcomes, and define the right intervention for the right patient. Future studies with a larger, more enriched database are warranted to generate further clinical insights using advanced analytics.

Notes

Conflict of Interest

Keni Lee is an employee of Sanofi, UK. Ramzi Argoubi is an employee of Cerner Enviza, an Oracle company, which received funding from Sanofi to conduct this study. Halley Costantino was an employee of Cerner Enviza at the time of conducting this study but is now an employee of BluePath Solutions.

Acknowledgments

This study was funded by Sanofi. Sanofi was involved in the design of the study.

The authors wish to thank Ingrid Diana Monteiro and Deepali Garg, employees of Sanofi, for assistance in writing and publication support.

Data Statement

Qualified researchers may request access to patient level data and related study documents including the clinical study report, study protocol with any amendments, statistical analysis plan, and dataset specifications. Patient level data will be anonymized, and study documents will be redacted to protect the privacy of our participants. Further details on Sanofi’s data sharing criteria, eligible studies, and process for requesting access can be found at: https://vivli.org/.

Supplementary Materials

Supplementary materials can be found via https://doi.org/10.4258/hir.2025.31.1.66.

hir-2025-31-1-66-Supplementary.pdf

References

1. Kraef C, van der Meirschen M, Free C. Digital telemedicine interventions for patients with multimorbidity: a systematic review and meta-analysis. BMJ Open 2020;10(10):e036904. https://doi.org/10.1136/bmj-open-2020-036904.

2. Timpel P, Oswald S, Schwarz PE, Harst L. Mapping the evidence on the effectiveness of telemedicine interventions in diabetes, dyslipidemia, and hypertension: an umbrella review of systematic reviews and meta-analyses. J Med Internet Res 2020;22(3):e16791. https://doi.org/10.2196/16791.

3. Seixas AA, Olaye IM, Wall SP, Dunn P. Optimizing Healthcare through digital health and wellness solutions to meet the needs of patients with chronic disease during the COVID-19 era. Front Public Health 2021;9:667654. https://doi.org/10.3389/fpubh.2021.667654.

4. Taylor ML, Thomas EE, Vitangcol K, Marx W, Campbell KL, Caffery LJ, et al. Digital health experiences reported in chronic disease management: an umbrella review of qualitative studies. J Telemed Telecare 2022;28(10):705–17. https://doi.org/10.1177/1357633X221119620.

5. Whitman A, De Lew N, Chappel A, Aysola V, Zuckerman R, Sommers BD. Addressing social determinants of health: examples of successful evidence-based strategies and current federal efforts [Internet] Washington (DC): Assistant Secretary for Planning and Evaluation; 2022. [cited at 2025 Jan 5]. Available from: https://aspe.hhs.gov/reports/sdoh-evidence-review.

6. Kaushik A, Patel S, Dubey K. Digital cardiovascular care in COVID-19 pandemic: a potential alternative? J Card Surg 2020;35(12):3545–50. https://doi.org/10.1111/jocs.15094.

7. Koehler F, Koehler K, Deckwart O, Prescher S, Wegscheider K, Winkler S, et al. Telemedical Interventional Management in Heart Failure II (TIM-HF2), a randomised, controlled trial investigating the impact of telemedicine on unplanned cardiovascular hospitalisations and mortality in heart failure patients: study design and description of the intervention. Eur J Heart Fail 2018;20(10):1485–93. https://doi.org/10.1002/ejhf.1300.

8. Koehler F, Koehler K, Prescher S, Kirwan BA, Wegscheider K, Vettorazzi E, et al. Mortality and morbidity 1 year after stopping a remote patient management intervention: extended follow-up results from the telemedical interventional management in patients with heart failure II (TIM-HF2) randomised trial. Lancet Digit Health 2020;2(1):e16–e24. https://doi.org/10.1016/S2589-7500(19)30195-5.

9. Roni RG, Tsipi H, Ofir BA, Nir S, Robert K. Disease evolution and risk-based disease trajectories in congestive heart failure patients. J Biomed Inform 2022;125:103949. https://doi.org/10.1016/j.jbi.2021.103949.

10. Komodo. Komodo’s healthcare map [Internet] New York (NY): Komodo; c2024. [cited at 2025 Jan 5]. Available from: https://www.komodohealth.com/healthcare-map.

11. Oracle Life Science. National Health Wellness Survey: patient-reported healthcare evidence [Internet] Austin (TX): Oracle; 2024. [cited at 2024 Dec 1]. Available from: https://www.oracle.com/a/ocom/docs/industries/life-sciences/factsheet-nhws-patient-reported-healthcare-evidence.pdf.

12. DeMartino JK, Wang R, Chen CY, Ahmad N, Bookhart B, Mascola L. Global implications for COVID-19 vaccine series completion: insights from real-world data from the United States. Vaccines (Basel) 2022;10(9):1561. https://doi.org/10.3390/vaccines10091561.

13. Zhou FL, Yeaw J, Karkare SU, DeKoven M, Berhanu P, Reid T. Impact of a structured patient support program on adherence and persistence in basal insulin therapy for type 2 diabetes. BMJ Open Diabetes Res Care 2018;6(1):e000593. https://doi.org/10.1136/bmjdrc-2018-000593.

14. Ware JE Jr, Sherbourne CD. The MOS 36-item short-form health survey (SF-36). I. Conceptual framework and item selection. Med Care 1992;30(6):473–83. https://doi.org/10.1097/00005650-199206000-00002.

15. Obradovic M, Lal A, Liedgens H. Validity and responsiveness of EuroQol-5 dimension (EQ-5D) versus Short Form-6 dimension (SF-6D) questionnaire in chronic pain. Health Qual Life Outcomes 2013;11:110. https://doi.org/10.1186/1477-7525-11-110.

16. Tromp J, Tay WT, Ouwerkerk W, Teng TK, Yap J, MacDonald MR, et al. Multimorbidity in patients with heart failure from 11 Asian regions: a prospective cohort study using the ASIAN-HF registry. PLoS Med 2018;15(3):e1002541. https://doi.org/10.1371/journal.pmed.1002541.

