Sage Journals: Discover world-class research

Abstract

Background:

Major atrial fibrillation (AF) guidelines recommend non-vitamin K antagonist oral anticoagulants (NOACs) over warfarin, except in rare clinical circumstances based on 4 randomized controlled trials comparing each NOAC with warfarin. We aimed to investigate the current NOAC prescription behaviors in alignment with the recent clinical evidence available.

Method:

We conducted a cross-sectional analysis of NOAC-using patients with non-valvular atrial fibrillation (NVAF) who were aged ≥65 years on the index date (July 1, 2018) based on nationwide claims data. The types of NOACs being taken were analyzed using chi-squared tests, and factors influencing NOAC selection were identified using multinomial logistic regression analysis.

Results:

A total of 6,061 patients were included. Among the 4 NOACs, rivaroxaban was the most used NOAC. Patients aged ≥75 years (odds ratio [OR] = 1.270, confidence interval [CI] = 1.089-1.450) and women (OR = 1.148, CI = 1.011-1.284) were more likely to use apixaban relative to rivaroxaban. Patients with prior stroke/transient ischemic attack/thromboembolism had higher odds of using dabigatran (OR = 1.508, CI = 1.312-1.704) and apixaban (OR = 1.186, CI = 1.026-1.346). Patients with renal disease had higher odds of using apixaban (OR = 1.466, 95% CI = 1.238-1.693). These findings are consistent with the efficacy and safety profiles reported in pivotal trials and observational studies comparing individual NOACs.

Conclusion:

Among the 4 NOACs, rivaroxaban was the most commonly used NOAC. Apixaban was preferred for patients aged ≥75 years, females, and patients with renal disease.

Keywords

atrial fibrillation stroke anticoagulants non-vitamin K antagonist oral anticoagulants selection factors

Introduction

The era of non-vitamin K antagonist oral anticoagulants (NOACs) began with the advent of dabigatran, the first oral direct thrombin inhibitor.¹ Initially, the European Medicines Agency granted marketing authorization for dabigatran for the prevention of venous thromboembolism in patients who underwent total hip- or knee-replacement surgery in March 2008.^2,3 Then, its indication was extended to the preventive therapy of stroke in patients with non-valvular atrial fibrillation (NVAF) based on the results of the Randomized Evaluation of Long-Term Anticoagulant Therapy (RE-LY) trial.^3,4 In October 2010, the Food and Drug Administration and Health Canada approved dabigatran for stroke prevention in patients with NVAF.⁵

Since then, factor Xa inhibitors (rivaroxaban, apixaban, and edoxaban) have been subsequently approved, and currently, 4 NOACs are being used.^6
-8 They have become a preferred treatment option, leading to a new chapter in the world of oral anticoagulants, where previously only vitamin K antagonists (VKAs) existed.

The risk of stroke in patients with atrial fibrillation (AF) is increased by up to 5 times,^9
-11 which is effectively prevented by anticoagulant therapy,¹² and VKAs have been responsible for this prevention for a long time.¹³ However, VKAs have several limitations such as delayed onset and offset of action, narrow therapeutic range demanding regular monitoring of the international normalized ratio, and many interactions with food and drugs.¹⁴ These limitations have been the cause of poor utilization despite the importance of anticoagulant therapy.¹⁵ With the introduction of NOACs, these shortcomings have been circumvented, and the anticoagulant utilization rate has been gradually increasing.^15,16

Currently, NOACs are favored over VKAs in patients with AF because they have non-inferior or superior efficacy and a lower risk of serious bleeding.^17

-21 Although there are differences in NOAC prescription patterns with respect to each country,^16,22
-24 it is now a global trend that significantly more NOACs are prescribed than VKAs for stroke prevention in patients with AF.^16,22

-25

Major AF guidelines recommend NOACs over warfarin, except in rare clinical circumstances (i.e., valvular AF) based on 4 randomized controlled trials comparing each NOAC with warfarin.^4,26

-30 Choosing one NOAC over another is a somewhat challenging next step for several reasons. First, direct comparisons of efficacy and safety between NOACs are not feasible because head-to-head trials are not available. In addition, there are significant differences in the eligibility criteria and other study protocols between the NOAC trials.^4,26
-28 Therefore, the relative efficacy and safety of NOACs can only be assessed through either indirect comparisons^31,32 or observational studies.^33,34 In the absence of direct comparison of NOACs, the individual evidence from each NOAC trial should be assessed and applied to a patient under specific circumstances. In addition, pharmacokinetic properties, dosing schedule, potential drug-drug interactions, cost, and local availability of each NOAC should also be considered before the final decision is made. To date, most studies regarding oral anticoagulant choice in patients with AF have focused on the selection of NOACs vs. warfarin, rather than NOAC vs. NOAC.^25,35
-37 Therefore, we aimed to investigate clinicians’ prescription behaviors and the factors influencing the choice between NOACs in alignment with the current clinical evidence. We wanted to see how patient-related clinical factors, such as age, sex, and comorbidities, have been incorporated in the process of NOAC selection in a real-world setting.

