Heart failure (HF) stands as a pressing public health issue, exerting a substantial negative impact on healthcare costs and quality of life for those affected.1–3 HF affects more than 64 million people worldwide, including 15 million in Europe.1,4,5 The Heart Failure Association Atlas 2019 report showed a median HF incidence in Europe of three per 1000 person-years for all ages and a median prevalence of 17 cases per 1000 people, while data from the EPICA study, published in 2002, reported a prevalence of 43.6 cases per 1000 people.2,6 Prevalence is expected to increase with an aging population, in addition to the extended survival observed in patients following guideline-directed medical therapy (GDMT).7–9 Recently, in Portugal, data from the PORTHOS observational study were presented showing an HF prevalence of 16.54% in the population aged 50 years or more, according to an updated and guideline-based diagnosis approach (Rui Baptista and Cristina Gavina, personal communication, December 12, 2023).

Heart failure has a major economic impact and implies high direct and indirect costs.10 Hospital care, due to the high rate of hospitalizations, is the major driver of HF costs.10 Not only does HF represent a significant economic burden to the healthcare systems, accounting for 1–3% of global health expenses, but the costs associated with in-hospital treatment are estimated to represent 39–75% of the total HF costs.3,11 More than one million hospitalizations deriving from HF have been reported in Europe and America per year, as well as readmission rates of 30% and multiple readmissions.1,12 Following an HF diagnosis, patients are hospitalized, on average, once per year.13 Globally, the number of hospitalizations may be expected to increase by up to 50% in the next 25 years, due to the aging population and the increased prevalence of comorbidities.14,15 In Portugal, 39% of the costs associated with HF derive from hospitalizations.3 As HF prevalence in Portugal is expected to increase by 30% by 2035, the total costs are also predicted to rise to EUR 503 million during the next decade.3 Simultaneously, a rise in the in-hospital mortality can be expected.16 Effective treatments that allow a reduction in HF-related hospitalizations (hHF) can play a significant role in the reduction of long-term HF costs.3

Several successful HF therapies have been developed in recent years. Among them, SGLT2 inhibitors (SGLT2i), including dapagliflozin, and empagliflozin, have been shown to reduce the risk of cardiovascular death and hospitalization in HF patients.17,18 First, SGLT2i have proven their efficacy in patients with HF and reduced ejection fraction (HFrEF).19,20 In 2021, the results of the SOLOIST-WHF trial showed the benefits of sotagliflozin, a dual SGLT2 and SGLT1 inhibitor, in reducing deaths from cardiovascular causes, hospitalizations, and urgent visits for HF in patients with diabetes and recent worsening HF, regardless of EF.21 Then, the EMPEROR-Preserved trial and DELIVER showed the benefit of SGLT2i in HF patients with mildly reduced (HfmrEF) and preserved ejection fraction (HfpEF), regardless of type 2 diabetes.22,23 Additionally, the DELIVER study showed benefits for patients who were hospitalized or had been recently hospitalized with HF, and for those with HF with improved LVEF.23,24

Both the European Society of Cardiology (ESC) and the American Heart Association/American College of Cardiology/Heart Failure Society of America (AHA/ACC/HFSA) guidelines recommend SGLT2i use in the treatment of patients with HFrEF, regardless of diabetes status (class of recommendation I, level of evidence A) due to the beneficial role in decreasing hHF and mortality.25,26 SGLT2i are also recommended for the treatment of patients with both HfmrEF and HfpEF (class of recommendation I, level of evidence A and class of recommendation II, level of evidence B, according to the ESC and AHA/ACC/HFSA guidelines, respectively).25,27

The benefit of SGLT2i across all HF patients, regardless of LVEF, represents a significant breakthrough, particularly for patients with HfpEF who had no previously effective therapeutic options except for control of comorbidities and symptomatic relief.22,23 Given the high prevalence of HfpEF, the opportunity SGLT2i bring is considerable. Therefore, considering the economic burden of HF and the robust evidence that consolidates the unique role of SGLT2i, we aimed to gather insights on the impact dapagliflozin might have on cost outcomes had it been used in a hospital setting. For this purpose, we hypothesized that the use of in-hospital dapagliflozin to all eligible patients will be associated with a decrease in healthcare costs associated with HF subsequent hospitalizations (HFHs) in six Portuguese hospitals.

Using a modeling approach that was informed by clinical study results, epidemiological data, and real-world evidence data, the aim of this paper was to project the economic impact of dapagliflozin on HFHs when started on patients who were hospitalized for HF.