17. Horiuchi Y, Tanimoto S, Latif AH, Urayama KY, Aoki J, Yahagi K, et al. Identifying novel phenotypes of acute heart failure using cluster analysis of clinical variables. Int J Cardiol 2018;262:57–63. https://doi.org/10.1016/j.ijcard.2018.03.098.

18. Smeets M, Henrard S, Aertgeerts B, Cools F, Janssens S, Vaes B. Methods to identify heart failure patients in general practice and their impact on patient characteristics: a systematic review. Int J Cardiol 2018;257:199–206. https://doi.org/10.1016/j.ijcard.2017.06.108.

19. Urbich M, Globe G, Pantiri K, Heisen M, Bennison C, Wirtz HS, et al. A systematic review of medical costs associated with heart failure in the USA (2014–2020). Pharmacoeconomics 2020;38(11):1219–36. https://doi.org/10.1007/s40273-020-00952-0.

20. Dyer MT, Goldsmith KA, Sharples LS, Buxton MJ. A review of health utilities using the EQ-5D in studies of cardiovascular disease. Health Qual Life Outcomes 2010;8:13. https://doi.org/10.1186/1477-7525-8-13.

21. van der Wal HH, van Deursen VM, van der Meer P, Voors AA. Comorbidities in heart failure. Handb Exp Pharmacol 2017;243:35–66. https://doi.org/10.1007/164_2017_27.

22. van Deursen VM, Urso R, Laroche C, Damman K, Dahlstrom U, Tavazzi L, et al. Co-morbidities in patients with heart failure: an analysis of the European Heart Failure Pilot Survey. Eur J Heart Fail 2014;16(1):103–11. https://doi.org/10.1002/ejhf.30.

23. Widmer F. Comorbidity in heart failure. Ther Umsch 2011;68(2):103–6. https://doi.org/10.1024/0040-5930/a000127.

24. Suran M. Increased use of Medicare telehealth during the pandemic. JAMA 2022;327(4):313. https://doi.org/10.1001/jama.2021.23332.

25. Givertz MM, Yang M, Hess GP, Zhao B, Rai A, Butler J. Resource utilization and costs among patients with heart failure with reduced ejection fraction following a worsening heart failure event. ESC Heart Fail 2021;8(3):1915–23. https://doi.org/10.1002/ehf2.13155.

26. Dunlay SM, Shah ND, Shi Q, Morlan B, VanHouten H, Long KH, et al. Lifetime costs of medical care after heart failure diagnosis. Circ Cardiovasc Qual Outcomes 2011;4(1):68–75. https://doi.org/10.1161/CIRCOUTCOMES.110.957225.

27. Nicholson G, Gandra SR, Halbert RJ, Richhariya A, Nordyke RJ. Patient-level costs of major cardiovascular conditions: a review of the international literature. Clinicoecon Outcomes Res 2016;8:495–506. https://doi.org/10.2147/CEOR.S89331.

28. Ahmad T, Lund LH, Rao P, Ghosh R, Warier P, Vaccaro B, et al. Machine Learning methods improve prognostication, identify clinically distinct phenotypes, and detect heterogeneity in response to therapy in a large cohort of heart failure patients. J Am Heart Assoc 2018;7(8):e008081. https://doi.org/10.1161/JAHA.117.008081.

29. Jain A. Data clustering: 50 years beyond K-means. Pattern Recognit Lett 2010;31(8):651–66. https://doi.org/10.1016/j.patrec.2009.09.011.

30. Tsoi KK, Chan NB, Yiu KK, Poon SK, Lin B, Ho K. Machine learning clustering for blood pressure variability applied to systolic blood pressure intervention trial (SPRINT) and the Hong Kong community cohort. Hypertension 2020;76(2):569–76. Data clustering: 50 years beyond K-means. https://doi.org/10.1161/HYPERTENSIONAHA.119.14213.

Article information Continued

This is an Open Access article distributed under the terms of the Creative Commons Attribution Non-Commercial License (http://creativecommons.org/licenses/by-nc/4.0/) which permits unrestricted non-commercial use, distribution, and reproduction in any medium, provided the original work is properly cited.

Figure 1

Study design. NHWS: National Health and Wellness Survey, HF: heart failure.

Figure 2

Patient selection flowchart. HF: heart failure, NHWS: National Health and Wellness Survey.

Figure 3

Hospitalization and re-hospitalizations in patients with HF and by subgroup. HF: heart failure, DHC: digital healthcare.

Figure 4

Mortality in patients with heart failure and subgroups of interest. DHC: digital healthcare.

Figure 5

(A) Heart failure-related hospitalizations and (b) all-cause hospitalizations by archetype. DHC: digital healthcare.

Table 1

Baseline socio-demographic characteristics (closed claims)

Characteristic	Non-inpatient index diagnosis (n = 241)	Inpatient index diagnosis (n = 112)	All patients (n = 353)	Archetype 1 (n = 126)	Archetype 2 (n = 227)	p-value
Age at index date^a (yr)						<0.001
Mean ± SD	63.88 ± 12.20	62.80 ± 12.16	63.54 ± 12.18	67.02 ± 10.74	61.61 ± 12.52
Median (IQR)	67.00 (57.00–73.00)	63.00 (55.00–73.00)	65.00 (56.00–73.00)	69.00 (60.00–75.75)	63.00 (54.00–71.00)

Sex						0.637
Male	101 (41.9)	50 (44.6)	151 (42.8)	56 (44.4)	95 (41.9)
Female	140 (58.1)	62 (55.4)	202 (57.2)	70 (55.6)	132 (58.1)

