Abstract
Background:
Landmark clinical trials have expended the indications for the direct oral anticoagulants (DOACs), but contemporary data on usage and expenditure patterns are lacking. This study aimed to assess annual trends in oral anticoagulants (OACs) utilization and expenditure across the United States from 2014–2020.
Methods:
We utilized the Medical Expenditure Panel Survey (MEPS) to study the trends of use and expenditures of warfarin, dabigatran, rivaroxaban, apixaban, and edoxaban between 2014 and 2020 in the United States. Survey respondents reported OAC use within the past year, which was verified against pharmacy records. Payment information was obtained from the respondent’s pharmacy and categorized as third-party or self/out-of-pocket. Potential indications and medical conditions of interest for OAC therapy were identified from respondent-reported medical conditions. We estimated the national number of OAC users and total expenditures across age, sex, race, ethnicity, insurance, and medical condition subgroups.
Results:
Between 2014 and 2020, the number of warfarin users decreased from 3.8 million (70% of all OAC users) to 2.2 million (29% of all OAC users) while the number of DOAC users increased from 1.6 million (30% of all OAC users) to 5.4 million (70% of all OAC users). The total expenditures of OACs in the United States increased from $3.4 billion in 2014 to $17.8 billion in 2020, which was driven by the increase in DOACs expenditures.
Conclusions:
DOACs have replaced warfarin as the preferred OAC in the United States. The increased costs associated with DOACs usage may decline when generic formulations are approved.
Keywords: oral anticoagulants, pharmacoepidemiology, direct oral anticoagulants, warfarin
Introduction:
Vitamin K antagonists have served as the primary oral anticoagulant (OAC) to prevent and treat thromboembolism since 1954.1 Direct oral anticoagulants (DOACs) provide a safer and more convenient alternative to vitamin K antagonists, albeit with a greater cost and without indications in certain conditions, such as mechanical heart valve implants.1 The direct thrombin inhibitor dabigatran was the first approved DOAC by the US Food and Drug Administration (FDA) in 2010. Subsequently, the FDA approved factor Xa inhibitors: rivaroxaban in 2011, apixaban in 2012, and edoxaban in 2015.1 Since their initial indications for stroke prevention in atrial fibrillation and venous thromboembolism treatment, additional landmark clinical trials have established DOACs for the prevention of major adverse cardiovascular and limb events2,3, as part of dual and triple antithrombotic therapy after percutaneous coronary intervention4–7, in atrial fibrillation ablation8 and to prevent cancer-associated venous thromboembolism9,10. Additionally, landmark clinical trials have clarified the safety and efficacy of DOACs in patients with mechanical heart valves11–13, left ventricular assist devices, embolic stroke of undetermined source14,15, and other special populations16,17. Additionally, reversal agents have provided clinicians and patients with comfort in prescribing DOACs.18–21
In light of the rapid evolution of the DOAC landscape, contemporary estimates of DOAC and VKA usage and expenditures are needed. Moreover, several questions about national trends in OAC use broadly remain unanswered. Many studies have focused on specific populations such as Medicare or Medicaid beneficiaries or were conducted on a regional level. The overall cost burden to the healthcare system and the breakdown between patient and third-party payer (any entity that pays prescription claims on behalf of the insured) contributions are unknown. Last, it is important to determine whether DOACs use and expenditures differ across demographic and socioeconomic factors. Therefore, this study aimed to analyze the annual national trends in OAC utilization and the associated expenditure between 2014 and 2020 using a nationally representative sample.
Methods
We utilized the publicly available data from the Medical Expenditure Panel Survey (MEPS) to analyze the trends of use and expenditures of warfarin, dabigatran, rivaroxaban, apixaban, and edoxaban between 2014 and 2020.22 MEPS administers household surveys to non-institutionalized individuals to estimate medication utilization and expenditures in the United States. MEPS collects social and demographic characteristics, as well as health status and health insurance information. A single member reports information for the entire household.
Study Population
MEPS participants are a randomly selected subset of the National Health Interview Survey (NHIS). The NHIS is an annual study that serves as a source of information for the healthcare of the noninstitutionalized population. It is conducted by the U.S. Bureau on behalf of the National Center for Health Statistics, which is part of the U.S. Centers for Disease Control and Prevention.22 The complex design of MEPS includes stratification and clustering, as well as oversampling of specific subgroups to provide more accurate representation of the U.S. population.23 We included all MEPS participants who were 18 years or older in this analysis.
Medication History
The medications of any household member and their use instructions are self-reported by the designated household survey respondent during the MEPS interview.24 MEPS staff, after receiving permission, contact the household’s outpatient pharmacies. The pharmacies provide detailed medication information, which includes payments, payers, quantities, date of prescription filling, and other pertinent information. Cost data are not adjusted for inflation. Figure 1 summarizes the logic for patient identification and data utilization we used in this study.
Figure 1.
