Abstract
Aim:
Healthcare resources usage and costs associated to nonvalvular atrial fibrillation (NVAF) were analyzed in Spain.
Methods:
This is an observational and retrospective study on patients with NVAF who started their treatment with apixaban or acenocoumarol between 1 January 2015 and 31 December 2017.
Results:
2160 patients treated with apixaban were paired (1:1) with patients treated with acenocoumarol (propensity score matching). Apixaban reduced the incidence of strokes and systemic embolisms, minor and major bleedings and deaths, versus acenocoumarol. Apixaban led to reductions of 80, 55 and 43% in costs related to nursing visits, hospitalizations, and emergency visits, respectively, leading to annual cost savings of €274/patient, from the perspective of society.
Conclusion:
Our results suggested that apixaban is a cost-effective alternative for patients with NVAF.
Keywords: acenocoumarol, apixaban, cardiovascular events, cost–effectiveness, healthcare resources
Atrial fibrillation (AF) is a supraventricular tachyarrhythmia with uncoordinated atrial electrical activation that leads to ineffective atrial contraction [1]. It is the most common cardiac arrhythmia, and its prevalence increases with age, affecting 4.4% of the population aged >40 years in Spain [2,3]. It is estimated that around 80% of AF patients have nonvalvular AF (NVAF), which is defined as AF without moderate or severe mitral stenosis, a mechanical or bioprosthetic heart valve, or mitral valve repair [4,5]. NVAF is associated with increased morbidity, as it may raise the risk of strokes up to five-times, which usually have more severity than those not related to AF [3,6,7]. Therefore, the prevention of systemic embolisms, even after the first episode of AF, is the cornerstone of treatment [7].
Oral anticoagulation with traditional vitamin K antagonists (VKA; acenocoumarol and warfarin) reduces the incidence of stroke and has an acceptable risk of bleeding compared with aspirin in patients at moderate/high risk of thromboembolic episodes [5]. However, new direct-acting oral anticoagulants (DOACs), such as non-vitamin K antagonist oral anticoagulants (including dabigatran, apixaban, rivaroxaban), showed better risk–benefit profiles than VKA in patients with NVAF in conventional clinical trials [8–10]. In addition, observational real-life studies confirmed the effectiveness and safety of DOAC in patients with NVAF, with generally favorable results for DOAC versus VKA [11–18]. DOAC were mainly associated with a lower incidence of stroke or systemic embolism and intracranial and other major bleedings in comparison to warfarin [12,13,18]. In this sense, the study carried out by Ramagopalan et al. in a Spanish population of 4,320 patients with NVAF who started anticoagulant treatment with apixaban or acenocoumarol estimated that apixaban reduced the risk of stroke and systemic embolism, and minor and major bleedings, compared with acenocoumarol [17]. In fact, the European Society of Cardiology recommends in their guidelines DOAC as the first-line treatment for NVAF [19].
DOAC and VKA have been compared in economic analyses in patients with NVAF [20,21]. A simulation study developed by Barrios et al., reported that poor coagulation control in patients treated with VKA would lead to an increase in ischemic strokes, major bleeding events and deaths in comparison to DOAC. The improvement in the prevention of cardiovascular events and deaths was associated to cost savings of around €30 million for the Spanish National Health System (SNHS) and €76 million for society [20]. In addition, a systematic review of cost–effectiveness analyses carried out by Pinyol et al., reported that according to the evaluations conducted in different countries, apixaban is generally a cost-effective alternative in comparison to VKA. However, the economic evaluations mainly considered the efficacy results of clinical trials in patients with NVAF [21].
Despite these clinical and economic results, in Spain, VKA are used in first line, and the prescription of DOAC is restricted to NVAF patients with indication of anticoagulant treatment (and no counterindications for anticoagulant treatment), able to adhere to the medication and to have a regular follow-up, and within these patients, those: a) with known hypersensitivity or with specific contraindication to the use of acenocoumarol or warfarin; b) with a history of intracranial bleeding (except during acute phase) where anticoagulation benefits surpass hemorrhagic risk; c) with ischemic stroke and high risk of cerebral hemorrhage; d) treated with VKA but suffer thromboembolic events despite having a controlled normalized international ratio (NIR); e) treated with VKA and who have good compliance with the treatment but cannot reach a good NIR control or; f) unable to access standard NIR control, according to the Therapeutic Positioning Report from the Ministry of Health, Social Services and Equality of Spain [22]. Therefore, there is a need to estimate the cost–effectiveness value of apixaban versus VKA in clinical practice to improve the care of these patients. As a consequence, our study aims to compare the use of healthcare resources and costs associated with the management of patients with NVAF treated with apixaban versus VKA, based on the results provided by Ramagopalan et al. [17] in Spain.
Materials & methods
Study design
The methods followed in this study were already published [17]. This is an observational and retrospective study based on the electronic medical records (EMRs) from the BIG-PAC® [23] database. It gathers a population of 1.8 million patients [24] from seven public health areas (primary care centers and hospitals) in seven Spanish autonomous communities. EMRs are anonymized in the centers/hospitals of origin, in compliance with the Spanish Organic Law 3/2018 of December 5 on the Protection of Personal Data and guarantee of digital rights [25]. The BIG-PAC® database demonstrated its representativeness of the Spanish population in previous studies [23,26].
Study population
The study population was patients with AF (International Classification of Diseases, 10th edition, Clinical Modification [ICD-10-CM] code: I48.91), without mitral valve heart disease (ICD-10-CM codes: I05.x, I08.0, I08.8, I08.9, I34.0, I34.8, Q20.0–Q20.5, Q20.8, Q21.0–Q21.3, Q21.8, Q21.9, I27.83, Z95.2, Z95.3), receiving new anticoagulant treatment with apixaban or VKA between 1 January 2015 and 31 December 2017 (patient recruitment period). These patients are the same from Ramagopalan et al., who compared the risk of stroke, systemic thromboembolism and bleeding after apixaban or acenocoumarol treatment after matching patients using propensity score matching [17]. Figure 1 shows the main findings of their study. The index date was defined as the date when patients started the new anticoagulant treatment for the treatment of NVAF during the recruitment period. Patients who initiated apixaban and who had had a previous switch with acenocoumarol were also included (or vice versa). The follow-up period covered from the index date until treatment interruption, the first event (hemorrhagic/ischemic stroke, thromboembolism, or major/minor bleeding), 12 months (end of study period), or death, whichever occurred first. This methodology is similar to that used in previous studies [17,18]. In Spain, the prescription of DOAC (e.g., apixaban) requires a specific authorization before starting the treatment. Patients were classified into two cohorts: apixaban and VKA.
Figure 1. . Clinical events per study group after propensity score matching.