Race/ethnicity						0.897
African American	29 (12.0)	9 (8.0)	38 (10.8)	12 (9.5)	26 (11.5)
American Indian	2 (0.8)	1 (0.9)	3 (0.8)	0 (0.0)	3 (1.3)
Asian	3 (1.2)	1 (0.9)	4 (1.1)	1 (0.8)	3 (1.3)
Hispanic	11 (4.6)	13 (11.6)	24 (6.8)	8 (6.3)	16 (7.0)
White	187 (77.6)	84 (75.0)	271 (76.8)	100 (79.4)	171 (75.3)
Mixed	9 (3.7)	4 (3.6)	13 (3.7)	5 (4.0)	8 (3.5)
Other	0 (0)	0 (0)	0 (0)	0 (0)	0 (0)

Marital status						0.596
Married/living with partner	121 (50.6)	59 (52.7)	180 (51.3)	67 (53.2)	113 (50.2)
Unmarried/not living with partner	118 (49.4)	53 (47.3)	171 (48.7)	59 (46.8)	112 (49.8)
Declined to answer	2	0	2	0	2

Education level						0.161
No University	174 (72.2)	80 (71.4)	254 (72.0)	85 (67.5)	169 (74.4)
University	67 (27.8)	32 (28.6)	99 (28.0)	41 (32.5)	58 (25.6)
Declined	0	0	0	0	0

Current employment status						0.179
Full time	29 (12.0)	17 (15.2)	46 (13.0)	13 (10.3)	33 (14.5)
Part time	23 (9.5)	7 (6.2)	30 (8.5)	7 (5.6)	23 (10.1)
Self-employed	16 (6.6)	8 (7.1)	24 (6.8)	7 (5.6)	17 (7.5)
Unemployed	44 (18.3)	27 (24.1)	71 (20.1)	23 (18.3)	48 (21.1)
Student	4 (1.7)	1 (0.9)	5 (1.4)	1 (0.8)	4 (1.8)
Retired	125 (51.9)	52 (46.4)	177 (50.1)	75 (59.5)	102 (44.9)

Household income (US dollar)						0.479
<25,000	63 (27.9)	32 (30.5)	95 (28.7)	28 (23.9)	67 (31.3)
25,000 to <50,000	66 (29.2)	23 (21.9)	89 (26.9)	36 (30.8)	53 (24.8)
50,000 to <75,000	35 (15.5)	28 (26.7)	63 (19.0)	25 (21.4)	38 (17.8)
75,000 to <100,000	30 (13.3)	7 (6.7)	37 (11.2)	11 (9.4)	26 (12.1)
>100,000	32 (14.2)	15 (14.3)	47 (14.2)	17 (14.5)	30 (14.0)
Declined	15	7	22	9	13

Insurance type at index date^a						0.012
Commercial	74 (30.7)	21 (27.7)	105 (29.7)	40 (31.7)	65 (28.6)
Managed Medicaid	43 (17.8)	28 (25.0)	71 (20.1)	14 (11.1)	57 (25.1)
Medicaid	0 (0)	2 (1.8)	2 (0.6)	1 (0.8)	1 (0.4)
Medicare	15 (6.2)	10 (8.9)	25 (7.1)	12 (9.5)	13 (5.7)
Medicare Advantage	87 (36.1)	35 (31.2)	122 (34.6)	52 (41.3)	70 (30.8)
TRICARE	0 (0.0)	1 (0.9)	1 (0.3)	0 (0)	1 (0.4)
Veterans Affairs	2 (0.8)	1 (0.9)	3 (0.8)	0 (0)	3 (1.3)
Worker’s Comp	0 (0)	0 (0)	0 (0)	0 (0)	0 (0)
NULL	20 (8.3)	4 (3.6)	24 (6.8)	7 (5.6)	17 (7.5)

Comorbidities
Obesity	-	-	174 (51.0)	-	-	-
Systemic hypertension	-	-	169 (47.9)	-	-
T2DM	-	-	120 (34.0)	-	-
CHD	-	-	77 (21.8)	-	-
COPD	-	-	58 (16.4)	-	-
Sleep apnea	-	-	48 (13.6)	-	-

Smoking status						0.002
Smoker	23 (9.5)	29 (25.9)	52 (14.7)	13 (10.3)	39 (17.2)
Smoker, trying to quit	19 (7.9)	5 (4.5)	24 (6.8)	10 (7.9)	14 (6.2)
Former smoker	87 (36.1)	36 (32.1)	123 (34.8)	59 (46.8)	64 (28.2)
Never smoked	112 (46.5)	42 (37.5)	154 (43.6)	44 (34.9)	110 (48.5)

Alcohol consumption						0.209
Daily	17 (7.1)	9 (8.0)	26 (7.4)	9 (7.1)	17 (7.5)
4–6 times a week	15 (6.2)	6 (5.4)	21 (5.9)	7 (5.6)	14 (6.2)
2–3 times a week	16 (6.6)	13 (11.6)	29 (8.2)	10 (7.9)	19 (8.4)
Once a week	16 (6.6)	6 (5.4)	22 (6.2)	5 (4.0)	17 (7.5)
2–3 times a month	21 (8.7)	9 (8.0)	30 (8.5)	5 (4.0)	25 (11.0)
≤1 a month	51 (21.2)	20 (17.9)	71 (20.1)	27 (21.4)	44 (19.4)
Abstains	105 (43.6)	49 (43.8)	154 (43.6)	63 (50.0)	91 (40.1)

Exercised in past month						0.576
>14 days	39 (16.2)	23 (20.5)	62 (17.6)	20 (15.9)	42 (18.5)
7–13 days	27 (11.2)	12 (10.7)	39 (11.0)	18 (14.3)	21 (9.3)
4–6 days	19 (7.9)	6 (5.4)	25 (7.1)	10 (7.9)	15 (6.6)
1–3 days	21 (8.7)	7 (6.2)	28 (7.9)	11 (8.7)	17 (7.5)
Did not exercise	135 (56.0)	64 (57.1)	199 (56.4)	67 (53.2)	132 (58.1)