Summary of the methods
We identified patients receiving prescriptions for warfarin, dabigatran, rivaroxaban, apixaban, and edoxaban. We defined a MEPS household member as an OAC user if at least one OAC prescription was filled in the year of their interview. Each OAC user was categorized according to their most recently used OAC if they switched agents during the previous year. We obtained the actual total cost, and further determined whether the source of payment was either a third payer and/or a self/out-of-pocket expense for each prescription.
Medical Conditions
The designated household respondent reports each household member’s medical conditions to the interviewer, which are converted by trained coders to International Classification of Diseases, Ninth Revision, Clinical Modification (ICD-9-CM) codes (for 2014 and 2015) and International Classification of Diseases, Tenth Revision, Clinical Modification (ICD-10-CM) codes (for 2016 to 2020) (Table 1). We utilized 3-digit ICD codes rather than the fully specified codes in accordance with MEPS standard analytical recommendations since fully specified codes may lack accuracy due to the self-reported nature of medical conditions.24
Table 1.
Variable definitions and medical condition diagnostic codes
| Variable Name | Definition |
|---|---|
| Age | Age at end of calendar year |
| Race | Asian, Black, or White among races reported |
| Ethnicity | Self-reported as Hispanic or non-Hispanic |
| Medicare Insurance | Had Medicare coverage at any time during the preceding year |
| Medicaid Insurance | Had Medicaid insurance coverage at any time during the preceding year |
| Private Insurance | Had private insurance coverage at any time during the preceding year |
| Out-of-pocket Expenditure | Payments made by an individual of the household |
| Third-party Expenditure | All payments not made by an individual of the household |
| VTE: DVT and PE |
PE: ICD9: 415 Pulmonary embolism and infarction ICD10:I26 Pulmonary embolism DVT: ICD9: 451 Phlebitis and thrombophlebitis (superficial/deep; upper/lower extremities) 453 Other venous embolism and thrombosis ICD10: I80 Phlebitis and thrombophlebitis I82 Other venous embolism and thrombosis |
| Valvular diseases |
ICD9: 424 Other diseases of endocardium + 394 Diseases of mitral valve 395 Diseases of aortic valve 396 Diseases of mitral and aortic valves 397 Diseases of other endocardial structures ICD10: I34 Nonrheumatic mitral valve disorders I35 Nonrheumatic aortic valve disorders I36 Nonrheumatic tricuspid valve disorders I37 Nonrheumatic pulmonary valve disorders + I38 Endocarditis, valve unspecified I39 Endocarditis and heart valve disorders in diseases classified elsewhere + I05 Rheumatic mitral valve diseases I06 Rheumatic aortic valve diseases I07 Rheumatic tricuspid valve diseases I08 Multiple valve diseases |
| Cardiac dysrhythmia |
ICD9: 427 (cardiac dysrhythmias) ICD10: I46 Cardiac arrest I47 Paroxysmal tachycardia I48 Atrial fibrillation and flutter I49 Other cardiac arrhythmias |
| Coronary heart disease |
Coronary artery disease: ICD9: 414 Other forms of chronic ischemic heart disease ICD10: I25 Chronic ischemic heart disease Angina: ICD9: 413 Angina pectoris ICD10: I20 Angina pectoris Myocardial infarction: ICD9: 410 Acute myocardial infarction ICD10: I21 Acute myocardial infarction |
| Cerebrovascular Disease |
ICD9: 433 Occlusion and stenosis of precerebral arteries 434 Occlusion of cerebral arteries 435 Transient cerebral ischemia 436 Acute, but ill-defined, cerebrovascular disease 437 Other and ill-defined cerebrovascular disease ICD10: I63 Cerebral infarction I65 Occlusion and stenosis of precerebral arteries, not resulting in cerebral infarction I66 Occlusion and stenosis of cerebral arteries, not resulting in cerebral infarction I67 (other cerebrovascular diseases) G45 (transient cerebral ischemic attacks and related syndromes) G46 (vascular syndromes of brain in cerebrovascular) |
| Chronic kidney Disease |
ICD9: 585 Chronic kidney disease (CKD) ICD10: Chronic kidney disease (CKD) N18 |
Data Analysis
We estimated the annual nationwide number of OAC users and OAC expenditures for each year between 2014 and 2020 after accounting for sampling weight, stratification, clustering, and multi-stage sampling. Standard errors were estimated using Taylor series linearization.25 We evaluated utilization and expenditures trends among all OAC users as a single cohort and across age, sex, race, ethnicity, insurance coverage, and medical condition subgroups. All analyses were conducted using Stata version 16.1 (Stata Corp, College Station, TX).
Results
Demographic Trends Among Oral Anticoagulant Users
The overall number of OAC users in the United States increased from 5.3 million in 2014 (1.7 % of non-institutionalized adults) to 7.6 million (2.3 % of non-institutionalized adults) in 2020. The mean age of OAC users in 2014 was 68.8 years, while the mean age was 70.3 years in 2020. Men comprised 52.5% and 53.8% of the OAC user population in 2014 and 2020, respectively. OAC users in 2020 were more likely to have Medicare insurance coverage in 2020 than in 2014 (78.2% vs. 74.6%). OAC users were predominantly white (88.8% in 2014 and 88.5% in 2020). More than 60% of the users had income that was 2 times or greater than the federal poverty level in each year of the study period (Table 2). We found that in every subcategory of cardiovascular disease concomitant anticoagulant utilization increased from 2014 to 2020 (Figure 2).