Results expressed as percentage of equivalent events per 100-person-year (n for each study group = 2160). *Stroke and systemic embolism: n = 124 (44 vs 80); HR = 0.54 (95% CI: 0.38–0.78; p = 0.001, **Minor bleeding: n = 392 (156 vs 236); HR = 0.64 (95% CI: 0.52–0.79); p < 0.001; ***Major bleeding: n = 152 (52 vs 100); HR = 0,51 (95% CI: 0.37–0.72); p < 0.001.
DOAC: Direct-acting oral anticoagulant; HR: Hazard ratio; p: Statistical significance.
Figure created using data from Ramagopalan et al. [17]
Inclusion criteria were age ≥18 years, diagnosis of AF (ICD-10-CM code: I48.91) starting a new anticoagulant treatment between 1 January 2015 and 31 December 2017, patients in the database for ≥12 months before the index date, inclusion in the chronic prescription program (≥2 prescriptions during the follow-up period), and regular patient monitoring (≥2 health records in the database). Exclusion criteria were patients diagnosed with AF (ICD-10-CM code: I48.91) with mitral valve heart disease (ICD-10-CM codes: I05.x, I08.0, I08.8, I08.9, I34.0, I34.8, Q20.0–Q20.5, Q20.8, Q21.0–Q21.3, Q21.8, Q21.9, I27.83, Z95.2, Z95.3), subjects transferred to other centers, displaced or out-of-area; residents of nursing homes, patients with a history of AF secondary to reversible causes (thyrotoxicosis, pericarditis), heart surgery, venous thromboembolism, hip or knee surgery in the 6 weeks before the index date, valvular heart disease and/or pregnancy, subjects with valvular AF (with mechanical heart valve or moderate-severe mitral stenosis) and end-stage kidney disease, dialysis or kidney transplantation.
Patients who initiated their treatment with VKA and switched to a DOAC other than apixaban during the follow-up period were excluded.
Variables
Demographic variables & comorbidities
The sociodemographic characteristics such as age (continuous and by range) and sex were recorded [17], along with the comorbidities of patients. As a summary variable of general comorbidity, the Charlson [27] comorbidity index were used as an approximation to severity (Supplementary Table 1). These variables and the CHA2DS2-VASc and HAS-BLED [28] (range: 0–8) scores were obtained at the index date (Supplementary Tables 2 & 3) [29].
Medication administered & treatment persistence/duration
The medication was collected from dispensing records. The prescription of medications was carried out at physicians' discretion. Medications were obtained using the Anatomical Therapeutic Chemical (ATC) classification system [30]: warfarin (code: B01AA03), acenocoumarol (code: B01AA07) and apixaban (code: B01AF02). The time from the diagnosis of NVAF to the first prescription and the dosages prescribed in the record of the first prescription were estimated. Treatment persistence/duration was calculated from the index date up to 1 year, or up to the development of a new event (hemorrhagic/ischemic strokes, bleedings), the switch to another anti-platelet/anticoagulant treatment other than that which motivated inclusion (in the succeeding 30 days), or interruption/discontinuation of medication (≥60 days without renewing the medication) or death, whichever occurs first. Treatment persistence was estimated at 6 and 12 months of follow-up. The date of interruption was 30 days from the date of the last prescription.
Effectiveness
The study considered episodes of ischemic stroke and systemic thromboembolism, and major and minor bleedings (genitourinary, and other minor bleedings). Major bleedings included intracranial, gastrointestinal, and other locations (liver, eye, spleen) requiring hospital admission, defined as acute or subacute manifest bleedings accompanied by ≥1 of the following criteria [31]: a) reduction in hemoglobin levels of ≥2 g/dl; b) transfusion of ≥2 red blood cell concentrates, and/or; c) fatal bleeding. These events were identified as early as 30 days after the initiation of anticoagulant medication (apixaban vs VKA) until the date of treatment discontinuation (described above; treatment period). Episodes were classified according to the ICD-10-CM coding system. The mortality rate was calculated using the number of deaths divided by the number of patients.
Use of healthcare resources, costs, & incremental cost–effectiveness ratio
The use of healthcare resources and costs were estimated during the follow-up period. Healthcare resources included medical visits (primary care, nursing, specialist care [neurology, vascular, cardiology, internal medicine, geriatrics, endocrinology, and hematology services], emergency medical visits), hospitalizations (number and percentage of hospitalized patients, annual rates of hospitalization and length of stay), diagnostic/therapeutic tests (laboratory tests, radiology, computed tomography, nuclear magnetic resonance, catheterization, angioplasty, endarterectomy/thrombectomy, echocardiogram and Doppler echocardiography) and cardiovascular medication.
Costs were estimated based on the use of healthcare resources and the unit costs (Supplementary Table 4). Medical prescriptions were quantified according to the retail price per pack at the time of prescription. Indirect costs were estimated according to the productivity loss (non-healthcare costs), the number of days of work disability and the mean salary for the Spanish population, estimated by the National Institute of Statistics (INE) [32]. The study did not include direct non-health costs, i.e., out of pocket costs or those paid for by the patient/family, as they are not registered in the database. Costs were expressed as average annual cost per patient in euros (2021).
The incremental cost–effectiveness ratio (ICER) per ischemic stroke/thromboembolism avoided and per surviving patient was estimated as (C1-C0)/(E1-E0), being C1 the cost in the intervention group (DOAC), C0 the cost in the control group (VKA), E1 the effectiveness in the intervention group (DOAC) and E0 the effectiveness in the control group (VKA). Costs were those estimated according to the social perspective. The survival rate (defined as the time between the index date and the patient's death, which is notified by the general practitioner in the EMR) was also estimated.
Statistical analysis
The search criteria in the database were based on computer statements (SQL script). Data were reviewed, looking for possible coding or recording errors. Data were validated to ensure the quality of the results.
Descriptive-univariate statistical analyses were carried out to describe the variables in each group. Qualitative variables were described using absolute and relative frequencies (n, %), while means and standard deviations (SDs) were used for quantitative variables with symmetric distributions and medians and interquartile ranges (IQR; P25-P75) were used for those with asymmetric distributions. In addition, 95% confidence interval (CI) were estimated for population parameters.
Propensity score matching (PSM) was used to maximize the comparability of study cohorts. Each case in cohort 1 (apixaban) was matched with one patient in cohort 2 (VKA) (1:1). The procedure was the greedy nearest neighbor algorithm, with substitution and accepting a tolerance of 0.20 (caliper width). Priority was given to exact matches randomly. Group homogeneity was assessed using a logistic regression model. Once PSM was carried out, standardized coefficients (standardized differences) were provided in subsequent comparisons. The variables (estimators; covariates) included were age, sex, Charlson comorbidity index [27], CHA2DS2-VASc scores and HAS-BLED scores [28].