BMI (kg/m²)						0.056^b
Underweight, <18.5	2 (0.9)	2 (1.9)	4 (1.2)	1 (0.8)	3 (1.4)
Normal, 18.5–24.9	48 (20.5)	22 (20.6)	70 (20.5)	16 (13.0)	54 (24.8)
Overweight, 25.0–29.9	59 (25.2)	34 (31.8)	93 (27.3)	36 (29.3)	57 (26.1)
Obese, ≥30	125 (53.4)	49 (45.8)	174 (51.0)	70 (56.9)	104 (47.7)
Declined to answer	7	5	12	3	9

Takes steps to prevent heart attack/problems						<0.001
Yes	109 (45.2)	41 (36.6)	150 (42.5)	126 (100)	24 (10.6)
No	132 (54.8)	71 (63.4)	203 (57.5)	0 (0)	203 (89.4)

Steps to prevent heart attack/problems			70 (19.8)	70 (55.6)	0 (0)	<0.001
Eat a low-fat diet	45 (18.7)	25 (22.3)	85 (24.1)	80 (63.5)	5 (2.2)	<0.001
Eat a low-sodium diet	57 (23.7)	28 (25.0)	51 (14.4)	49 (38.9)	2 (0.9)	<0.001
Exercise regularly	36 (14.9)	15 (13.4)	105 (29.7)	100 (79.4)	5 (2.2)	<0.001
Lower my blood pressure	79 (32.8)	26 (23.2)	92 (26.1)	86 (68.3)	6 (2.6)	<0.001
Lower my cholesterol	70 (29.0)	22 (19.6)	32 (9.1)	28 (22.2)	4 (1.8)	<0.001
Take aspirin	24 (10.0)	8 (7.1)	71 (20.1)	66 (52.4)	5 (2.2)	<0.001
Take baby/low dose aspirin	52 (21.6)	19 (17.0)	13 (3.7)	13 (10.3)	0 (0)	<0.001^b
Take Plavix/generic Plavix/Effient/Brilinta	10 (4.1)	3 (2.7)	76 (21.5)	67 (53.2)	9 (4.0)	<0.001
Take other prescription medication	55 (22.8)	21 (18.8)	40 (11.3)	38 (30.2)	2 (0.9)	<0.001
Quit smoking/smoke less	27 (11.2)	13 (11.6)	63 (17.8)	60 (47.6)	3 (1.3)	<0.001
Weight loss/maintain healthy weight	43 (17.8)	20 (17.9)	5 (1.4)	5 (4.0)	0 (0)	0.005 ^b
Take aspirin cardio	3 (1.2)	2 (1.8)	72 (20.4)	66 (52.4)	6 (2.6)	<0.001^b
Take vitamins/dietary supplements	54 (22.4)	18 (16.1)	51 (14.4)	50 (39.7)	1 (0.4)	<0.001
Try to manage stress	34 (14.1)	17 (15.2)	4 (1.1)	4 (3.2)	0 (0)	0.016^b
Other	2 (0.8)	2 (1.8)	2 (0.6)	0 (0)	2 (0.9)	0.540^b
None of the above	1 (0.4)	1 (0.9)	203 (57.5)	0 (0)	203 (89.4)	<0.001

Answered no to: take steps to prevent heart problems	132 (54.8)	71 (63.4)	70 (19.8)	70 (55.6)	0 (0)	<0.001

Values are presented as number (%). Percentages are calculated out of non-missing sample unless stated otherwise. Age at index date was calculated as the difference between the index date and the date of birth.

T2DM: type 2 diabetes mellitus, CHD: coronary heart disease, COPD: chronic obstructive pulmonary disease, BMI: body mass index, SD: standard deviation, IQR: interquartile range.

p-values in bold are statistically significant.

This item comes from the Komodo claims database.

Fisher exact test was used for categorical items with expected counts <5.

Table 2

Baseline quality of life measures

Characteristic	All patients (n = 353)	Archetype 1 (n = 126)	Archetype 2 (n = 227)	p-value
SF-36 Mental Component Score				0.731
Mean ± SD	48.80 ± 11.29	48.53 ± 11.43	48.96 ± 11.23
Median (IQR)	52.02 (40.91–57.99)	52.17 (39.74–57.20)	51.68 (41.49–58.23)

SF-36 Physical Component Score				0.057
Mean ± SD	41.99 ± 10.59	40.55 ± 10.12	42.79 ± 10.78
Median (IQR)	42.92 (33.68–50.57)	41.63 (32.31–48.02)	43.71 (33.93–51.48)

EQ-5D Index Score				0.519
Mean ± SD	0.752 ± 0.178	0.743 ± 0.161	0.756 ± 0.187
Median (IQR)	0.790 (0.694–0.853)	0.784 (0.649–0.827)	0.790 (0.700–0.861)

EQ-5D - Mobility				0.221^a
I have no problems walking	135 (38.2)	41 (32.5)	94 (41.4)
I have slight problems walking	88 (24.9)	34 (27.0)	54 (23.8)
I have moderate problems walking	80 (22.7)	31 (24.6)	49 (21.6)
I have severe problems walking	42 (11.9)	19 (15.1)	23 (10.1)
I am unable to walk	8 (2.3)	1 (0.8)	7 (3.1)

EQ-5D - Self-care				0.667^a
I have no problems washing or dressing myself	280 (79.3)	99 (78.6)	181 (79.7)
I have slight problems washing or dressing myself	47 (13.3)	19 (15.1)	28 (12.3)
I have moderate problems washing or dressing myself	19 (5.4)	7 (5.6)	12 (5.3)
I have severe problems washing or dressing myself	3 (0.8)	1 (0.8)	2 (0.9)
I am unable to wash or dress myself	4 (1.1)	0 (0)	4 (1.8)

EQ-5D - Usual Activities				0.588^a
I have no problems doing my usual activities	151 (42.8)	51 (40.5)	100 (44.1)
I have slight problems doing my usual activities	100 (28.3)	33 (26.2)	67 (29.5)
I have moderate problems doing my usual activities	78 (22.1)	34 (27.0)	44 (19.4)
I have severe problems doing my usual activities	21 (5.9)	7 (5.6)	14 (6.2)
I am unable to do my usual activities	3 (0.8)	1 (0.8)	2 (0.9)

