Healthcare utilization and costs in the first two years after heart failure diagnosis : an observational study by phenotype in southwestern Sweden
Davidge, Jason LU ; Halling, Anders LU and Agvall, Björn LU
(2026)
In BMC Health Services Research
26(1).
- Abstract
Aim: Assess healthcare utilization and direct costs for heart failure (HF) during the first two years post-diagnosis across HF subgroups. Methods: This retrospective population-based study included patients with HF aged 40–90 years in Region Halland. HF subgroups were defined based on echocardiographic ejection fraction, including HF with reduced ejection fraction (HFrEF), mildly reduced ejection fraction (HFmrEF), preserved ejection fraction (HFpEF), and no defined phenotype (HF-NDP). NT-proBNP and comorbidities were analyzed as clinical features. HF phenotypes were identified via algorithmic electronic medical records analysis. Data included primary and hospital care (inpatient and outpatient) healthcare utilization and costs during a... (More)
Aim: Assess healthcare utilization and direct costs for heart failure (HF) during the first two years post-diagnosis across HF subgroups. Methods: This retrospective population-based study included patients with HF aged 40–90 years in Region Halland. HF subgroups were defined based on echocardiographic ejection fraction, including HF with reduced ejection fraction (HFrEF), mildly reduced ejection fraction (HFmrEF), preserved ejection fraction (HFpEF), and no defined phenotype (HF-NDP). NT-proBNP and comorbidities were analyzed as clinical features. HF phenotypes were identified via algorithmic electronic medical records analysis. Data included primary and hospital care (inpatient and outpatient) healthcare utilization and costs during a two-year post-diagnosis during 2015–2017. Utilization covered visits to physicians, nurses, paramedical staff, and inpatient days. Costs were estimated using the Patient Encounter Costing model. HF phenotypes, differentiated by ejection fraction, were identified via algorithmic analysis of electronic medical records. ANOVA with Bonferroni correction compared subgroups; Poisson regression assessed factors associated with longer hospital stays. Results: A total of 1769 patients were included: 472 (27%) with HFrEF, 318 (18%) HFmrEF, 505 (28%) HFpEF, and 474 (27%) HF-NDP. Hospitalizations represented the largest cost component, accounting for 68% of first-year expenditures. Average total costs per patient were €15,771 in year one and €7,459 in year two. Subgroup-specific costs declined over time: HFrEF (€18,682 to €7,083), HFpEF (€17,052 to €9,289), HFmrEF (€16,513 to €7,946), and HF-NDP (€11,009 to €5,556). HFpEF patients incurred the highest costs in the second year, indicating a sustained burden. Higher NT-proBNP levels and cardiovascular comorbidities were associated with longer lengths of stay (LoS) in both years. In year one, HFrEF served as reference, while HFmrEF had IRR 0.89 (95% CI:0.88–0.93), HFpEF IRR 0.98 (95% CI:0.94–1.01), and HF-NDP IRR 0.75 (95% CI:0.72–0.79). By year two, LoS risk increased for all other subgroups compared to HFrEF, most notably HFpEF (IRR 1.81; 95% CI:1.70–1.94; p < 0.001). Conclusions: Hospitalizations drove first-year costs, especially for HFrEF. Costs declined in year two for all subgroups, but HFpEF remained highest. Continuous care should target HFpEF and other high-risk phenotypes to reduce long-term burden.
(Less)
- author
- Davidge, Jason
LU
; Halling, Anders
LU
and Agvall, Björn
LU
- organization
- publishing date
- 2026
- type
- Contribution to journal
- publication status
- published
- subject
- keywords
- Healthcare economics, Healthcare utilization, Heart failure, Heart failure phenotypes, Pharmacotherapy
- in
- BMC Health Services Research
- volume
- 26
- issue
- 1
- article number
- 148
- publisher
- BioMed Central (BMC)
- external identifiers
-
- scopus:105028706691
- pmid:41530745
- ISSN
- 1472-6963
- DOI
- 10.1186/s12913-026-14020-4
- language
- English
- LU publication?
- yes
- id
- 420106b0-c1ad-4042-ae5a-31ab64172625
- date added to LUP
- 2026-02-17 13:24:06
- date last changed
- 2026-02-18 03:00:11
@article{420106b0-c1ad-4042-ae5a-31ab64172625,
abstract = {{<p>Aim: Assess healthcare utilization and direct costs for heart failure (HF) during the first two years post-diagnosis across HF subgroups. Methods: This retrospective population-based study included patients with HF aged 40–90 years in Region Halland. HF subgroups were defined based on echocardiographic ejection fraction, including HF with reduced ejection fraction (HFrEF), mildly reduced ejection fraction (HFmrEF), preserved ejection fraction (HFpEF), and no defined phenotype (HF-NDP). NT-proBNP and comorbidities were analyzed as clinical features. HF phenotypes were identified via algorithmic electronic medical records analysis. Data included primary and hospital care (inpatient and outpatient) healthcare utilization and costs during a two-year post-diagnosis during 2015–2017. Utilization covered visits to physicians, nurses, paramedical staff, and inpatient days. Costs were estimated using the Patient Encounter Costing model. HF phenotypes, differentiated by ejection fraction, were identified via algorithmic analysis of electronic medical records. ANOVA with Bonferroni correction compared subgroups; Poisson regression assessed factors associated with longer hospital stays. Results: A total of 1769 patients were included: 472 (27%) with HFrEF, 318 (18%) HFmrEF, 505 (28%) HFpEF, and 474 (27%) HF-NDP. Hospitalizations represented the largest cost component, accounting for 68% of first-year expenditures. Average total costs per patient were €15,771 in year one and €7,459 in year two. Subgroup-specific costs declined over time: HFrEF (€18,682 to €7,083), HFpEF (€17,052 to €9,289), HFmrEF (€16,513 to €7,946), and HF-NDP (€11,009 to €5,556). HFpEF patients incurred the highest costs in the second year, indicating a sustained burden. Higher NT-proBNP levels and cardiovascular comorbidities were associated with longer lengths of stay (LoS) in both years. In year one, HFrEF served as reference, while HFmrEF had IRR 0.89 (95% CI:0.88–0.93), HFpEF IRR 0.98 (95% CI:0.94–1.01), and HF-NDP IRR 0.75 (95% CI:0.72–0.79). By year two, LoS risk increased for all other subgroups compared to HFrEF, most notably HFpEF (IRR 1.81; 95% CI:1.70–1.94; p < 0.001). Conclusions: Hospitalizations drove first-year costs, especially for HFrEF. Costs declined in year two for all subgroups, but HFpEF remained highest. Continuous care should target HFpEF and other high-risk phenotypes to reduce long-term burden.</p>}},
author = {{Davidge, Jason and Halling, Anders and Agvall, Björn}},
issn = {{1472-6963}},
keywords = {{Healthcare economics; Healthcare utilization; Heart failure; Heart failure phenotypes; Pharmacotherapy}},
language = {{eng}},
number = {{1}},
publisher = {{BioMed Central (BMC)}},
series = {{BMC Health Services Research}},
title = {{Healthcare utilization and costs in the first two years after heart failure diagnosis : an observational study by phenotype in southwestern Sweden}},
url = {{http://dx.doi.org/10.1186/s12913-026-14020-4}},
doi = {{10.1186/s12913-026-14020-4}},
volume = {{26}},
year = {{2026}},
}