Table 2.
Characteristics of all oral anticoagulant users
| Characteristic | 2014 | 2015 | 2016 | 2017 | 2018 | 2019 | 2020 |
|---|---|---|---|---|---|---|---|
| Number of persons | 5,329,269 | 5,946,867 | 5,560,267 | 5,626,619 | 6,925,087 | 7,503,126 | 7,621,187 |
| Age, years1 | 68.8 (0.72) | 69.7 (0.62) | 69.4 (0.81) | 69.9 (0.81) | 71.1 (0.54) | 71.5 (0.55) | 70.3 (0.7) |
| Men | 52.5% | 52.4% | 55.2% | 57.7% | 55.4% | 54.6% | 53.8% |
| Race | |||||||
| Asian | 2.5% | 2% | 2% | 1.8% | 1.8% | 2.1% | 1.2% |
| Black | 8.9% | 9% | 9.4% | 10.8% | 10.3% | 9.7% | 10.4% |
| White | 88.8% | 88.4% | 87.2% | 87.2% | 87.7% | 87.8% | 88.5% |
| Ethnicity | |||||||
| Hispanic | 6.1% | 4.5% | 4.8% | 6.6% | 5.2% | 4.1% | 5.5% |
| Non-Hispanic | 93.9% | 95.5% | 95.2% | 93.4% | 94.8% | 95.9% | 94.5% |
| Family income level | |||||||
| <100% FPL2 | 11% | 9.7% | 11.2% | 8.7% | 11.1% | 10.2% | 10.7% |
| 100–125% FPL2 | 7.6% | 4.7% | 4.9% | 4.7% | 4.1% | 4.4% | 3.8% |
| 125–200%FPL2 | 16.8% | 16.6% | 16.6% | 18.1% | 16.4% | 17.5% | 14% |
| 200–400% FPL2 | 27.1% | 28.9% | 30% | 33.4% | 30.9% | 28.9% | 29.3% |
| ≥400% FPL2 | 37.5% | 40% | 37.2% | 35% | 37.5% | 39% | 42.2% |
| Insurance type 3 | |||||||
| Medicare | 74.6% | 75.2% | 76.4% | 80.2% | 80.6% | 80% | 78.2% |
| Medicaid | 11.4% | 11.4% | 14.7% | 11.5% | 11.7% | 11.1% | 11.2% |
| Private | 61.5% | 59.7% | 57.2% | 57% | 56.8% | 55.2% | 53.4% |
| None | 2.2% | 3.3% | 3.4% | 0.6% | 0.3% | 0.5% | 0.3% |
Mean (SE);
Federal Poverty Level;
Part of the patients have more than one insurance coverage
Figure 2.
Anticoagulant use Across Medical Conditions
Between 2014 and 2020, the number of warfarin users decreased from 3.8 million (70% of all OAC users) to 2.2 million (29% of all OAC users) while the number of DOAC users increased from 1.6 million (30% of all OAC users) to 5.4 million (70% of all OAC users) (Figure 3). Whereas the most frequently prescribed DOAC in 2014 was rivaroxaban (49% of DOAC users), the most frequently prescribed DOAC in 2020 was apixaban (59% of DOAC users). The total number of dabigatran users declined from 520,000 in 2014 to 207,000 in 2020. About two-thirds of DOAC users had private insurance in 2014, while 55.7% had private insurance in 2020. DOAC users with family income level ≥400% federal poverty level were 41.3% of the DOAC population in 2014 and 44.8% in 2020. Of warfarin users, 58.4% and 49% had private insurance in 2014 and 2020, respectively. Warfarin users with family income level ≥400% federal poverty level were 35.9% of the warfarin population in 2014 and 35.4% in 2020. The proportion of warfarin users with Medicare coverage was 83.4%, while 76.2% of the DOAC users had Medicaid coverage in 2020 (Table 3).
Figure 3.
Distribution of warfarin and direct oral anticoagulant use in the United States between 2014 and 2020
Table 3.