EQ-5D - Pain/Discomfort				0.798^a
I have no pain or discomfort	283 (80.2)	100 (79.4)	183 (80.6)
I have slight pain or discomfort	46 (13.0)	18 (14.3)	28 (12.3)
I have moderate pain or discomfort	19 (5.4)	7 (5.6)	12 (5.3)
I have severe pain or discomfort	2 (0.6)	1 (0.8)	1 (0.4)
I have extreme pain or discomfort	3 (0.8)	0 (0)	3 (1.3)

EQ-5D - Anxiety/Depression				0.743^a
I am not anxious or depressed	295 (83.6)	105 (83.3)	190 (83.7)
I am slightly anxious or depressed	37 (10.5)	15 (11.9)	22 (9.7)
I am moderately anxious or depressed	16 (4.5)	5 (4.0)	11 (4.8)
I am severely anxious or depressed	2 (0.6)	1 (0.8)	1 (0.4)
I am extremely anxious or depressed	3 (0.8)	0 (0)	3 (1.3)

Values are presented as number (%). Percentages are calculated out of non-missing sample unless stated otherwise.

SF-36: The Medical Outcomes Study Short Form 36, EQ-5D: The EuroQol 5-Dimension 5-Level Health Questionnaire, SD: standard deviation, IQR: interquartile range.

Fisher exact test was used for categorical items with expected counts <5.

Table 3

Interventions of interest by archetype

Interventions^a	1-yr baseline period (n = 353)	1-yr F/U period (n = 353)	1-yr F/U period or until HF-related (re)hospitalization^c (n = 353)	Interventions of interest by archetype

				1-yr baseline period			1-yr F/U period

				Archetype 1 (n = 126)	Archetype 2 (n = 227)	p-value	Archetype 1 (n = 126)	Archetype 2 (n = 227)	p-value
Digital healthcare interventions	14 (4.0)	32 (9.1)	29 (8.2)	6 (4.8)	8 (3.5)	0.579^b	13 (10.3)	16 (7.0)	0.284
Telehealth	14 (4.0)	32 (9.1)	29 (8.2)	6 (4.8)	8 (3.5)	0.579^b	13 (10.3)	16 (7.0)	0.284
Diet/nutrition management	0 (0)	1 (0.3)	1 (0.3)	0 (0)	0 (0)	N/A	1 (0.8)	0 (0)	0.357^b
Exercise	0 (0)	1 (0.3)	1 (0.3)	0 (0)	0 (0)	N/A	1 (0.8)	0 (0)	0.357^b

Non-digital healthcare interventions	268 (75.9)	289 (81.9)	279 (79.0)	105 (83.3)	163 (71.8)	0.015	104 (82.5)	175 (77.1)	0.228
Smoking cessation counselling	8 (2.3)	12 (3.4)	9 (2.5)	2 (1.6)	6 (2.6)	0.717^b	5 (4.0)	4 (1.8)	0.290^b
Weight management	59 (16.7)	77 (21.8)	70 (19.8)	24 (19.0)	35 (15.4)	0.381	29 (23.0)	41 (18.1)	0.263
Clinical evaluation - ECG	208 (58.9)	234 (66.3)	226 (64.0)	88 (69.8)	120 (52.9)	0.002	90 (71.4)	136 (59.9)	0.031
Clinical evaluation - SpO₂	24 (6.8)	37 (10.5)	32 (9.1)	8 (6.3)	16 (7.0)	0.803	8 (6.3)	24 (10.6)	0.185
Clinical evaluation - blood pressure	41 (11.6)	48 (13.6)	41 (11.6)	14 (11.4)	27 (12.3)	0.796	16 (12.7)	25 (11.0)	0.636
Diet/nutrition management	7 (2.0)	8 (2.3)	8 (2.3)	4 (3.2)	3 (1.3)	0.253^b	3 (2.4)	5 (2.2)	1.000^b
Exercise	88 (24.9)	98 (27.8)	86 (24.4)	39 (31.0)	49 (21.6)	0.051	28 (22.2)	58 (25.6)	0.485
Physical/training evaluation	20 (5.7)	40 (11.3)	35 (9.9)	7 (5.6)	13 (5.7)	0.947	10 (7.9)	25 (11.0)	0.354
Psychosocial and wellness counselling and programs	16 (4.5)	24 (6.8)	20 (5.7)	4 (3.2)	12 (5.3)	0.361	6 (4.8)	14 (6.2)	0.584
Advanced care planning	10 (2.8)	16 (4.5)	12 (3.4)	6 (4.8)	4 (1.8)	0.176^b	8 (6.3)	4 (1.8)	0.031^b
Transitional care management	12 (3.4)	35 (9.9)	25 (7.1)	5 (4.0)	7 (3.1)	0.761^b	8 (6.3)	17 (7.5)	0.689