Heart failure is related to significant morbidity and mortality, representing an important social and economic burden.35 The increasing aging of the population, and the improvement in healthcare including innovative secondary prevention therapy and rehabilitation programs, are factors that highly contribute to the rising incidence and prevalence of this condition.3 According to the literature, HF costs represent 2.6% of total Portuguese public health expenditure.3 Furthermore, costs related to HF admissions represent 1.6% of Portuguese hospitals’ budgets, and over 41% of these costs are attributed to patients that are re-hospitalized at least once within one year after previous discharge.31 Given this approach, the approval of therapies that significantly reduce hHF risk can have an important impact on healthcare costs.

Dapagliflozin was associated with a reduction in the number of HF hospitalizations and subsequent hospitalizations.33,34,36 In this paper, we projected the impact of in-hospital dapagliflozin use on healthcare costs related to HF readmissions by analyzing aggregated data collected from different Portuguese hospitals. The complexity of HF management in an hHF setting varies significantly, influenced by factors such as whether it is the first or a repeated event, comorbidities, the severity of the condition, the type of hospitalization and medical specialty. Therefore, for simplification, in this projection, the data were aggregated into three different approaches. Although the data used for the projection may not be representative of the national reality, this approach aimed to create a simple and comprehensive approach that reflects the economic impact of dapagliflozin use in all eligible patients in the different healthcare infrastructures (in terms of size) where an HF patient can be managed. Categorization of the data collected by the hospital sources provides a snapshot of these savings in these three different realities. We also evaluated data from HF clinics. Although these clinics can be placed in university and non-university hospitals, we also aimed to evaluate the impact of dapagliflozin in this setting since it is an increasing practice for patient management in our country. Our findings demonstrate that the use of in-hospital dapagliflozin for all eligible HF patients is associated with a significant decrease in costs in all considered approaches. Our projection model is based on the reported reduction of hHF risk associated with dapagliflozin use. Specifically, we project the HFHs cost savings associated with dapagliflozin use based on extrapolations for expected HFHs relative reductions based on data available from DAPA-HF and DELIVER clinical trials.33,34 Both DAPA-HF and DELIVER studies were multinational, double-blinded, placebo-controlled randomized clinical trials in patients with HFrEF and with mildly reduced, preserved, or improved LVEF, respectively. These studies evaluated the efficacy and safety of 10 mg dapagliflozin, daily, in comparison with placebo over a follow-up of 18.2 months (DAPA-HF) and 2.3 years (DELIVER).19,23

In scenario A, dapagliflozin was associated with HFHs cost savings ranging from EUR 32986.12, when applied to HF clinics in 2019, considering a conservative approach, to EUR 1390808.69, when considering the complex approach with data collected from 2019 in university hospitals. Similarly, if we consider the scenario B, cost savings can reach EUR 1296549.56 in the case of university hospitals in 2019. For these projections, we specifically focused on the medication expenses attributed to dapagliflozin only. This decision stemmed from our assumption that the utilization of other HF medications would likely maintain a similar rate, irrespective of dapagliflozin use. Moreover, our analysis presumed that the benefits associated with dapagliflozin remain constant regardless of variations in other patterns of HF treatment, as previously reported in the literature.37,38

In the university hospitals cluster, the use of dapagliflozin in the represented years could involve savings of EUR 813916.92 per hospital in an average approach. Considering that Portugal has seven university hospitals, savings could reach EUR 5697418.43 in these years.39 Additionally, for non-university hospitals, extrapolation we only considered hospitals equipped a cardiology unit. We estimate an average cost reduction of EUR 752377.95 per hospital resulting in a potential total cost reduction of EUR 22571338.43 for all the 30 Portuguese non-university hospitals with a structured cardiology unit.39 Notwithstanding, it is relevant to highlight that many HF patients are managed by internal medicine, and the role of internal medicine is paramount in the management of patients with acute HF. For simplification purposes, we took into account only these 30 hospitals, however even hospitals without a structured cardiology unit might benefit from cost savings associated with dapagliflozin use. Moreover, non-university hospitals vary in size, resources, and population served, and thus this value may be biased; the projection indicates the economic impact that dapagliflozin can have among Portuguese hospitals.

Regarding the HF clinics, although they have grown significantly internationally and even nationally in recent years, there is no information available on the exact number of HF clinics in our country. Therefore, we are unable to estimate the total savings that the use of dapagliflozin can generate in this context. Nevertheless, from the independent projection, our model pointed to potential significant savings. Overall, according to scenario B, Portuguese healthcare costs could be reduced in EUR 28268756.86 through the treatment with dapagliflozin to all eligible patients. It is also important to notice that even after excluding a significant percentage of patients corresponding to the HF diabetic population already treated with SGLT2i, our projections still indicate a significant cost saving associated with dapagliflozin use.