Bivariate analyses were conducted (ANOVA and Chi-square tests) to compare the demographic variables, comorbidities and medication in the groups of the study. In addition, these tests were used to compare the incidence rates of thromboembolic and bleeding events between both study groups. A covariance analysis (ANCOVA; generalized linear model; estimate of marginal means; Bonferroni adjustment) was used to correct costs. Covariates were age, sex, Charlson comorbidity index [27] and CHA2DS2-VASc and HAS-BLED score [28]. The 95% CI were calculated by non-parametric resampling (1000 bootstrap iterations).
The treatment persistence/duration was analyzed using a Kaplan–Meier survival analysis (procedure: log-rank test). Cox proportional risk regression was used to determine the treatment persistence and events during follow-up (hazard ratio [HR]; censored data). Percentage results were obtained, equivalent to one per 100 persons-year (incidence rate; accumulated risk). Data were censored in case of absence of the event.
The SPSSWIN version 27 statistical program was used, and values of two-sided p < 0.05 were considered statistically significant.
Ethics approval & consent to participate
This study was approved by the Ethics Committee of the Hospital of Terrassa (Barcelona) (code: 02-21-399-094) on 27 September 2021. The Strengthening the Reporting of Observational Studies in Epidemiology (STROBE) statement guidelines for reporting observational studies were followed. The individual consent was not necessary, according to Article 5 of Royal Decree 957/2020, of November 3rd, which regulates observational studies with medicines for human use [33].
Results
Study population
The study estimated that 36,542 patients with NVAF started an anticoagulation treatment (DOAC or VKA) between 1 January 2015 and 31 December 2017. Most of them (n = 32,925) were treated with apixaban and/or acenocoumarol/warfarin and were divided into those treated with acenocoumarol/warfarin (n = 30,765) and those who received apixaban (n = 2160). PSM paired 2160 patients in the apixaban group with 2160 patients treated with acenocoumarol (Figure 2). The characteristics of the population before carrying out the PSM can be seen in Supplementary Table 1.
Figure 2. . Patient flow diagram.
No differences were observed in sociodemographic characteristics, comorbidities and Charlson comorbidity index of patients treated with apixaban or acenocoumarol, highlighting the high degree of comparability between both groups after the PSM (Table 1).
Table 1. . Baseline characteristics after propensity score matching.
| Study groups PSM Patients |
Apixaban (n = 2160) | Acenocoumarol (n = 2160) | p-value | Standardized difference |
|---|---|---|---|---|
| Sociodemographic characteristics | ||||
| Mean age (SD), years | 71.2 (12.8) | 71.6 (10.1) | 0.271 | -0.041 |
| Ranges: | ||||
| 40–64 years | 20.2% | 21.6% | 0.552 | -0.014 |
| 65–74 years | 38.6% | 37.8% | 0.391 | 0.009 |
| ≥75 years | 41.2% | 40.6% | 0.475 | 0.005 |
| Sex (male) | 47.6% | 47.8% | 0.903 | -0.003 |
| Associated comorbidities | ||||
| Arterial hypertension | 70.9% | 71.1% | 0.893 | -0.021 |
| Mellitus diabetes | 31.7% | 30.6% | 0.430 | 0.012 |
| Dyslipidaemia | 45.0% | 45.6% | 0.714 | -0.006 |
| Ischemic heart disease | 23.0% | 22.2% | 0.561 | 0.009 |
| Heart failure | 20.2% | 19.8% | 0.761 | 0.004 |
| COPD | 15.0% | 15.2% | 0.865 | -0.023 |
| Cerebrovascular accident | 14.8% | 15.2% | 0.733 | -0.005 |
| Previous bleeding | 14.6% | 14.5% | 0.887 | 0.058 |
| Renal failure | 10.9% | 10.7% | 0.845 | 0.013 |
| Anemia | 10.3% | 9.7% | 0.222 | 0.022 |
| Scales | ||||
| Charlson comorbidity index (SD) | 2.5 (2.0) | 2.6 (1.9) | 0.758 | 0.016 |
| CHA2-DS2-VASc (SD) | 3.3 (1.9) | 3.4 (1.7) | 0.009 | -0.015 |
| HAS-BLED (SD) | 2 (1.3) | 2.1 (1.1) | <0.001 | -0.017 |
Values expressed as a percentage or mean (SD).
COPD: Chronic obstructive pulmonary disease; p: Statistical significance; PSM: Propensity score matching; SD: Standard deviation.
Effectiveness
Mortality rate was lower in patients who received apixaban (3.7%) in comparison to those treated with acenocoumarol (8.3%, p < 0.001) (Table 2).
Table 2. . Duration and persistence of treatments per study group after propensity score matching.
| Study groups PSM Patients |
Apixaban (n = 2160) | Acenocoumarol (n = 2160) | p-value |
|---|---|---|---|
| Time since diagnosis, years | |||
| Mean (SD) | 0.9 (1.4) | 0.9 (1.5) | 0.611 |
| Median (P25–P75) | 0.1 (0.0–1.8) | 0.1 (0.0–1.9) | |
| Treatment duration, days | |||
| Mean (SD) | 300.3 (104.0) | 266.4 (121.1) | <0.001 |
| Median (P25–P75) | 330 (216–365) | 304 (136–365) | |
| Previous use of antithrombotic medication | |||
| Acenocoumarol initial | 28.2% | 100.0% | 0.001 |
| Apixaban | 71.8% | 0.0% | 0.001 |
| Not treatment persistence at 1 year | 28.9% | 39.4% | <0.001 |
| Treatment abandonment | 5.6% | 5.2% | 0.570 |
| Dose reduction | 2.2% | 7.2% | <0.001 |
| Switch to heparin | 6.5% | 5.5% | 0.421 |
| Switch to antiplatelet | 11.3% | 4.3% | <0.001 |
| Switch to other NOAC | 3.3% | 17.2% | <0.001 |
| Treatment persistence | |||
| 6 months | 79.8% | 70.4%† | <0.001 |
| 12 months | 71.1% | 60.6%‡ | <0.001 |
| Mortality | 3.7% | 8.3% | <0.001 |
Values expressed as a percentage or mean (SD).
HR: 1.2 (95% CI: 1.0–1.3) p = 0.006.
HR: 1.3 (95% CI: 1.1–1.4) p = 0.012.
CI: Confidence interval; COPD: Chronic obstructive pulmonary disease; HR: Hazard ratio; DOAC: Non-vitamin K antagonist oral anticoagulants; p: statistical significance; P: Percentile; PSM: Propensity score matching; SD: Standard deviation.
Duration & persistence to treatments
The time from diagnosis was similar in both groups (p = 0.611), but the treatment lasted longer in patients on treatment with apixaban in comparison to those treated with acenocoumarol (300 days vs 266.4 days, p < 0.001). Therefore, the persistence to treatment was higher in patients treated with apixaban (79.8% and 71.1% at 6 and 12 months, respectively), in comparison to patients on treatment with acenocoumarol (70.4% and 60.6% at 6 and 12 months, respectively).