Methods

Study Data

The Aged Population Sample (APS) data of approximately 700,000 individuals (10% extraction rate) aged ≥65 years in Korea collected by the Health Insurance Review and Assessment service (HIRA) in 2018 (HIRA-APS-2018-0039) were used.³⁸

The Korean Classification of Diseases-7 codes (KCD-7 codes; the Korean version of International Classification of Disease, Tenth Revision [ICD-10]) were used to verify patients with particular diseases or conditions. Ingredient codes were used to verify 4 NOACs (dabigatran, rivaroxaban, apixaban, and edoxaban) and warfarin (Online Appendix A).

The social health insurance system of Korea is composed of 3 different insurance types: National Health Insurance (NHI), Medical Aid (MedAid), and Patriots & Veterans Insurance (PVI). All of them are combined into a single payer system, and all citizens are required to subscribe to the system.³⁹ Any type of social insurance is subject to the same benefit criteria and differs only in the patients’ pay rate for copayments, ranging from 0% to 30%, depending on the type of insurance.³⁹ The NHI covers approximately 97% of the population and charges a duty on income, while the MedAid is a guaranteed system for the low-income group and the PVI is for national veterans.³⁹

The Institutional Review Board (IRB) of Pusan National University approved this study (PNU IRB/2020_93_HR). Written consent was waived by the IRB because the data used in this study were secondary data provided by the HIRA and did not contain any personal identification information.

Study Subjects

Patients who were diagnosed with AF or atrial flutter as a primary diagnosis from January 1, 2018, to June 30, 2018 were identified. Among those patients, patients with valvular heart disease were excluded. We classified patients with valve replacement or valvular disease based on KCD-7 codes as “patients with valvular heart disease.” Patients with NVAF whose CHA₂DS₂-VASc (congestive heart failure, hypertension, age ≥75 years [2 points], diabetes mellitus, prior stroke or transient ischemic attack [TIA] [2 points], vascular disease, age 65-74 years, and sex) score ≥2 were considered appropriate candidates for anticoagulant therapy in this study. Patients taking NOACs on the index date (July 1, 2018) were selected as study subjects. NOACs taken on the index date were estimated by considering the prescription date and the number of days of administration since the last prescription of each patient before the index date.

The following factors influencing NOAC selection were investigated: age, sex, insurance type, presence of comorbid diseases, ATRIA bleeding risk score, AF type, antiplatelet use, and medical institution type and region. Insurance types were divided into NHI and MedAid/PVI. Comorbid diseases included congestive heart failure, hypertension, diabetes mellitus, prior stroke/TIA/thromboembolism, vascular disease, renal disease, and cancer. The medical institution types were divided into tertiary hospitals, general hospitals, and primary medical institutions/others. The region was divided into the capital city, 6 metropolitan cities, and other regions.

Statistical Analysis

Demographic and clinical characteristics of the study subjects are presented as numbers and percentages using frequency analysis since all of them are categorical variables. The P-values for the association of each patient characteristic with NOAC selection were determined using chi-square tests. We performed multinomial logistic regression analysis to identify factors influencing NOAC selection and reported the odds ratios (ORs) and 95% confidence intervals (CIs). The likelihood-ratio test (presented as chi-square) and pseudo-R² (by McFadden method) were used to measure the goodness-of-fit for the models. All data manipulation and statistical analyses were conducted using R Statistical Software (version 4.0.3; R Foundation for Statistical Computing, Vienna, Austria), and the significance level was set at P < .05.

Results

Subject Characteristics

A total of 6,061 patients with NVAF using NOACs on the index date were included in the analysis (Figure 1). Table 1 summarizes the demographic and clinical characteristics of the study subjects. Patients aged ≥75 years comprised 56.6% of the study population, and the proportion of male patients was slightly higher than that of female patients (50.8% vs. 49.2%). Hypertension was the most common comorbidity (89.5%), followed by congestive heart failure (48.7%).

Figure 1.