Characteristics of warfarin and direct oral anticoagulant users
| 2014 | 2015 | 2016 | 2017 | 2018 | 2019 | 2020 | ||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Estimated population of the U.S. | 318,440,423 | 321,423,251 | 323,141,687 | 324,779,909 | 326,327,888 | 327,396,693 | 328,545,297 | |||||||
| Characteristic | Warfarin | DOAC | Warfarin | DOAC | Warfarin | DOAC | Warfarin | DOAC | Warfarin | DOAC | Warfarin | DOAC | Warfarin | DOAC |
| Number of persons | 3,760,421 | 1,568,848 | 3,485,756 | 2,461,111 | 2,849,872 | 2,710,395 | 2,609,934 | 3,016,685 | 2,660,183 | 4,264,904 | 2,585,528 | 4,917,598 | 2,231,073 | 5,390,114 |
| Age, years1 | 69.2 (0.97) | 67.8 (1.2) | 70.9 (0.78) | 67.8 (1) | 70.8 (1) | 67.9 (1.1) | 71.9 (0.9) | 68.2 (1.2) | 72.3 (0.87) | 70.5 (0.65) | 71.6 (0.89) | 71.3 (0.67) | 71.7 (1) | 69.6 (0.8) |
| Men | 52.9% | 51.4% | 53.7% | 51.1% | 56.9% | 53.7% | 61% | 54.7% | 55.7% | 55.2% | 53.7% | 54.8% | 57.4% | 53% |
| Race | ||||||||||||||
| Asian | 3.2% | 1% | 1.9% | 2% | 2.3% | 1.7% | 1.5% | 2.1% | 1% | 2.3% | 1.3% | 2.5% | 1.1% | 1.3% |
| Black | 7.4% | 12.5% | 9.5% | 8.3% | 9.8% | 9.1% | 9.2% | 12.2% | 10.3% | 10.2% | 9.9% | 9.8% | 9.4% | 11.7% |
| White | 89.5% | 87% | 88.6% | 88.2% | 87.6% | 86.9% | 89.2% | 85.5% | 89.4% | 86.8% | 89.4% | 86.8% | 90.1% | 86.9% |
| Ethnicity | ||||||||||||||
| Hispanic | 5.3% | 8.1% | 4.7% | 4.4% | 5.6% | 4% | 6.7% | 6.5% | 5.7% | 4.8% | 2.9% | 4.6% | 3.8% | 6.2% |
| Non-Hispanic | 94.7% | 91.9% | 95.3% | 95.6% | 94.4% | 96% | 93.3% | 93.5% | 94.3% | 95.2% | 97.1% | 95.4% | 96.2% | 93.8% |
| Family income level | ||||||||||||||
| <100% FPL2 | 11.3% | 10.4% | 11.1% | 7.7% | 10.7% | 11.7% | 7.9% | 9.5% | 9.8% | 11.9% | 12.3% | 8.9% | 15.4% | 8.7% |
| 100–125% FPL2 | 8.8% | 4.7% | 4.5% | 5.1% | 3.9% | 6.1% | 3.3% | 5.9% | 4.1% | 4% | 4.5% | 4.2% | 1.9% | 4.6% |
| 125–200% FPL2 | 19% | 11.4% | 17.6% | 15% | 19.5% | 13.8% | 17.7% | 18.5% | 18.6% | 15.6% | 18.8% | 16.9% | 13% | 14.2% |
| 200–400% FPL2 | 25% | 32.1% | 27.3% | 31.1% | 29.2% | 30.7% | 37.3% | 30% | 33.2% | 29.4% | 29.9% | 28.5% | 34.3% | 27.7% |
| ≥400% FPL2 | 35.9% | 41.3% | 39.5% | 41.1% | 36.7% | 37.6% | 33.7% | 36.1% | 34.2% | 39.1% | 34.5% | 41.4% | 35.4% | 44.8% |
| Insurance type 3 | ||||||||||||||
| Medicare | 76.9% | 69.1% | 78% | 70.4% | 79.4% | 73.4% | 86.2% | 75% | 85.5% | 77.8% | 80.8% | 78.9% | 83.4% | 76.2% |
| Medicaid | 12.3% | 9.3% | 11.6% | 11.2% | 13% | 16.4% | 8.5% | 14.1% | 12.3% | 11.2% | 9.3% | 12% | 9.9% | 11.5% |
| Private | 58.4% | 68.8% | 56.8% | 64.2% | 55% | 59.7% | 54.1% | 59.5% | 49% | 61.5% | 52.1% | 56.9% | 49% | 55.7% |
| None | 1.6% | 3.7% | 2.9% | 3.7% | 1.9% | 4.9% | 0.7% | 0.5% | 0.4% | 0.2% | 0.3% | 0.6% | 0.3% | 0.3% |
| DOAC | ||||||||||||||
| Apixaban | -- | 282,698 | -- | 821,406 | -- | 1,024,603 | -- | 1,601,534 | -- | 2,462,452 | -- | 2,796,672 | -- | 3,189,415 |
| Dabigatran | -- | 523,819 | -- | 403,733 | -- | 294,380 | -- | 267,357 | -- | 291,583 | -- | 232,887 | -- | 207,564 |
| Rivaroxaban | -- | 762,331 | -- | 1,229,804 | -- | 1,432,215 | -- | 1,147,795 | -- | 1,548,433 | -- | 1,888,040 | -- | 2,028,055 |
Mean (SE);
Federal Poverty Level;
Part of the patients have more than one insurance coverage
Expenditures
The total expenditure of OACs in the United States increased from $3.4 billion ($629/OAC user) in 2014 to $17.8 billion ($2,336/OAC user) in 2020 (272% increase), mostly driven by the increase in DOAC expenditures ($2.8 billion ($1,785/DOAC user) in 2014 to $17.6 billion ($3,265/DOAC user) in 2020). Third-party payers constituted 67% ($370 million) of the total warfarin cost in 2014 and 66% ($130 million) in 2020. On the other hand, DOACs third payers paid between 86% ($1,540/DOAC user) and 90% ($2,968/DOAC user) throughout the period between 2014 and 2020 (Figure 4). Out-of-pocket costs also increased during this period, from $380 million ($242/person) to $1.6 billion ($297/person). DOACs and warfarin cost details per person is shown in table 4.