Use of resources
Apixaban patients required a lower use of healthcare resources compared with those needed by the acenocoumarol group. The highest differences between both groups were reported in the number of nursing visits (14.7 [SD: 9.8] vs 3 [SD: 2.3], respectively; p < 0.001), and primary care visits (10.9 [SD: 7.0] vs 6.3 [SD: 4.3], respectively; p < 0.001). Patients on treatment with apixaban had fewer specialist visits in comparison to acenocoumarol patients (p < 0.001), mostly to cardiologists and neurologists (Table 3).
Table 3. . Resource consumption per study group after propensity score matching.
| Study groups PSM Patients |
Apixaban (n = 2160) | Acenocoumarol (n = 2160) | p-value |
|---|---|---|---|
| Healthcare resources | |||
| Visits | |||
| Primary care visits | 6.3 (4.3) | 10.9 (7) | <0.001 |
| Nursing visits | 3 (2.3) | 14.7 (9.8) | <0.001 |
| Specialist visit | 2.1 (1.9) | 2.8 (2) | <0.001 |
| Cardiology | 0.4 (0.6) | 0.5 (0.6) | <0.001 |
| Intern medicine | 0.3 (0.5) | 0.4 (0.5) | 0.009 |
| Endocrinology | 0.3 (0.5) | 0.4 (0.5) | 0.003 |
| Vascular | 0.3 (0.5) | 0.4 (0.5) | <0.001 |
| Neurology | 0.4 (0.6) | 0.5 (0.6) | <0.001 |
| Hematology | 0 (0.1) | 0.4 (0.5) | <0.001 |
| Geriatrics | 0.3 (0.5) | 0.4 (0.5) | 0.023 |
| Emergency visits | 0.2 (0.4) | 0.3 (0.5) | <0.001 |
| Hospitalizations | |||
| Hospitalizations (%) | 6.6% | 10.8% | <0.001 |
| Days, mean (SD) | 0.6 (2.5) | 1.37 (4.2) | <0.001 |
| Clinical tests and procedures | |||
| Laboratory tests | 2.2 (1.6) | 3.2 (2.3) | <0.001 |
| Radiology | 0.8 (0.8) | 1 (0.8) | <0.001 |
| Computerized tomography | 0.3 (0.5) | 1 (0.8) | <0.001 |
| Nuclear magnetic resonance | 0.3 (0.5) | 0.8 (0.8) | <0.001 |
| Other tests† | 1.2 (1) | 2.4 (1.7) | <0.001 |
| Catheterization | 0.2 (0.4) | 0.2 (0.4) | 0.154 |
| Angioplasty | 0.2 (0.4) | 0.2 (0.4) | 0.709 |
| Endarterectomy | 0 (0.1) | 0.2 (0.4) | <0.001 |
| Indirect resources | |||
| Temporary labor loss | <0.001 | ||
| Labor loss (%) | 12.4% | 15.6% | |
| Days, mean (SD) | 1.7 (4.7) | 2.5 (6.3) | |
Values expressed as a percentage or mean (SD).
Includes echocardiogram and Doppler echocardiography.
p: Statistical significance; PSM: Propensity score matching; SD: Standard deviation.
It was observed that 6.6% of patients treated with apixaban were hospitalized during the follow-up period, with an average length of hospital stay of 0.6 days (SD: 2.5), whereas in patients with acenocoumarol, these results were 10.8% and 1.37 days (SD: 4.2) (Table 3).
In general, patients treated with apixaban required fewer clinical tests and procedures during the follow-up period. The healthcare resources most frequently used were laboratory tests (apixaban: 2.2 [SD: 1.6] and acenocoumarol: 3.2 [SD: 2.3]), followed by other tests (apixaban: 1.2 [SD: 1.0] and acenocoumarol: 2.4 [SD: 1.7]), which included echocardiogram and Doppler echocardiography. However, there were no differences in the use of catheterization (p = 0.154) and angioplasty (p = 0.709) (Table 3).
Productivity losses were lower in the apixaban group, in comparison to the acenocoumarol group. It was estimated that 12.4% of patients in the former group required sick leave (average: 1.7 days [SD: 4.7]), whereas 15.6% of patients in the acenocoumarol group had to take days off work because of their disease (2.5 days [SD: 6.3]) (Table 3).
Costs
Annual healthcare costs were lower in apixaban patients, in comparison to those treated with acenocoumarol. The highest annual costs were associated to hospitalizations (apixaban: €293 [SD: 1196] vs acenocoumarol: €658 [SD: 2036]; p < 0.001), followed by nursing visits (apixaban: €60 [SD: 45] vs acenocoumarol: €294 [SD: 195]; p < 0.001) and specialist visits (apixaban: €193 [SD: 175] vs acenocoumarol: €257 [SD: 185]; p < 0.001). Apixaban showed reductions of 80%, 55% and 43% in the annual costs related to nursing visits, hospitalizations and emergency visits, respectively (Table 4).
Table 4. . Annual cost (€, 2021) per patient by study group after propensity score matching.
| Study groups, PSM Patients |
Apixaban (n = 2160) | Acenocoumarol (n = 2160) | p-value | |
|---|---|---|---|---|
| Healthcare cost | ||||
| Visits | ||||
| Primary care visits | 146 (99) | 252 (163) | <0.001 | |
| Nursing visits | 60 (45) | 294 (195) | <0.001 | |
| Specialist visit | 193 (175) | 257 (185) | <0.001 | |
| Emergency visits | 21 (48) | 37 (63) | <0.001 | |
| Hospitalizations | 293 (1196) | 658 (2036) | <0.001 | |
| Clinical tests and procedures | ||||
| Laboratory tests | 48 (35) | 70 (51) | <0.001 | |
| Radiology | 15 (14) | 18 (15) | <0.001 | |
| Computerized tomography | 31 (45) | 97 (80) | <0.001 | |
| Nuclear magnetic resonance | 60 (84) | 148 (136) | <0.001 | |
| Other tests† | 43 (37) | 89 (65) | <0.001 | |
| Catheterization | 96 (192) | 104 (198) | 0.154 | |
| Angioplasty | 100 (195) | 102 (197) | 0.709 | |
| Endarterectomy | 2 (29) | 99 (194) | <0.001 | |
| Medicines | ||||
| Cardiovascular medicines cost | 899 (327) | 42 (20) | <0.001 | |
| Healthcare cost | 2008 (1490) | 2268 (2251) | <0.001 | |
| Non-healthcare cost | 169 (478) | 252 (641) | <0.001 | |
| Total cost | 2177 (1488) | 2520 (2268) | <0.001 | |
| Cost correction‡ | Apixaban | Acenocoumarol | Difference | p |
|---|---|---|---|---|
| Healthcare cost | 2046 | 2224 | -178 | 0.001 |
| 95% CI | (1986–2107) | (2128–2315) | (-282–[-73]) | |
| Non-healthcare cost | 162 | 259 | -97 | <0.001 |
| 95% CI | (145–181) | (234–286) | (-126–[-67]) | |
| Total cost | 2208 | 2482 | -274 | <0.001 |
| 95% CI | (2146–2271) | (2380–2584) | (-387–[-157]) |
Values expressed as a percentage or mean (SD).