Selection of study subjects. AF indicates atrial fibrillation; AFL, atrial flutter; CHA2DS2-VASc, congestive heart failure, hypertension, age ≥75 years [2 points], diabetes mellitus, prior stroke or transient ischemic attack [TIA] [2 points], vascular disease, age 65-74 years, and sex; HIRA-APS, Health Insurance Review & Assessment Service-Aged Population Sample; NOAC, non-vitamin K antagonist oral anticoagulant; NVAF, non-valvular atrial fibrillation; OAC, oral anticoagulant.

Table 1.

Demographic Characteristics and Non-Vitamin K Antagonist Oral Anticoagulants Utilization.

	N	(%)	Percentage of patients using NOAC on the index date
	N	(%)	Dabigatran	(%)	Rivaroxaban	(%)	Apixaban	(%)	Edoxaban	(%)	P-value
Overall	6,061		757	(12.5)	1,992	(32.9)	1,624	(26.8)	1,688	(27.9)
Age group											<0.001
65-69	1,236	(20.4)	165	(13.4)	402	(32.5)	283	(22.9)	386	(31.2)
70-74	1,395	(23.0)	194	(13.9)	487	(34.9)	337	(24.2)	377	(27.0)
≥75	3,430	(56.6)	398	(11.6)	1,103	(32.2)	1,004	(29.3)	925	(27.0)
Gender											0.001
Male	3,079	(50.8)	429	(13.9)	1,018	(33.1)	779	(25.3)	853	(27.7)
Female	2,982	(49.2)	328	(11.0)	974	(32.7)	845	(28.3)	835	(28.0)
Insurance type											0.560
NHI	5,607	(92.5)	700	(12.5)	1,856	(33.1)	1,498	(26.7)	1,553	(27,7)
MedAid/PVI	454	(7.5)	57	(12.6)	136	(30.0)	126	(27.8)	135	(29.7)
CHF											0.047
No	3,110	(51.3)	416	(13.4)	1,004	(32.3)	803	(25.8)	887	(28.5)
Yes	2,951	(48.7)	341	(11.6)	988	(33.5)	821	(27.8)	801	(27.1)
Hypertension											0.492
No	639	(10.5)	89	(13.9)	201	(31.5)	179	(28.0)	170	(26.6)
Yes	5,422	(89.5)	668	(12.3)	1,791	(33.0)	1,445	(26.7)	1,518	(28.0)
Diabetes mellitus											0.594
No	3,680	(60.7)	456	(12.4)	1,206	(32.8)	971	(26.4)	1,047	(28.5)
Yes	2,381	(39.3)	301	(12.6)	786	(33.0)	653	(27.4)	641	(26.9)
Prior stroke/TIA/TE											<0.001
No	4,692	(77.4)	541	(11.5)	1,553	(33.1)	1,220	(26.0)	1,378	(29.4)
Yes	1,369	(22.6)	216	(15.8)	439	(32.1)	404	(29.5)	310	(22.6)
Vascular disease											0.611
No	5,702	(94.1)	713	(12.5)	1,884	(33.0)	1,519	(26.6)	1,586	(27.8)
Yes	359	(5.9)	44	(12.3)	108	(30.1)	105	(29.2)	102	(28.4)
Renal disease											<0.001
No	5,528	(91.2)	712	(12.9)	1,831	(33.1)	1,420	(25.7)	1,565	(28.3)
Yes	533	(8.8)	45	(8.4)	161	(30.2)	204	(38.3)	123	(23.1)
Cancer											0.834
No	5,421	(89.4)	679	(12.5)	1,788	(33.0)	1,443	(26.6)	1,511	(27.9)
Yes	640	(10.6)	78	(12.2)	204	(31.9)	181	(28.3)	177	(27.2)
Type of medical institution											<0.001
Tertiary	2,755	(45.5)	366	(13.3)	791	(28.7)	880	(31.9)	718	(26.1)
General	2,791	(46.0)	342	(12.3)	964	(34.5)	640	(22.9)	845	(30.3)
Primary & others	515	(8.5)	49	(9.5)	237	(46.0)	104	(20.2)	125	(24.3)
Region											<0.001
Capital	1,813	(29.9)	259	(14.3)	483	(26.6)	551	(30.4)	520	(28.7)
Metropolitan	1,640	(27.1)	222	(13.5)	577	(35.2)	486	(29.6)	355	(21.6)
Others	2,608	(43.0)	276	(10.6)	932	(35.7)	587	(22.5)	813	(31.2)
AF type											0.004
Paroxysmal	2,456	(40.5)	346	(14.1)	796	(32.4)	656	(26.7)	658	(26.8)
Persistent	928	(15.3)	99	(10.7)	327	(35.2)	251	(27.0)	251	(27.0)
Permanent	674	(11.1)	75	(11.1)	241	(35.8)	151	(22.4)	207	(30.7)
AFL/Others	2,003	(33.0)	237	(11.8)	628	(31.4)	566	(28.3)	572	(28.6)
Antiplatelet Use											0.690
No	5,455	(90.0)	686	(12.6)	1,788	(32.8)	1,453	(26.6)	1,528	(28.0)
Yes	606	(10.0)	71	(11.7)	204	(33.7)	171	(28.2)	160	(26.4)
ATRIA score											0.045
≤ 4	5,700	(94.0)	722	(12.7)	1,881	(33.0)	1,506	(26.4)	1,591	(27.9)
> 4	361	(6.0)	35	(9.7)	111	(30.7)	118	(32.7)	97	(26.9)