Figure 4.
Oral anticoagulant out-of-pocket and third-party payer costs
Table 4.
Oral anticoagulants number of users and expenditure.
| 2014 | 2015 | 2016 | 2017 | 2018 | 2019 | 2020 | |
|---|---|---|---|---|---|---|---|
| Warfarin users | 3,760,421 (1.2%) | 3,485,756 (1.1%) | 2,849,872 (0.9%) | 2,609,934 (0.8%) | 2,660,183 (0.8%) | 2,585,528 (0.8%) | 2,231,073 (0.7%) |
| Warfarin third-party cost | 371,000,000 | 355,000,000 | 377,000,000 | 313,000,000 | 310,000,000 | 165,000,000 | 131,000,000 |
| Warfarin third-party cost $/person/year |
98.66 | 101.84 | 132.3 | 119.93 | 116.53 | 63.82 | 58.72 |
| Warfarin out-of-pocket cost | 182,000,000 | 194,000,000 | 174,000,000 | 102,000,000 | 105,000,000 | 106,000,000 | 67,100,000 |
| Warfarin out-of-pocket cost $/person/year |
48.4 | 55.66 | 61.1 | 39.1 | 39.5 | 41.0 | 30.1 |
| Warfarin total cost | 553,000,000 | 549,000,000 | 552,000,000 | 415,000,000 | 416,000,000 | 272,000,000 | 198,000,000 |
| Warfarin: $/person/year | 147.06 | 157.50 | 193.69 | 159.01 | 156.38 | 105.2 | 88.75 |
| Total DOACs users | 1,568,848 (0.5%) | 2,461,111 (0.8%) | 2,710,395 (0.8%) | 3,016,685 (0.9%) | 4,264,904 (1.3%) | 4,917,598 (1.5%) | 5,390,114 (1.6%) |
| DOACs third-party cost | 2,420,000,000 | 4,400,000,000 | 6,590,000,000 | 7,280,000,000 | 10,200,000,000 | 15,200,000,000 | 16,000,000,000 |
| DOACs third-party cost $/person/year |
1,542.53 | 1,787.81 | 2,431.4 | 2,413.25 | 2,391.61 | 3,090.94 | 2,968.4 |
| DOACs out-of-pocket cost | 380,000,000 | 584,000,000 | 772,000,000 | 1,360,000,000 | 1,240,000,000 | 1,610,000,000 | 1,630,000,000 |
| DOACs out-of-pocket cost $/person/year |
242.22 | 237.3 | 284.83 | 450.83 | 290.75 | 327.4 | 302.41 |
| DOACs total cost | 2,800,000,000 | 4,990,000,000 | 7,360,000,000 | 8,640,000,000 | 11,500,000,000 | 16,800,000,000 | 17,600,000,000 |
| DOACs: $/person/year | 1,784.75 | 2,027.54 | 2,715.47 | 2,864.07 | 2,696.43 | 3,416.3 | 3,265.24 |
| Total OACs users | 5,329,268 (1.7%) | 5,902,893 (1.8%) | 5,547,818 (1.7%) | 5,626,619 (1.7%) | 6,882,955 (2.1%) | 7,503,126 (2.3%) | 7,621,187 (2.3%) |
| OACs third-party cost | 2,790,000,000 | 4,760,000,000 | 6,970,000,000 | 7,600,000,000 | 10,600,000,000 | 15,400,000,000 | 16,100,000,000 |
| OACs third-party cost $/person/year |
523.52 | 806.4 | 1,256.35 | 1,350.72 | 1,540.04 | 2,052.48 | 2,112.53 |
| OACs out-of-pocket cost | 561,000,000 | 778,000,000 | 946,000,000 | 1,460,000,000 | 1,340,000,000 | 1,720,000,000 | 1,700,000,000 |
| OACs third-party cost $/person/year |
105.27 | 131.8 | 170.52 | 259.48 | 194.68 | 229.24 | 223.06 |
| Total expenditure | 3,350,000,000 | 5,540,000,000 | 7,910,000,000 | 9,050,000,000 | 11,900,000,000 | 17,100,000,000 | 17,800,000,000 |
| Total: $/person/year | 628.60 | 938.52 | 1,425.79 | 1,608.43 | 1,728.91 | 2,279.05 | 2,335.59 |
| Number of prescriptions | 31,500,000 | 31,900,000 | 32,700,000 | 31,700,000 | 34,800,000 | 38,100,000 | 34,200,000 |
Discussion
Since their initial approvals for stroke prevention in atrial fibrillation and venous thromboembolism treatment, the landscape of DOAC indications has evolved rapidly. Landmark clinical trials have not only established new indications for DOAC therapy but also clarified areas where other antithrombotics offer better safety and/or efficacy profiles. Approval of reversal agents has also altered DOAC usage. Altogether, these clinical trials have brought considerable clarity to the use of DOAC therapy. The number of people using OACs increased significantly between 2014 and 2020 in the United States. This increase was mainly driven by the rise in DOACs use, despite the decline in warfarin use during the same period. Among individual DOACs, apixaban use increased the most, followed by rivaroxaban. Conversely, dabigatran use declined during that period. Moreover, the total cost of OACs had a substantial increase form 3.4 billion to 17.8 billion, despite a decline in warfarin total cost from 0.6 billion to 0.2 billion, driven by the increase of DOACs total cost from 2.8 billion in 2014 to 17.6 billion in 2020. Apixaban led the increase in total costs, followed by rivaroxaban, while dabigatran costs decreased overall.