Medications	248 (70.3)	276 (78.2)	263 (74.5)	92 (73.0)	156 (68.7)	0.398	98 (77.8)	165 (72.7)	0.293
ACE inhibitors	100 (28.3)	97 (27.5)	88 (24.9)	29 (23.0)	71 (31.3)	0.099	27 (21.4)	61 (26.9)	0.257
Beta-blockers	133 (37.7)	179 (50.7)	166 (47.0)	54 (42.9)	79 (34.8)	0.135	62 (49.2)	104 (45.8)	0.541
ARB	67 (19.0)	90 (25.5)	84 (23.8)	32 (25.4)	35 (15.4)	0.022	41 (32.5)	43 (18.9)	0.004
MRA	12 (3.4)	33 (9.3)	21 (5.9)	6 (4.8)	6 (2.6)	0.360^b	8 (6.3)	13 (5.7)	0.813
ARNI	1 (0.3)	9 (2.5)	6 (1.7)	1 (0.8)	0 (0)	0.357^b	4 (3.2)	2 (0.9)	0.192^b
If channel blockers	0 (0)	1 (0.3)	0 (0)	0 (0)	0 (0)	N/A	0 (0)	0 (0)	N/A
Diuretics	135 (38.2)	179 (50.7)	165 (46.7)	50 (39.7)	85 (37.4)	0.679	59 (46.8)	106 (46.7)	0.981
Digoxin	7 (2.0)	12 (3.4)	10 (2.8)	2 (1.6)	5 (2.2)	1.000^b	3 (2.4)	7 (3.1)	1.000^b
Antiplatelets	31 (8.8)	56 (15.9)	48 (13.6)	17 (13.5)	14 (6.2)	0.020	23 (18.3)	25 (11.0)	0.057
Anticoagulants	48 (13.6)	79 (22.4)	67 (19.0)	22 (17.5)	26 (11.5)	0.115	28 (22.2)	39 (17.2)	0.247
Antiarrhythmics	12 (3.4)	22 (6.2)	19 (5.4)	7 (5.6)	5 (2.2)	0.125^b	9 (7.1)	10 (4.4)	0.275
LDL/LDC lowering therapy	160 (45.3)	182 (51.6)	167 (47.3)	69 (54.8)	91 (40.1)	0.008	67 (53.2)	100 (44.1)	0.100
SGLT2 inhibitors	4 (1.1)	11 (3.1)	10 (2.8)	2 (1.6)	2 (0.9)	0.619^b	3 (2.4)	7 (3.1)	1.000^b

Values are presented as number (%).

ECG: electrocardiogram, SpO₂: peripheral capillary oxygen saturation, ACE: angiotensin-converting enzyme, ARB: angiotensin receptor blockers, ARNI: angiotensin receptor neprilysin inhibitors, LDL: low-density lipoprotein, LDC: low-density cholesterol, SGLT2: sodium-glucose co-transporter-2; HF: heart failure, F/U: follow-up, SD: standard deviation, IQR: interquartile range.

Among digital healthcare (DHC) interventions, alcohol use therapy, smoking cessation counselling, weight management; ECG, SpO₂, blood pressure, physical/training evaluation, psychosocial and wellness counselling and programs, behavioral health, patient education, advanced care planning, chronic care management, and transitional care management as clinical evaluations were not used.

p-values in bold are statistically significant.

At least one code/prescription recorded during the period of interest.

For categorical items with samples <5 patients, Fisher exact test was used.

For patients with a HF-related (re)hospitalization within 1 year after the index date, the frequency of DHC intervention, non-DHC intervention, medication use is for the period between the index date and the date of (re)hospitalization. For patients without a HF-related (re)hospitalization within 1 year after the index date, the frequency of DHC intervention, non-DHC intervention, medication use is for the 1-year period following the index date.

Table 4

Costs during 1-year baseline and follow-up period

Total costs (US dollar)	1-yr baseline period				1-yr follow-up period

	All patients (n = 353)	Archetype 1 (n = 126)	Archetype 2 (n = 227)	p-value^b	All patients (n = 353)	Archetype 1 (n = 126)	Archetype 2 (n = 227)	p-value^b
Sample without outliers^a				N/A				N/A
Valid	328 (92.9)	119 (94.4)	209 (92.1)		328 (92.9)	119 (94.4)	209 (92.1)
Missing	25 (7.1)	7 (5.6)	18 (7.9)		25 (7.1)	7 (5.6)	18 (7.9)

Total costs				0.066				0.022
Mean ± SD	22,240 ± 43,234	28,049 ± 48,279	18,932 ± 39,824		45,702 ± 97,374	62,023 ± 120,712	36,409 ± 80,004
Median (IQR)	7,503 (2,024–22,222)	10,198 (3,235–27,545)	6,074 (1,717–19,505)		14,619 (5,092–34,688)	19,476 (6,615–47,715)	12,320 (4,562–30,568)
Min–Max	0–409,822	0–278,123	0–409,822		0–879,798	0–879,798	0–800,554

Inpatient costs				0.317				0.050
Mean ± SD	3,569 ± 16,336	4,767 ± 19,690	2,888 ± 14,082		16,170 ± 60,109	24,802 ± 76,561	11,256 ± 47,832
Median (IQR)	0 (0–0)	0 (0–0)	0 (0–0)		0 (0–1,700)	0 (0–2,463)	0 (0–1,435)
Min–Max	0–168,181	0–168,181	0–128,183		0–482,632	0–482,632	0–405,236

Emergency room costs				0.219				0.049
Mean ± SD	198 ± 550	148 ± 469	226 ± 590		319 ± 1,091	162 ± 404	409 ± 1,326
Median (IQR)	0 (0–0)	0 (0–0)	0 (0–0)		0 (0–0)	0 (0–0)	0 (0–28)
Min–Max	0–4,030	0–3,429	0–4,030		0–13,396	0–1,755	0–13,396

Outpatient costs				0.463				0.976
Mean ± SD	10,886 ± 31,545	12,583 ± 29,617	9,919 ± 32,622		16,002 ± 43,490	16,096 ± 28,267	15,948 ± 50,201
Median (IQR)	2,427 (460–8,615)	3,238 (772–10,665)	1,919 (315–7,247)		4,313 (1,171–12,912)	6,222 (1,403–19,929)	3,506 (1,065–10,979)
Min–Max	0–409,822	0–214,974	0–409,822		0–598,922	0–211,264	0–598,922

Pharmacy costs				0.192				0.064
Mean ± SD	4,824 ± 15,827	6,336 ± 20,366	3,964 ± 12,508		5,483 ± 17,200	7,816 ± 25,065	4,155 ± 10,188
Median (IQR)	30 (0–2,217)	0 (0–2,262)	44 (0–2,084)		168 (0–4,723)	161 (0–6,365)	175 (0–4,498)
Min–Max	0–182,259	0–182,259	0–113,777		0–239,958	0–239,958	0–110,207

SD: standard deviation, IQR: interquartile range.