Abbreviations: AF, atrial fibrillation; AFL, atrial flutter; ATRIA, anticoagulation and risk factors in AF; CHF, congestive heart failure; MedAid, Medical Aid; NHI, National Health Insurance; NOAC, non-vitamin K antagonist oral anticoagulants; PVI, Patriots & Veterans Insurance; TE, thromboembolism; TIA, transient ischemic attack.

NOAC Selection

Rivaroxaban was the most commonly used NOAC (32.9%), followed by edoxaban (27.9%), apixaban (26.8%), and dabigatran (12.5%) on the index date. The distribution of the use of the 3 NOACs (as dabigatran was infrequently used) is as follows: in the ≥75-year age group, rivaroxaban, apixaban, and edoxaban were used in that order. The same order was observed regarding NOAC use among women and patients with prior stroke/TIA/TE. In patients with renal disease and in tertiary hospitals, the order of frequency of use was apixaban, rivaroxaban, and edoxaban. Depending on regions, the distributions were different, namely apixaban, edoxaban, and rivaroxaban in the capital city; rivaroxaban, apixaban, and edoxaban in metropolitan cities; and rivaroxaban, edoxaban, and apixaban in other regions (Table 1).

Factors Influencing NOAC Selection

Table 2 presents the results of the multinomial logistic regression analysis using rivaroxaban as a reference. Analysis according to the age group of the patients showed that compared with the 65-69-year age group, the ≥75-year age group was more likely to use apixaban (OR = 1.270, 95% CI = 1.089-1.450) relative to rivaroxaban. Women had lower odds of using dabigatran (OR = 0.821, 95% CI = 0.647-0.995) and higher odds of using apixaban (OR = 1.148, 95% CI = 1.011-1.284). In terms of comorbidities, patients with prior stroke/TIA/TE were more likely to use dabigatran (OR = 1.508, 95% CI = 1.312-1.704) and apixaban (OR = 1.186, 95% CI = 1.026-1.346) and less likely to use edoxaban (OR = 0.793, 95% CI = 0.627-0.960) relative to rivaroxaban. Patients with the renal disease had higher odds of using apixaban (OR = 1.466, 95% CI = 1.238-1.693). Rivaroxaban was generally preferred over dabigatran, apixaban, or edoxaban in medical institutions, excluding tertiary hospitals, and in regions, excluding the capital city. For example, compared with tertiary hospitals, primary medical institutions were less likely to use dabigatran (OR = 0.506, 95% CI = 0.163-0.849), apixaban (OR = 0.441, 95% CI = 0.181-0.701), or edoxaban (OR = 0.574, 95% CI = 0.325-0.824) relative to rivaroxaban. Metropolitan residents had lower odds of using dabigatran (OR = 0.736, 95% CI = 0.510-0.962), apixaban (OR = 0.779, 95% CI = 0.599-0.959), or edoxaban (OR = 0.582, 95% CI = 0.393-0.770) relative to rivaroxaban, compared with capital residents. Patients with persistent AF were less likely to use dabigatran (OR = 0.683, 95% CI = 0.423-0.943) in comparison with rivaroxaban. The P value of likelihood ratio test of this multinomial logistic regression model was <0.001, and the McFadden R² was 0.022.

Table 2.

Adjusted Odds Ratios and 95% Confidence Intervals From Multinomial Logistic Regression Analysis of Non-Vitamin K Antagonist Oral Anticoagulants Utilization.