The DOACs rivaroxaban, apixaban and dabigatran offer greater convenience and safety than warfarin. The impact of this medication class on clinical practice and the healthcare system, however, remains unclear. This analysis of a nationally representative medication use and expenditure survey not only demonstrates that DOACs have surpassed warfarin as the preferred OAC therapy in the United States but also quantifies the financial burden imposed on the healthcare system.
Several state-level and national studies focusing on specific indications showed an increase in the prescription of DOACs over the years studied.26–30 Nonetheless, despite that increase in use, DOACs might be underutilized as shown in several regional studies in patients with AF compared with warfarin.31–35 In this study, we found that in every subcategory of cardiovascular disease concomitant anticoagulant utilization increased from 2014 to 2020, despite medical conditions and corresponding ICD coding did not identically match with FDA approved anticoagulant indications. Our analysis builds upon the aforementioned studies by improving generalizability with a nationally representative study sample.
The initial uptake of DOACs therapy was hindered by their higher costs, lack of familiarity with pharmacological and clinical trial differences across agents, concerns about the inability to quantify the antithrombotic effects and the lack of specific reversal agents. These barriers have been overcome through the development of idarucizumab and andexanet reversal agents for factor Xa inhibitors and dabigatran, respectively, and greater comfort with the safety profile of these agents. More information has emerged in regard to in special population such as patients with obesity and renal insufficiency.1,36,37 Furthermore, our data indicate that third-party public and private payers account for a substantial proportion of the total cost of therapy. Thus, the improved safety and greater convenience of these agents have motivated clinicians and patients to increase use of DOACs.
There are limitations to our study. Our data is subject to the inherent limitations of self-reported code generated database including reporting bias and errors in coding. The availability and accuracy of respondents’ medical histories is limited, which could result in possible underrepresentation or overrepresentation of certain medical conditions. MEPS only covers the non-institutionalized, civilian population in the United States and OAC patterns may differ in other populations. Our analysis accounted for medications cost but not other healthcare costs like preventing cardiovascular events. Finally, we did not account for inflation rate in this study.
Conclusions
Rivaroxaban, apixaban and dabigatran have overtaken warfarin as the preferred oral anticoagulant in the United States. Increased use of these agents has led to greater total expenditures on oral anticoagulant therapy, which may abate when generic options become available.
References
- 1.Roberti R, Iannone LF, Palleria C, et al. Direct Oral Anticoagulants: From Randomized Clinical Trials to Real-World Clinical Practice. Front Pharmacol. 2021;12(May):1–14. doi: 10.3389/fphar.2021.684638 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 2.Eikelboom JW, Connolly SJ, Bosch J, et al. Rivaroxaban with or without Aspirin in Stable Cardiovascular Disease. N Engl J Med. 2017;377(14):1319–1330. doi: 10.1056/NEJMoa1709118 [DOI] [PubMed] [Google Scholar]
- 3.Bonaca MP, Bauersachs RM, Anand SS, et al. Rivaroxaban in Peripheral Artery Disease after Revascularization. N Engl J Med. 2020;382(21):1994–2004. doi: 10.1056/NEJMoa2000052 [DOI] [PubMed] [Google Scholar]
- 4.Gibson CM, Mehran R, Bode C, et al. Prevention of Bleeding in Patients with Atrial Fibrillation Undergoing PCI. N Engl J Med. 2016;375(25):2423–2434. doi: 10.1056/NEJMoa1611594 [DOI] [PubMed] [Google Scholar]