Includes echocardiogram and Doppler echocardiography.
Covariates: age, sex, time from diagnosis and Charlson comorbidity index.
CI: Confidence interval; p: Statistical significance; PSM: Propensity score matching; SD: Standard deviation.
In general, apixaban patients had lower annual costs in most of clinical tests and procedures, mainly in endarterectomy (apixaban: €2 [SD: 29] vs acenocoumarol: €99 [SD: 195]; p < 0.001) and nuclear magnetic resonance (apixaban: €60 [SD: 84] vs acenocoumarol: €148 [SD: 136]; p < 0.001). The deepest reductions in annual healthcare costs were recorded in the costs associated to endarterectomy (97%), computed tomography (68%) and nuclear magnetic resonance (59%). However, there were no differences in the costs associated to catheterization (p = 0.154) and angioplasty (p = 0.709) (Table 4).
The annual cost associated with medicines in patients with apixaban was €899 (SD: 327), whereas in patients treated with acenocoumarol, it amounted to €42 (p < 0.001) (Table 4) (Figure 3).
Figure 3. . Cost per patient year by type of cost.
After the adjustments, the annual total healthcare cost amounted to €2046 (95% CI: 1986–2107) in patients with apixaban, whereas it was €2224 (95% CI: 2128–2315) in patients with acenocoumarol (p = 0.001). Non-healthcare costs also were lower in patients on apixaban in comparison to those who received acenocoumarol (€162 [95% CI: 145–181] vs €259 [95% CI: 234–286], p < 0.001). The annual total costs were lower in patients treated with apixaban versus the acenocoumarol group (€2208 [95% CI: 2146–2271] vs €2482 [95% CI: 2380–2584], p < 0.001). Therefore, the use of apixaban led to annual cost savings of €274 (95% CI: €157–€387) from the perspective of the society (Table 4) (Figure 4).
Figure 4. . Corrected cost per study group.
Incremental cost–effectiveness ratios
Based on the effectiveness and costs results, the ICER of apixaban versus acenocoumarol was estimated. Apixaban patients showed a reduction in the incidence of stroke and systemic embolism events, and minor and major bleedings in comparison to acenocoumarol patients. Therefore, for every 27 patients treated with apixaban, 1 minor bleeding would be avoided in patients treated with acenocoumarol, for every 46 patients treated with apixaban, 1 major bleeding would be avoided in patients treated with acenocoumarol, and for every 59 patients treated with apixaban, 1 stroke or systemic embolism would be avoided in patients treated with acenocoumarol (Table 5). In addition, it was observed that the survival rate in patients treated with apixaban was 55.4%, higher than that of acenocoumarol-treated patients during the follow-up period (Table 6).
Table 5. . Results of the CEA per event avoided.
| Event type | Study group | Total cost† | Percentage of patients with event | Effect.‡ | Incremental cost | Incremental effect. | NNT§ | ICER | Total cost/ event avoided¶ |
|---|---|---|---|---|---|---|---|---|---|
| Stroke and systemic embolism | Api. | €2208 | 0.02 | 0.98 | €2253 | ||||
| Acen. | €2482 | 0.037 | 0.963 | -€274 | 0.017 | 59 | - €16,118 | €2577 | |
| Minor bleeding | Api. | €2208 | 0.072 | 0.928 | €2379 | ||||
| Acen. | €2482 | 0.109 | 0.891 | -€274 | 0.037 | 27 | - €7,405 | €2786 | |
| Major bleeding | Api. | €2208 | 0.024 | 0.976 | €2262 | ||||
| Acen. | €2482 | 0.046 | 0.954 | -€274 | 0.022 | 46 | - €12,455 | €2602 |
Total corrected cost including healthcare and non-healthcare costs.
Estimated as the percentage of patients without events.
Estimated as the number of patients needed to treat to prevent one additional event.
Estimated as the total cost divided by the effectiveness.
Acen: Acenocoumarol; Api: Apixaban; CEA: Cost–effectiveness analysis; ICER: Incremental cost–effectiveness ratio; NNT: Number of patients needed to treat.
Table 6. . Results of the CEA per surviving patient.
| Study group | Total cost† | Mortality rate | Effectiveness (survival rate) | Incremental cost | Incremental effectiveness | ICER | Total cost per surviving patient‡ |
|---|---|---|---|---|---|---|---|
| Apixaban | €2208 | 0.037 | 0.963 | – | – | – | €2293 |
| Acenocoumarol | €2482 | 0.083 | 0.917 | -€274 | 0.046 | -€5957 | €2707 |
Total corrected cost including healthcare and non-healthcare costs.
Estimated as the total cost divided by the effectiveness.
CEA: Cost–effectiveness analysis; ICER: Incremental cost–effectiveness ratio.
In addition, apixaban patients required fewer healthcare resources than patients treated with acenocoumarol, leading to lower management and non-healthcare costs from the perspective of society. Therefore, it is considered that apixaban is a dominant alternative from the perspective of the SNHS and society (Tables 5 & 6).
Discussion
Our results showed that apixaban reduced the use of healthcare resources, particularly nursing and specialist visits for treatment follow-up, and hospitalizations. In addition, apixaban reduced healthcare costs and productivity loses, leading to annual costs savings of €274 (95% CI: 157–387) per patient with NVAF, in comparison to acenocoumarol. Lastly, the mortality rate in apixaban-treated patients was more than half-fold change lower than that of patients who had been administered acenocoumarol. Ramagopalan et al., whose results served as a basis for our study, had already observed the benefit of apixaban versus VKA on clinical outcomes (bleedings, stroke and systemic thromboembolisms risk) in patients with NVAF. Apixaban reduced the incidence of strokes and systemic embolisms and minor and major bleedings (Figure 1) (p ≤ 0.001 in all comparisons) [17], and it is thus a dominant alternative for the SNHS and society in Spain.