		Dabigatran vs rivaroxaban		Apixaban vs rivaroxaban		Edoxaban vs rivaroxaban
		Adj. OR	95% CI	Adj. OR	95% CI	Adj. OR	95% CI
Age group	65-69 (R)
	70-74	0.996	(0.747, 1.244)	0.981	(0.773, 1.190)	0.820	(0.625, 1.016)
	≥75	0.933	(0.712, 1.154)	1.270**	(1.089, 1.450)	0.881	(0.711, 1.051)
Gender	Male (R)
Gender	Female	0.821*	(0.647, 0.995)	1.148*	(1.011, 1.284)	1.036	(0.901, 1.170)
Insurance type	NHI (R)
Insurance type	MedAid/PVI	1.202	(0.874, 1.530)	1.148	(0.889, 1.406)	1.259	(1.007, 1.511)
CHF	No (R)
CHF	Yes	0.926	(0.751, 1.100)	1.117	(0.980, 1.255)	0.965	(0.830, 1.100)
Hypertension	No (R)
Hypertension	Yes	0.870	(0.601, 1.139)	0.902	(0.684, 1.119)	1.007	(0.789, 1.226)
Diabetes mellitus	No (R)
Diabetes mellitus	Yes	1.038	(0.862, 1.213)	1.016	(0.878, 1.155)	0.942	(0.806, 1.079)
Prior stroke/TIA/TE	No (R)
Prior stroke/TIA/TE	Yes	1.508**	(1.312, 1.704)	1.186*	(1.026, 1.346)	0.793**	(0.627, 0.960)
Vascular disease	No(R)
Vascular disease	Yes	1.144	(0.773, 1.514)	1.236	(0.949, 1.523)	1.208	(0.922, 1.494)
Renal disease	No(R)
Renal disease	Yes	0.689	(0.338, 1.040)	1.466**	(1.238, 1.693)	0.879	(0.627, 1.131)
Cancer	No(R)
Cancer	Yes	0.923	(0.643, 1.204)	1.009	(0.792, 1.227)	0.984	(0.767, 1.201)
Type of medical institution	Tertiary (R)
	General	0.862	(0.672, 1.051)	0.639**	(0.490, 0.789)	0.996	(0.848, 1.145)
	Primary & others	0.506**	(0.163, 0.849)	0.441**	(0.181, 0.701)	0.574**	(0.325, 0.824)
Region	Capital (R)
	Metropolitan	0.736**	(0.510, 0.962)	0.779**	(0.599, 0.959)	0.582**	(0.393, 0.770)
	Others	0.606**	(0.386, 0.826)	0.659**	(0.484, 0.834)	0.866	(0.697, 1.036)
AF type	Paroxysmal (R)
	Persistent	0.683**	(0.423, 0.943)	0.901	(0.704, 1.099)	0.920	(0.724, 1.116)
	Permanent	0.732	(0.441, 1.023)	0.793	(0.561, 1.025)	1.057	(0.842, 1.271)
	AFL/Others	0.775*	(0.574, 0.976)	0.977	(0.818, 1.135)	1.025	(0.868, 1.182)
Antiplatelet Use	No(R)
Antiplatelet Use	Yes	0.817	(0.522, 1.112)	0.974	(0.749, 1.199)	0.949	(0.722, 1.176)
ATRIA score	≤ 4
	>4	0.912	(0.506, 1.318)	1.111	(0.827, 1.395)	1.127	(0.833, 1.421)
McFadden R2	0.022
Log Likelihood	−7,912.4
LR Test	351.09 (P < 0.001)

Abbreviations: Adj. OR, adjusted odds ratio; AF, atrial fibrillation; AFL, atrial flutter; ATRIA, anticoagulation and risk factors in AF; CHF, congestive heart failure; CI, confidence interval; LR, likelihood ratio; MedAid, Medical Aid; NHI, National Health Insurance; PVI, Patriots & Veterans Insurance; (R), reference; TE, thromboembolism; TIA, transient ischemic attack.

*P < 0.05.