- 5.Cannon CP, Bhatt DL, Oldgren J, et al. Dual Antithrombotic Therapy with Dabigatran after PCI in Atrial Fibrillation. N Engl J Med. 2017;377(16):1513–1524. doi: 10.1056/NEJMoa1708454 [DOI] [PubMed] [Google Scholar]
- 6.Lopes RD, Heizer G, Aronson R, et al. Antithrombotic Therapy after Acute Coronary Syndrome or PCI in Atrial Fibrillation. N Engl J Med. 2019;380(16):1509–1524. doi: 10.1056/NEJMoa1817083 [DOI] [PubMed] [Google Scholar]
- 7.Vranckx P, Valgimigli M, Eckardt L, et al. Edoxaban-based versus vitamin K antagonist-based antithrombotic regimen after successful coronary stenting in patients with atrial fibrillation (ENTRUST-AF PCI): a randomised, open-label, phase 3b trial. Lancet. 2019;394(10206):1335–1343. doi: 10.1016/S0140-6736(19)31872-0 [DOI] [PubMed] [Google Scholar]
- 8.Calkins H, Willems S, Gerstenfeld EP, et al. Uninterrupted Dabigatran versus Warfarin for Ablation in Atrial Fibrillation. N Engl J Med. 2017;376(17):1627–1636. doi: 10.1056/NEJMoa1701005 [DOI] [PubMed] [Google Scholar]
- 9.Carrier M, Abou-Nassar K, Mallick R, et al. Apixaban to Prevent Venous Thromboembolism in Patients with Cancer. N Engl J Med. 2019;380(8):711–719. doi: 10.1056/NEJMoa1814468 [DOI] [PubMed] [Google Scholar]
- 10.Agnelli G, Becattini C, Meyer G, et al. Apixaban for the Treatment of Venous Thromboembolism Associated with Cancer. N Engl J Med. 2020;382(17):1599–1607. doi: 10.1056/NEJMoa1915103 [DOI] [PubMed] [Google Scholar]
- 11.Van Mieghem NM, Unverdorben M, Hengstenberg C, et al. Edoxaban versus Vitamin K Antagonist for Atrial Fibrillation after TAVR. N Engl J Med. 2021;385(23):2150–2160. doi: 10.1056/NEJMoa2111016 [DOI] [PubMed] [Google Scholar]
- 12.Connolly SJ, Karthikeyan G, Ntsekhe M, et al. Rivaroxaban in Rheumatic Heart Disease–Associated Atrial Fibrillation. N Engl J Med. 2022;387(11):978–988. doi: 10.1056/NEJMoa2209051 [DOI] [PubMed] [Google Scholar]
- 13.Dangas GD, Tijssen JGP, Wöhrle J, et al. A Controlled Trial of Rivaroxaban after Transcatheter Aortic-Valve Replacement. N Engl J Med. 2020;382(2):120–129. doi: 10.1056/NEJMoa1911425 [DOI] [PubMed] [Google Scholar]
- 14.Diener HC, Sacco RL, Easton JD, et al. Dabigatran for Prevention of Stroke after Embolic Stroke of Undetermined Source. N Engl J Med. 2019;380(20):1906–1917. doi: 10.1056/NEJMoa1813959 [DOI] [PubMed] [Google Scholar]
- 15.Hart RG, Sharma M, Mundl H, et al. Rivaroxaban for Stroke Prevention after Embolic Stroke of Undetermined Source. N Engl J Med. 2018;378(23):2191–2201. doi: 10.1056/NEJMoa1802686 [DOI] [PubMed] [Google Scholar]
- 16.Zannad F, Anker SD, Byra WM, et al. Rivaroxaban in Patients with Heart Failure, Sinus Rhythm, and Coronary Disease. N Engl J Med. 2018;379(14):1332–1342. doi: 10.1056/NEJMoa1808848 [DOI] [PubMed] [Google Scholar]
- 17.Devereaux PJ, Duceppe E, Guyatt G, et al. Dabigatran in patients with myocardial injury after non-cardiac surgery (MANAGE): an international, randomised, placebo-controlled trial. Lancet. 2018;391(10137):2325–2334. doi: 10.1016/S0140-6736(18)30832-8 [DOI] [PubMed] [Google Scholar]
- 18.Pollack C V, Reilly PA, Eikelboom J, et al. Idarucizumab for Dabigatran Reversal. N Engl J Med. 2015;373(6):511–520. doi: 10.1056/NEJMoa1502000 [DOI] [PubMed] [Google Scholar]
- 19.Pollack C V, Reilly PA, van Ryn J, et al. Idarucizumab for Dabigatran Reversal — Full Cohort Analysis. N Engl J Med. 2017;377(5):431–441. doi: 10.1056/NEJMoa1707278 [DOI] [PubMed] [Google Scholar]
- 20.Connolly SJ, Milling TJ, Eikelboom JW, et al. Andexanet Alfa for Acute Major Bleeding Associated with Factor Xa Inhibitors. N Engl J Med. 2016;375(12):1131–1141. doi: 10.1056/NEJMoa1607887 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 21.Siegal DM, Curnutte JT, Connolly SJ, et al. Andexanet Alfa for the Reversal of Factor Xa Inhibitor Activity. N Engl J Med. 2015;373(25):2413–2424. doi: 10.1056/NEJMoa1510991 [DOI] [PubMed] [Google Scholar]
- 22.NHIS - National Health Interview Survey. Centers for Disease Control and Prevention. Accessed December 8, 2022. https://www.cdc.gov/nchs/nhis/index.htm