Patients with any cost category greater than 1 million dollar were considered outliers and excluded from the cost analysis.

Statistical tests performed include t-tests for continuous variables and chi-square tests for categorical variables.

Table 5

Characterization of each archetype according to clustering variables — lift values

Characteristic	Lift value		Archetype 1 (n = 126)	Archetype 2 (n = 227)	All patients (n = 353)

	Archetype 1	Archetype 2
Age at index date (yr)
Mean ± SD	N/A	N/A	67.02 ± 10.74	61.61 ± 12.52	63.54 ± 12.18
Median (IQR)	N/A	N/A	69.00 (60.00–75.75)	63.00 (54.00–71.00)	65.00 (56.00–73.00)

10-year age groups at index date (yr)
18–29	0.5	1.3	1 (0.8)	5 (2.2)	6 (1.7)
30–39	0.3	1.4	1 (0.8)	10 (4.4)	11 (3.1)
40–49	0.5	1.3	5 (4.0)	23 (10.1)	28 (7.9)
50–59	0.8	1.1	20 (15.9)	49 (21.6)	69 (19.5)
60–69	1.0	1.0	40 (31.7)	70 (30.8)	110 (31.2)
70–79	1.2	0.9	45 (35.7)	61 (26.9)	106 (30.0)
80+	1.7	0.6	14 (11.1)	9 (4.0)	23 (6.5)

Broad age groups at index date (yr)
18–59	0.7	1.2	27 (21.4)	87 (38.3)	114 (32.3)
60–69	1.0	1.0	40 (31.7)	70 (30.8)	110 (31.2)
70+	1.3	0.8	59 (46.8)	70 (30.8)	129 (36.5)

Sex
Male	1.0	1.0	56 (44.4)	95 (41.9)	151 (42.8)
Female	1.0	1.0	70 (55.6)	132 (58.1)	202 (57.2)

Race/ethnicity
White	1.0	1.0	100 (79.4)	171 (75.3)	271 (76.8)
African American	0.9	1.1	12 (9.5)	26 (11.5)	38 (10.8)
Other	0.9	1.1	14 (11.1)	30 (13.2)	44 (12.5)

Marital status
Married/living with partner	1.0	1.0	67 (53.2)	113 (49.8)	180 (51.0)
Not married/not living with partner/declined to answer	1.0	1.0	59 (46.8)	114 (50.2)	173 (49.0)

Education level
University educated	1.2	0.9	41 (32.5)	58 (25.6)	99 (28.0)
Not University educated	0.9	1.0	85 (67.5)	169 (74.4)	254 (72.0)

Employment status
Currently employed	0.8	1.1	27 (21.4)	73 (32.2)	100 (28.3)
Currently not employed	0.9	1.1	24 (19.0)	52 (22.9)	76 (21.5)
Retired	1.2	0.9	75 (59.5)	102 (44.9)	177 (50.1)

Household income (US dollar)
<25,000	0.8	1.1	28 (22.2)	67 (29.5)	95 (26.9)
25,000 to <50,000	1.1	0.9	36 (28.6)	53 (23.3)	89 (25.2)
50,000 to <75,000	1.1	0.9	25 (19.8)	38 (16.7)	63 (17.8)
75,000 to <100,000	0.8	1.1	11 (8.7)	26 (11.5)	37 (10.5)
100,000 or more	1.0	1.0	17 (13.5)	30 (13.2)	47 (13.3)
Decline to answer	1.1	0.9	9 (7.1)	13 (5.7)	22 (6.2)

Smoking status
Current smoker	0.9	1.1	23 (18.3)	53 (23.3)	76 (21.5)
Former smoker	1.3	0.8	59 (46.8)	64 (28.2)	123 (34.8)
Never smoked	0.8	1.1	44 (34.9)	110 (48.5)	154 (43.6)

Alcohol use
Drinks	0.9	1.1	63 (50.0)	136 (59.9)	199 (56.4)
Abstains	1.1	0.9	63 (50.0)	91 (40.1)	154 (43.6)

Exercise in past month
Did exercise	1.1	1.0	59 (46.8)	95 (41.9)	154 (43.6)
Did not exercise	0.9	1.0	67 (53.2)	132 (58.1)	199 (56.4)

Insurance type at index date
Commercial	1.1	1.0	40 (31.7)	65 (28.6)	105 (29.7)
Managed Medicaid	0.6	1.2	14 (11.1)	57 (25.1)	71 (20.1)
Medicare Advantage	1.2	0.9	52 (41.3)	70 (30.8)	122 (34.6)
Other	1.0	1.0	20 (15.9)	35 (15.4)	55 (15.6)

Steps to prevent heart attack/problems
Eat a low-fat diet	2.8	0.0	70 (55.6)	0 (0.0)	70 (19.8)
Eat a low sodium diet	2.6	0.1	80 (63.5)	5 (2.2)	85 (24.1)
Exercise regularly	2.7	0.1	49 (38.9)	2 (0.9)	51 (14.4)
Lower my blood pressure	2.7	0.1	100 (79.4)	5 (2.2)	105 (29.7)
Lower my cholesterol	2.6	0.1	86 (68.3)	6 (2.6)	92 (26.1)
Take aspirin	2.4	0.2	28 (22.2)	4 (1.8)	32 (9.1)
Take baby/low dose aspirin	2.6	0.1	66 (52.4)	5 (2.2)	71 (20.1)
Take Plavix/generic Plavix/Effient/Brilinta	2.8	0.0	13 (10.3)	0 (0)	13 (3.7)
Take other medication	2.5	0.2	67 (53.2)	9 (4.0)	76 (21.5)
Quit smoking/smoke less	2.7	0.1	38 (30.2)	2 (0.9)	40 (11.3)
Weight loss/healthy weight	2.7	0.1	60 (47.6)	3 (1.3)	63 (17.8)
Take aspirin cardio	2.9	0.0	5 (4.0)	0 (0)	5 (1.4)
Take vitamins/dietary supplements	2.6	0.1	66 (52.4)	6 (2.6)	72 (20.4)
Try to manage stress	2.8	0.0	50 (39.7)	1 (0.4)	51 (14.4)
Does not take steps to prevent heart problems	0.0	1.6	0 (0)	205 (90.3)	205 (58.1)