The cost–effectiveness of apixaban over acenocoumarol in our country was previously estimated by Barón et al. They adapted a simulation model from the UK to the Spanish perspective and showed that the administration of apixaban would avoid 18 strokes, 71 bleeding events, 2 acute myocardial infarctions, 1 systemic embolism and 23 cardiovascular deaths during the lifetime of a cohort of 1000 patients with NVAF. They estimated that apixaban would improve the survival and the quality of life of patients in comparison to acenocoumarol, and the costs per life year gained would be €13,305, whereas the cost per quality-adjusted life year would be €9,765, from the perspectives of the SNHS and society, respectively [34]. Baron et al. used a simulation (Markov model) to perform their cost–effectiveness analysis, and patient data was obtained from a randomized trial comparing apixaban versus warfarin, therefore assuming that warfarin and acenocoumarol were therapeutically equivalent. We used real-life data from a bigger group of patients (2160 patients) who started treatment with apixaban or acenocoumarol between 2015 and 2017. Our results are in line with their study, suggesting that apixaban had better clinical results than acenocoumarol, leading to a reduction in the costs of disease management in these patients. Variations with Baron et al. are due to differences in the calculation of healthcare and productivity costs, as they considered higher unit costs and various severity grades of stroke. In addition, the use of healthcare resources was collected from clinical trials and registries from other countries. On the other hand, Escobar et al. developed a prevalence-based Markov model to estimate the clinical and economic impact associated to an increase in the use of DOAC versus VKA in patients with NVAF in Spain. They considered efficacy, safety, and mortality data from real-life studies and in line with our results, they showed that the use of apixaban reduced healthcare costs from the perspective of the SNHS [35].
To our knowledge, this is the first study that estimates the cost–effectiveness of apixaban versus acenocoumarol in Spanish patients with NVAF, based on data from medical practice. Nevertheless, the protocol of a study based on observational data from patients with NVAF treated with DOAC or VKA in the primary care service of the Institut Català de la Salut was recently published [36]. This study will estimate the cost–effectiveness of DOAC versus VKA from the SNHS and societal perspectives, in a regional setting. However, our results are based on the BIG-PAC® database, which records data from the overall Spanish population. Our results might be of interest to decision makers, as they complement other economic evaluations of apixaban versus acetylsalicylic acid [37], and other studies that recommend the use of DOAC over acenocoumarol and warfarin [20,38,39]. Therefore, the restrictions in the use of apixaban in Spain might be limited, improving the management of patients with NVAF.
Our study also has some limitations. First, since BIG-PAC® is an administrative database, its use in observational studies may have deficiencies, such as missing data of the study population, particularly of those who received medical care in public or private centers outside of its area of influence. Second, there may be limitations on the categorization of the NVAF, as the ICD-10-CM coding system did not allow to differentiate by type of AF (permanent, persistent and/or paroxysmal). Third, although the groups of the study were compared after carrying out a PSM, other factors not considered in this statistical procedure, such as the medication used, may have influenced the results. Fourth, we could not obtain data on the degree of control of anticoagulation in patients on acenocoumarol (time in therapeutic range), which is relevant to estimate the benefits and risks of these drugs. Fifth, primary care and nursing visits might be associated to the results of the same NIR test in the same patient, leading to an increase in the costs associated to these resources. Other limitations are those associated to the possible patient classification and prescription bias since the medication is not randomly prescribed in real-life studies.
Conclusion
Our results suggested that apixaban improves the health status of patients with NVAF, leading to a reduction in the use of healthcare resources and costs. Therefore, apixaban is cost-effective in comparison to acenocoumarol in a real-life setting in Spain. These results are considered of interest to improve clinical outcomes and quality of life for patients, while reducing the economic burden of NVAF for the SNHS and society.
Summary points
Apixaban reduced the incidence of strokes and systemic embolisms and minor and major bleedings, in comparison with acenocoumarol.
Patients treated with apixaban had a lower mortality rate than those on treatment with acenocoumarol.
After 6 and 12 months, patients treated with apixaban had higher persistence rates that those on treatment with acenocoumarol.
Patients on treatment with apixaban required fewer healthcare resources compared with those needed by the acenocoumarol, particularly nursing, primary care, and specialist visits.
Hospital admissions were less frequent and shorter in the apixaban group than in the acenocoumarol group.
Apixaban patients required fewer healthcare resources than patients treated with acenocoumarol, leading to lower management and non-healthcare costs from the perspective of society.
Apixaban reduced healthcare costs and productivity loses, leading to annual costs savings of €274 (95% CI: 157–387) per patient with NVAF, in comparison to acenocoumarol.
Due to the improvements in the clinical outcomes, and the reduction in healthcare and non-healthcare costs, apixaban is considered a dominant alternative versus acenocoumarol, from the perspective of the Spanish National Health System and society.
Supplementary Material
Footnotes
Supplementary data
To view the supplementary data that accompany this paper please visit the journal website at: https://bpl-prod.literatumonline.com/doi/10.57264/cer-2023-0007
Author contributions
The study was conceived by A Sicras Mainar, D Vilanova Larena, A Sicras Mainar, D Vilanova Larena, J Comín Colet and O Delgado Sánchez participated and contributed to the study design. Data collection and the statistical analyses were made by A Sicras Mainar. All authors interpreted the results, and critically reviewed and approved the final version of the manuscript.
Financial & competing interests disclosure
This study was funded by Pfizer and Bristol Myers Squibb. A Sicras Mainar is an employee of Atrys Health SA, who was paid consultant to BMS in connection with this manuscript. D Vilanova Larena, J Salazar-Mendiguchía, MI del Campo Alonso and A Echeto are employees of Bristol Myers Squibb and have BMS stocks. J Comín Colet and O Delgado Sánchez declare to have received fees as coordinator investigators of this study by Bristol Myers Squibb. The authors have no other relevant affiliations or financial involvement with any organization or entity with a financial interest in or financial conflict with the subject matter or materials discussed in the manuscript apart from those disclosed.
No writing assistance was utilized in the production of this manuscript.
Ethical conduct of research
This study was approved by the Ethics Committee of the Hospital of Terrassa, Barcelona, (no. 02-21-399-094) on 27 September 2021. The Strengthening the Reporting of Observational Studies in Epidemiology (STROBE) statement guidelines for reporting observational studies were followed. Individual consent was not necessary, according to Article 5 of Royal Decree 957/2020, of November 3rd, which regulates observational studies with medicines for human use.
Data sharing statement
The data that support the findings of this study are available from the corresponding author upon reasonable request.