- 23.Chowdhury SR, Machlin SR GK. Sample Designs of the Medical Expenditure Panel Survey Household Component, 1996–2006 and 2007–2016. Methodology Report #33. Agency for Healthcare Research and Quality, Rockville, MD. [Google Scholar]
- 24.Hill SC, Roemer M SM. Outpatient Prescription Drugs: Data Collection and Editing in the 2011 Medical Expenditure Panel Survey. Methodology Report #29. Agency Healthc Res Qual; Rockville, MD. [Google Scholar]
- 25.Binder DA. On the Variances of Asymptotically Normal Estimators from Complex Surveys. Int Stat Rev / Rev Int Stat. 1983;51(3):279. doi: 10.2307/1402588 [DOI] [Google Scholar]
- 26.Wong SL, Marshall LZ, Lawson KA. Direct oral anticoagulant prescription trends, switching patterns, and adherence in Texas medicaid. Am J Manag Care. 2018;24(8):SP309–SP314. [PubMed] [Google Scholar]
- 27.Sur NB, Wang K, Di Tullio MR, et al. Disparities and Temporal Trends in the Use of Anticoagulation in Patients With Ischemic Stroke and Atrial Fibrillation. Stroke. 2019;50(6):1452–1459. doi: 10.1161/STROKEAHA.118.023959 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 28.Lutsey PL, Walker RF, MacLehose RF, Alonso A, Adam TJ, Zakai NA. Direct oral anticoagulants and warfarin for venous thromboembolism treatment: Trends from 2012 to 2017. Res Pract Thromb Haemost. 2019;3(4):668–673. doi: 10.1002/rth2.12222 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 29.Alalwan AA, Voils SA, Hartzema AG. Trends in utilization of warfarin and direct oral anticoagulants in older adult patients with atrial fibrillation. Am J Heal Pharm. 2017;74(16):1237–1244. doi: 10.2146/ajhp160756 [DOI] [PubMed] [Google Scholar]
- 30.Rodwin BA, Salami JA, Spatz ES, et al. Variation in the Use of Warfarin and Direct Oral Anticoagulants in Atrial Fibrillation and Associated Cost Implications. Am J Med. 2019;132(1):61–70.e1. doi: 10.1016/j.amjmed.2018.09.026 [DOI] [PubMed] [Google Scholar]
- 31.Wehbe RM, Yadlapati A. Underuse of Oral Anticoagulants for Nonvalvular Atrial Fibrillation: Past, Present, and Future. Texas Hear Inst J. 2016;43(4):287–290. doi: 10.14503/THIJ-16-5785 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 32.Alamneh EA, Chalmers L, Bereznicki LR. Suboptimal use of oral anticoagulants in atrial fibrillation: Has the introduction of direct oral anticoagulants improved prescribing practices? Am J Cardiovasc Drugs. 2016;16(3):183–200. doi: 10.1007/s40256-016-0161-8 [DOI] [PubMed] [Google Scholar]
- 33.Hsu JC, Maddox TM, Kennedy KF, et al. Oral Anticoagulant Therapy Prescription in Patients With Atrial Fibrillation Across the Spectrum of Stroke Risk: Insights From the NCDR PINNACLE Registry. JAMA Cardiol. 2016;1(1):55–62. doi: 10.1001/jamacardio.2015.0374 [DOI] [PubMed] [Google Scholar]
- 34.Hsu JC, Maddox TM, Kennedy K, et al. Aspirin Instead of Oral Anticoagulant Prescription in Atrial Fibrillation Patients at Risk for Stroke. J Am Coll Cardiol. 2016;67(25):2913–2923. doi: 10.1016/j.jacc.2016.03.581 [DOI] [PubMed] [Google Scholar]
- 35.Kattoor AJ, Pothineni NV, Goel A, et al. Prescription Patterns and Outcomes of Patients With Atrial Fibrillation Treated With Direct Oral Anticoagulants and Warfarin: A Real-World Analysis. J Cardiovasc Pharmacol Ther. 2019;24(5):428–434. doi: 10.1177/1074248419841634 [DOI] [PubMed] [Google Scholar]
- 36.White EM, Coons JC. Direct Oral Anticoagulant Use in Special Populations: Elderly, Obesity, and Renal Failure. Curr Cardiol Rep. 2021;23(4):27. doi: 10.1007/s11886-021-01456-9 [DOI] [PubMed] [Google Scholar]
- 37.Elshafei MN, Mohamed MFH, El-Bardissy A, et al. Comparative effectiveness and safety of direct oral anticoagulants compared to warfarin in morbidly obese patients with acute venous thromboembolism: systematic review and a meta-analysis. J Thromb Thrombolysis. 2021;51(2):388–396. doi: 10.1007/s11239-020-02179-4 [DOI] [PMC free article] [PubMed] [Google Scholar]