Use of any tools or devices to manage general wellness on a regular basis
Yes	1.3	0.9	46 (36.5)	57 (25.1)	103 (29.2)
No	0.9	1.1	80 (63.5)	170 (74.9)	250 (70.8)

Nature of HF diagnosis at index date
Inpatient	0.8	1.1	32 (25.4)	80 (35.2)	112 (31.7)
Non-inpatient	1.1	0.9	94 (74.6)	136 (64.8)	241 (68.3)

EQ-5D Index Score
Mean ± SD	N/A	N/A	0.743 ± 0.161	0.756 ± 0.187	0.752 ± 0.178
Median (IQR)	N/A	N/A	0.784 (0.649–0.827)	0.790 (0.700–0.861)	0.790 (0.694–0.853)

EQ-5D Index Score
<0.5	0.8	1.1	9 (7.1)	23 (10.1)	32 (9.1)
0.5–0.6	1.7	0.6	13 (10.3)	9 (4.0)	22 (6.2)
0.6–0.7	1.0	1.0	13 (10.3)	25 (11.0)	38 (10.8)
0.7–0.8	1.1	1.0	41 (32.5)	66 (29.1)	107 (30.3)
0.8–0.9	1.0	1.0	39 (31.0)	70 (30.8)	109 (30.9)
0.9–1.0	0.7	1.2	11 (8.7)	34 (15.0)	45 (12.7)

Comorbidities
CHD	1.6	0.6	45 (35.7)	32 (14.1)	77 (21.8)
Other cerebrovascular disease or stroke	1.7	0.6	11 (8.7)	7 (3.1)	18 (5.1)
Systemic hypertension	1.1	0.9	66 (52.4)	103 (45.4)	169 (47.9)
Peripheral arterial disease	1.6	0.7	12 (9.5)	9 (4.0)	21 (5.9)
Atrial fibrillation and flutter	1.5	0.7	23 (18.3)	21 (9.3)	44 (12.5)
Other cardiac arrhythmias	1.4	0.8	18 (14.3)	17 (7.5)	35 (9.9)
T2DM	1.2	0.9	50 (39.7)	70 (30.8)	120 (34.0)
Chronic kidney failure	1.6	0.6	24 (19.0)	17 (7.5)	41 (11.6)
COPD	1.2	0.9	25 (19.8)	33 (14.5)	58 (16.4)
Valvular heart disease	1.1	0.9	16 (12.7)	24 (10.6)	40 (11.3)
Depression and anxiety	1.0	1.0	13 (10.3)	24 (10.6)	37 (10.5)
Anemia and iron deficiency	1.3	0.8	22 (17.5)	25 (11.0)	47 (13.3)
Thyroid disorders/function disorders	1.0	1.0	14 (11.1)	26 (11.5)	40 (11.3)
Sleep apnea	0.9	1.0	16 (12.7)	32 (14.1)	48 (13.6)
Fatigue	1.0	1.0	13 (10.3)	22 (9.7)	35 (9.9)
Obesity	1.1	0.9	70 (55.6)	104 (45.8)	174 (49.3)

Tests and labs
ECG	1.2	0.9	88 (69.8)	120 (52.9)	208 (58.9)
Echocardiogram	1.3	0.9	55 (43.7)	68 (30.0)	123 (34.8)
Chest X-ray	0.9	1.0	15 (11.9)	30 (13.2)	45 (12.7)
Coronary angiogram and cardiac catheterization	1.1	0.9	7 (5.6)	11 (4.8)	18 (5.1)
Exercise stress test	1.3	0.8	25 (19.8)	29 (12.8)	54 (15.3)
BNP/NO, NT-proBNP, CNP	1.3	0.8	21 (16.7)	24 (10.6)	45 (12.7)
Potassium	1.3	0.9	10 (7.9)	12 (5.3)	22 (6.2)
Magnesium	1.3	0.8	32 (25.4)	37 (16.3)	69 (19.5)
Calcium	1.1	0.9	8 (6.3)	12 (5.3)	20 (5.7)
Urea	0.9	1.1	8 (6.3)	17 (7.5)	25 (7.1)
Creatinine	1.3	0.8	46 (36.5)	54 (23.8)	100 (28.3)
Basic metabolic panel	1.1	0.9	53 (42.1)	79 (34.8)	132 (37.4)
Comprehensive metabolic panel	1.0	1.0	93 (73.8)	157 (69.2)	250 (70.8)
eGFR	1.2	0.9	18 (14.3)	24 (10.6)	42 (11.9)
CBC	1.0	1.0	93 (73.8)	158 (69.6)	251 (71.1)
Troponin	1.2	0.9	26 (20.6)	36 (15.9)	62 (17.6)
Other blood tests	1.0	1.0	22 (17.5)	37 (16.3)	59 (16.7)
Other cardiac tests	1.9	0.5	10 (7.9)	5 (2.2)	15 (4.2)

Values are presented as number (%).

EQ-5D, EuroQol 5-Dimension 5-Level Health Questionnaire; HF, heart failure, CHD: coronary heart disease, T2DM: type 2 diabetes mellitus, COPD: chronic obstructive pulmonary disease, ECG: electrocardiogram, BNP: B-type natriuretic peptide, NO NT-proBNP: N-terminal-pro-BNP, CNP: C-type natriuretic peptide, eGFR:estimated glomerular filtration rate, CBC: complete blood count, SD: standard deviation, IQR: interquartile range.

Bold values indicate a lift score >1, which means that the variable is more prevalent in the cluster than in the overall population, indicating that the variable is enriched or overrepresented in that cluster. This suggests the variable is a key characteristic of the cluster.