Open access
This work is licensed under the Attribution-NonCommercial-NoDerivatives 4.0 Unported License. To view a copy of this license, visit https://creativecommons.org/licenses/by-nc-nd/4.0/
References
Papers of special note have been highlighted as: • of interest; •• of considerable interest
- 1.Hindricks G, Potpara T, Dagres N et al. 2020 ESC Guidelines for the diagnosis and management of atrial fibrillation developed in collaboration with the European Association for Cardio-Thoracic Surgery (EACTS). Eur. Heart J. 42(5), 373–498 (2021). [DOI] [PubMed] [Google Scholar]
- 2.Gómez-Doblas JJ, Muñiz J, Martin JJA et al. Prevalencia de fibrilación auricular en España. Resultados del estudio OFRECE. Rev. Esp. Cardiol. 67(4), 259–269 (2014).24774588 [Google Scholar]
- 3.Baena-Díez JM, Grau M, Forés R et al. Prevalence of atrial fibrillation and its associated factors in Spain: an analysis of six population-based studies. DARIOS Study. Rev. Clínica Esp. Engl. Ed. 214(9), 505–512 (2014). [DOI] [PubMed] [Google Scholar]
- 4.García JP, Ortega DAV, Pérez FF, Unzueta I, de Cabo SF, Chaves J. Profiling of patients with non-valvular atrial fibrillation and moderate-to-high risk of stroke not receiving oral anticoagulation in Spain. Semergen Rev. Esp. Med. Fam. (6), 396–405 (2019). [DOI] [PubMed] [Google Scholar]
- 5.January CT, Wann LS, Alpert JS et al. 2014 AHA/ACC/HRS Guideline for the Management of Patients With Atrial Fibrillation: Executive Summary. Circulation 130(23), 2071–2104 (2014). [DOI] [PubMed] [Google Scholar]
- 6.Rodríguez-Capitán J, Fernández-Meseguer A, Romero-González E, Calvo-Bonacho E, Gómez-Doblas JJ, de Teresa-Galván E. Frequency of Atrial Fibrillation in a Large Sample of Young Adults Selected From the Spanish Working Population. Rev. Espanola Cardiol. Engl. Ed 71(6), 498–500 (2018). [DOI] [PubMed] [Google Scholar]
- 7.Heidenreich PA, Estes NAM, Fonarow GC et al. 2020 Update to the 2016 ACC/AHA clinical performance and quality measures for adults with atrial fibrillation or atrial flutter: a report of the American College of Cardiology/American Heart Association Task Force on Performance Measures. Circ. Cardiovasc. Qual. Outcomes 14(1), (2021). [DOI] [PubMed] [Google Scholar]; • This document updates the previous guidelines regarding the management of patients with atrial fibrillation. It was developed by the American College of Cardiology/American Heart Association Task Force on Performance Measures.
- 8.Lip GYH, Potpara T, Boriani G, Blomström-Lundqvist C. A tailored treatment strategy: a modern approach for stroke prevention in patients with atrial fibrillation. J. Intern. Med. 279(5), 467–476 (2016). [DOI] [PubMed] [Google Scholar]
- 9.de Vries TAC, Hirsh J, Xu K et al. Apixaban for stroke prevention in atrial fibrillation: why are event rates higher in clinical practice than in randomized trials? A systematic review. Thromb. Haemost. 120(9), 1323–1329 (2020). [DOI] [PubMed] [Google Scholar]
- 10.Ruff CT, Giugliano RP, Braunwald E et al. Comparison of the efficacy and safety of new oral anticoagulants with warfarin in patients with atrial fibrillation: a meta-analysis of randomised trials. Lancet Lond. Engl. 383(9921), 955–962 (2014). [DOI] [PubMed] [Google Scholar]
- 11.Li G, Holbrook A, Jin Y et al. Comparison of treatment effect estimates of non-vitamin K antagonist oral anticoagulants versus warfarin between observational studies using propensity score methods and randomized controlled trials. Eur. J. Epidemiol. 31(6), 541–561 (2016). [DOI] [PubMed] [Google Scholar]
- 12.Gómez-Outes A, Terleira-Fernández AI, Calvo-Rojas G, Suárez-Gea ML, Vargas-Castrillón E. Dabigatran, rivaroxaban, or apixaban versus warfarin in patients with nonvalvular atrial fibrillation: a systematic review and meta-analysis of subgroups. Thrombosis 2013, 640723 (2013). [DOI] [PMC free article] [PubMed] [Google Scholar]
- 13.Lip GYH, Keshishian A, Kamble S et al. Real-world comparison of major bleeding risk among non-valvular atrial fibrillation patients initiated on apixaban, dabigatran, rivaroxaban, or warfarin. A propensity score matched analysis. Thromb. Haemost. 116(5), 975–986 (2016). [DOI] [PubMed] [Google Scholar]
- 14.Martinez BK, Sood NA, Bunz TJ, Coleman CI. Effectiveness and Safety of Apixaban, Dabigatran, and Rivaroxaban Versus Warfarin in Frail Patients With Nonvalvular Atrial Fibrillation. J. Am. Heart Assoc. 7(8), (2018). [DOI] [PMC free article] [PubMed] [Google Scholar]
- 15.Tepper PG, Mardekian J, Masseria C et al. Real-world comparison of bleeding risks among non-valvular atrial fibrillation patients prescribed apixaban, dabigatran, or rivaroxaban. PloS One 13(11), e0205989 (2018). [DOI] [PMC free article] [PubMed] [Google Scholar]
- 16.Briere J-B, Bowrin K, Coleman C et al. Real-world clinical evidence on rivaroxaban, dabigatran, and apixaban compared with vitamin K antagonists in patients with nonvalvular atrial fibrillation: a systematic literature review. Expert Rev. Pharmacoecon. Outcomes Res. 19(1), 27–36 (2019). [DOI] [PubMed] [Google Scholar]
- 17.Ramagopalan SV, Sicras-Mainar A, Polanco-Sanchez C, Carroll R, de Bobadilla JF. Patient characteristics and stroke and bleeding events in nonvalvular atrial fibrillation patients treated with apixaban and vitamin K antagonists: a Spanish real-world study. J. Comp. Eff. Res. 8(14), 1201–1212 (2019). [DOI] [PubMed] [Google Scholar]; •• This study compares the risk of strokes, systemic thromboembolisms and bleedings in patients initiating apixaban or acenocoumarol for the treatment of nonvalvular atrial fibrillation (NVAF). It is considered the first part of our research, as we focused this manuscript on the use of healthcare resources, costs and cost–effectiveness associated to NVAF.
- 18.Yao X, Abraham NS, Sangaralingham LR et al. Effectiveness and Safety of Dabigatran, Rivaroxaban, and Apixaban Versus Warfarin in Nonvalvular Atrial Fibrillation. J. Am. Heart Assoc. 5(6), (2016). [DOI] [PMC free article] [PubMed] [Google Scholar]
- 19.Kirchhof P, Benussi S, Kotecha D et al. 2016 ESC Guidelines for the management of atrial fibrillation developed in collaboration with EACTS. Eur. J. Cardiothorac. Surg. 50(5), e1–e88 (2016). [DOI] [PubMed] [Google Scholar]
- 20.Barrios V, Cinza-Sanjurjo S, Gavín O et al. Cost and burden of poor anticoagulation control with vitamin K antagonists in patients with nonvalvular atrial fibrillation in Spain. Rev. Esp. Cardiol. Engl. Ed. 74(9), 773–780 (2020). [DOI] [PubMed] [Google Scholar]; • This study aims to evaluate the burden and cost of complications due to poor anticoagulation control in patients with NVAF treated with vitamin K antagonists (VKA) in Spain. The results of this research complements our own results.