Although we did not intend it to be a cost-effectiveness study, the results from our projections are aligned with previous cost-effectiveness studies, in which dapagliflozin was a cost-effective treatment for HFrEF relative to GDMT when evaluated in various healthcare systems, including the US,40 UK, German, and Spain.41 A similar conclusion is indicated by recent cost-effectiveness studies of dapagliflozin in HfmrEF and HfpEF patients.42,43

In this projection, we considered data from HF readmissions coded as the main diagnosis, opting for conservative yet more accurate projection. Nevertheless, the cost savings could potentially be higher in clinical practice, when considering all hHF, given that hHF is often recorded as a secondary cause of readmission. Additionally, we focused only on the direct cost of patients’ readmissions to estimate cost savings. However, HF also represents a burden on patients’ quality of life, impacting the daily living activities of patients and their caregivers.10 Given the impact of dapagliflozin in improving HF-related outcomes, its implementation can also contribute to improvements in-patients’ quality of life.

We also applied our methodology to calculate the savings associated with hHF according to the Portuguese published data. Between 2012 and 2014, 51310 patients with a main diagnosis of HF have been hospitalized in Portuguese hospitals,31 leading to 68565 admissions and 665000 bed-days. The inpatient hospital services cost associated with these hospitalizations amounted to EUR 193 million, which corresponds to nearly 1.6% of the total costs from the national healthcare service.31 From the 51310 admitted patients, 22.1% had an average of 2.5 inpatient admissions per year, while 21.8% of these patients were re-hospitalized within 30 days, and 54.9% within 90 days of discharge.31 Considering this data and an average approach, the national health service could have saved up to EUR 82261202.84 if dapagliflozin had been available and implemented at that time.

On average, 19% of our sample HF patients were re-hospitalized at least once after a previous HF hospitalization. This aligns with published national results showing a subsequent hospitalizations rate of 22.10%.31 Our data are also consistent with the literature regarding the average age of patients (78 years for both) and gender distribution (51% female patients in our sample compared with 56% in published data).31 According to our data, the average length of stay for acute HF subsequent hospitalizations was approximately 10 days, which is lower than that reported in previous studies.31

Heart failure has a significant economic impact on healthcare costs, and this economic impact is expected to increase in the coming years. A considerable proportion of HF costs are attributed to hospitalizations, which are frequently recurrent. We had no intention of developing a comprehensive cost-effectiveness model, but this simplified projection based on real-world data acts as a significant addition to the translation of clinical trial data into clinical practice. This paper underscores the vital importance of implementing scientific knowledge to everyday medical practice. Our projections point to a significant cost reduction in all tested approaches, highlighting the importance of dapagliflozin use in an economic context. In addition to the reported value of dapagliflozin in improving HF outcomes, our projection clearly indicates a substantial cost reduction associated with subsequent HF hospitalizations when dapagliflozin is started during a HF hospitalization in all HF eligible patients.

Financial support for the preparation of this article was provided by AstraZeneca.

DB received speaker and consultant fees from AstraZeneca, Bayer, Bial, Bristol-Myers Squibb, Novartis, Pfizer, and Vifor Pharma. CF received speaker and consultant fees from AstraZeneca, Bayer, Bial, Boehringer-Ingelheim, CSL Vifor, Novartis, Novo Nordisk, Pfizer, Servier. RB received speaker and consultant fees from AstraZeneca, Bayer, Bial, Boehringer-Ingelheim, Novartis, Servier, and CSL Vifor. JP received speaker and consultant fees from AstraZeneca, Bayer, Bial, Boehringer-Ingelheim, GSK, Lilly, Merck, Novartis, Pfizer, Roche Diagnostics, Servier, and CSL Vifor. IM received speaker and consultant fees from AstraZeneca, Bayer, Bial, CSL Vifor, Daiichi-Sankyo, Novartis, Novo Nordisk, Pfizer, Roche Diagnostics, and Servier. SG received speaker and consultant fees from AstraZeneca, Bayer, Novartis, and Boehringer-Ingelheim. FF received speaker and consultant fees from AstraZeneca, Bayer, Novartis, Boehringer-Ingelheim and Novo Nordisk. JSC has received speaker and consultant fees, advisory board participation fees, or investigational grants from Abbott, AstraZeneca Pharmaceuticals, Bial, Boehringer-Ingelheim, Menarini, Merck Serono, Merck Sharp & Dohme, Novartis, Orion, Pfizer, Sanofi, Servier, and Vifor Pharma. ML and MA are employees of AstraZeneca Portugal (medical department). JS, RR and VL do not have any conflicts of interest to declare. None of the authors did receive any fee related to the preparation of this article.