- 21.Pinyol C, Cepeda JM, Roldan I et al. A Systematic Literature Review on the cost–effectiveness of Apixaban for Stroke Prevention in Non-valvular Atrial Fibrillation. Cardiol. Ther. 5(2), 171–186 (2016). [DOI] [PMC free article] [PubMed] [Google Scholar]
- 22.Ministerio de Sanidad, Seguros Sociales e Igualdad. Informe de Posicionamiento Terapéutico. Criterios y recomendaciones generales para el uso de los anticoagulantes orales directos (ACOD) en la prevención del ictus y la embolia sistémica en pacientes con fibrilación auricular no valvular. (2016). ; •• This document describes the conditions of use and prescription of antithrombotic agents currently available in Spain. It was published by the Spanish Ministry of Health.
- 23.Sicras-Mainar A, Enriquez JL, Hernández I, Sicras-Navarro A, Aymerich T, Leon M. PMU146 Validation and representativeness of the Spanish BIG-PAC database: integrated computerized medical records for research into epidemiology, medicines and health resource use (real world evidence). Value Health 22, S734 (2019). [Google Scholar]
- 24.European Network of Centres for Pharmacoepidemiology and Pharmacovigilance. ‘Big-Pac’ (2021). http://www.encepp.eu/encepp/viewResource.htm?id=29236
- 25.Boletín Oficial del Estado. Ley Orgánica 3/2018, de 5 de diciembre, de Protección de Datos Personales y garantía de los derechos digitales. BOE number 294,(BOE-A-2018-16673) 1–68 (2018).
- 26.Sicras-Mainar A, Sicras-Navarro A, Palacios B, Varela L, Delgado JF. Epidemiología y tratamiento de la insuficiencia cardiaca en España: estudio PATHWAYS-HF. Rev. Esp. Cardiol. 149(6), 240–247 (2020). [Google Scholar]
- 27.Deyo RA, Cherkin DC, Ciol MA. Adapting a clinical comorbidity index for use with ICD-9-CM administrative databases. J. Clin. Epidemiol. 45(6), 613–619 (1992). [DOI] [PubMed] [Google Scholar]
- 28.Dzeshka MS, Lane DA, Lip GYH. Stroke and Bleeding Risk in Atrial Fibrillation: Navigating the Alphabet Soup of Risk-Score Acronyms (CHADS2, CHA2DS2-VASc, R2CHADS2, HAS-BLED, ATRIA, and More). Clin. Cardiol. 37(10), 634–644 (2014). [DOI] [PMC free article] [PubMed] [Google Scholar]
- 29.Pérez-Copete J, Esteve-Pastor MA, Roldán V, Valdés M, Marín F. Escalas de evaluación del riesgo tromboembólico y hemorrágico en la fibrilación auricular. Rev. Esp. Cardiol. Supl. 16, 25–32 (2016). [Google Scholar]
- 30.World Health Organization (WHO). ‘The Anatomical Therapeutic Chemical Classification System with Defined Daily Doses (ATC/DDD)’. https://www.who.int/standards/classifications/other-classifications/the-anatomical-therapeutic-chemical-classification-system-with-defined-daily-doses
- 31.Roskell NS, Samuel M, Noack H, Monz BU. Major bleeding in patients with atrial fibrillation receiving vitamin K antagonists: a systematic review of randomized and observational studies. EP Eur. 15(6), 787–797 (2013). [DOI] [PMC free article] [PubMed] [Google Scholar]
- 32.Instituto Nacional de Estadística. ‘Ganancia media laboral por edad y sexo.’. https://www.ine.es/dynt3/inebase/index.htm?padre=4563&capsel=4563
- 33.Ministerio de la Presidencia, Relaciones con las Cortes y Memoria Democrática. Real Decreto 957/2020, de 3 de noviembre, por el que se regulan los estudios observacionales con medicamentos de uso humano. BOE number 310,(BOE-A-2020-14960) 104907–104925 (2020).
- 34.Barón Esquivias G, Escolar Albaladejo G, Zamorano JL et al. cost–effectiveness Analysis Comparing Apixaban and Acenocoumarol in the Prevention of Stroke in Patients With Nonvalvular Atrial Fibrillation in Spain. Rev. Esp. Cardiol. Engl. Ed. 68(8), 680–690 (2015). [DOI] [PubMed] [Google Scholar]; • This is a cost–effectiveness study based on a simulation model, which compares apixaban vs acenocoumarol in the prevention of stroke in patients with nonvalvular atrial fibrillation in Spain.
- 35.Escobar Cervantes C, Martí-Almor J, Cabeza AP, Prades M, Aceituno S. PCV23 Clinical and economic impact of non-VKA anticoagulants increased use in the treatment of nonvalvular atrial fibrillation in Spain based on real-life data. Value Health 22, S545 (2019). [Google Scholar]
- 36.Hernández Rojas Z, Dalmau Llorca MR, Aguilar Martín C et al. cost–effectiveness of direct oral anticoagulants versus vitamin K antagonist in atrial fibrillation: a study protocol using Real-World Data from Catalonia (FantasTIC Study). Medicine (Baltimore) 99(36), (2020). [DOI] [PMC free article] [PubMed] [Google Scholar]; • This is the study protocol of an analysis regarding the cost–effectiveness relationship of using of direct-acting oral anticoagulants vs vitamin K antagonists to prevent ischemic stroke in patients with nonvalvular atrial fibrillation in a Spanish region.
- 37.Escolar-Albaladejo G, Barón-Esquivias G, Zamorano JL et al. Análisis coste-utilidad de apixabán frente al ácido acetilsalicílico en la prevención del ictus en pacientes con fibrilación auricular no valvular en España. Aten. Primaria 48(6), 394–405 (2016). [DOI] [PMC free article] [PubMed] [Google Scholar]
- 38.Lopes RD, Berger SE, Di Fusco M et al. A review of global health technology assessments of non-VKA oral anticoagulants in non-valvular atrial fibrillation. Int. J. Cardiol. 319, 85–93 (2020). [DOI] [PubMed] [Google Scholar]
- 39.Rubio-Terrés C, Graefenhain de Codes R, Rubio-Rodríguez D, Evers T, Grau Cerrato S. cost–effectiveness Analysis of Rivaroxaban versus Acenocoumarol in the Prevention of Stroke in Patients with Non-valvular Atrial Fibrillation in Spain. J. Health Econ. Outcomes Res. 4(1), 19–34 (2016). [DOI] [PMC free article] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.