**P < 0.01.

Discussion

In this retrospective study of patients with NVAF taking OACs on the index date, we found that NOACs accounted for the majority of OAC use (87.7%). Among NOACs, rivaroxaban was the most commonly used NOAC, followed by edoxaban, apixaban, and dabigatran. The most commonly used NOAC varies with respect to country or period; however, the domination of NOACs in anticoagulant therapy is observed as a global trend. According to the results of studies conducted in the UK, US, and Denmark, the use of NOACs increased significantly from 2012 to 2015^22
-24,40; moreover, the most used NOACs were rivaroxaban in the UK and apixaban in the US and Denmark.^22
-24

In Korea, the change in health insurance coverage policy significantly influenced NOAC prescription.^41,42 On July 1, 2015, the reimbursement criteria were expanded to allow NOACs as the first-line therapy regardless of whether warfarin was tried. The ratios of NOACs prescribed to total OACs for patients with AF were 5.1%, 36.2%, and 60.8% in 2014, 2015, and 2016, respectively.⁴¹

Our finding that patients with AF who were older than 75 years were more likely to take apixaban vs. rivaroxaban is consistent with that of previous studies.^23,24 This is concordant with the recently published OAC recommendations of apixaban as the first choice for patients older than 75 years.^43
-45 According to the results of a study evaluating the efficacy and safety of apixaban vs. warfarin with respect to age in patients with AF, the absolute benefits of apixaban were greater with increasing age.⁴⁶ The increased use of apixaban vs. rivaroxaban in women is also consistent with that reported in a previous study.²⁴ This could perhaps be associated with apixaban having a weight-related dose-reduction criterion. Apixaban is recommended to be reduced to 2.5 mg twice a day if ≥1 of the following additional criteria are present: age ≥80 years, body weight ≤60 kg, and serum creatinine ≥1.5 mg/dL.^43,44 Considering that women had lower body weights and higher bleeding risk than men, it is presumed that women may have preferred the drug that could maintain its effect with a lower dose.

With respect to comorbidities, significant differences were observed in patients with prior stroke/TIA/TE or renal disease. In our study, in patients with prior stroke/TIA/TE, the use of dabigatran and apixaban increased, and the use of edoxaban decreased compared to rivaroxaban. These results might reflect the relative efficacy between NOACs, manifested as the number of strokes and systemic embolic events in the pivotal NOAC studies compared to warfarin.^4,26
-28 While dabigatran 150 mg and apixaban 5 mg induced significantly lower risks of stroke and systemic embolic events than warfarin (RR = 0.66, 95% CI = 0.53-0.82; HR = 0.79, 95% CI = 0.66-0.95, respectively)^4,27; other agents showed similar²⁸ or opposite tendencies,²⁶ which were not statistically significant. In contrast, in a recent NOAC comparative study of Asians using real-world data, apixaban and edoxaban showed lower rates of ischemic stroke than rivaroxaban and dabigatran.³³ Although no specific NOAC has yet been recommended for secondary stroke prevention,^43,44 preference may change as more research results accumulate in the future.

Increased use of apixaban and decreased use of dabigatran in the presence of renal disease are consistent with reports on renal elimination rates. The renal elimination rates for each NOAC are as follows: 80% for dabigatran, 50% for edoxaban, 33% for rivaroxaban, and 27% for apixaban,^43,47 and according to recent guidelines, apixaban can be used with relatively no dose adjustment for a wide range of renal functions.²⁹

The trends of NOAC usage according to the type of medical institution and region are that the use of apixaban surpasses rivaroxaban only in tertiary hospitals and in the capital city. This is presumed to be due to the more rapid adoption of new treatment trends. Several studies have shown a recent trend of apixaban usage overtaking rivaroxaban usage. According to the previously mentioned study that analyzed variation in OAC choice in patients with AF from 2010 to 2017 using the US claims data, dabigatran prescription peaked in 2011 and then decreased sharply, while rivaroxaban prescription peaked in 2014, after which apixaban surpassed all of them and became the most prescribed OAC for patients with AF since 2015.²⁴ Although edoxaban was not included in that study, the prescription of a newly developed drug can be seen to gradually increase and replace existing drugs. Likewise, in each study conducted in Denmark and England, the results showed high usage of apixaban, rivaroxaban, and dabigatran in that order at the end of the study period.^23,40 This pattern is consistent with the result of a recent meta-analysis that apixaban provided the most favorable efficacy and safety profile.⁴⁸ Meanwhile, according to a study comparing the prescription ratio of NOACs to OACs according to regions of Korea in the latter half of 2016, the ratio of NOACs to OACs in most areas was approximately 60% but was less than 40% in Jeju-do, which is the region farthest from the capital.⁴¹ This result may be in line with that of our study, reflecting the phenomenon of regional variations in the use of new drugs in Korea.

Several observational studies have been conducted to explore the comparative safety of NOACs. A French study involving claims data reported that apixaban had a lower risk of major bleeding than rivaroxaban (HR = 0.67, 95% CI = 0.63-0.72) and comparable risks of major bleeding with dabigatran (HR = 0.93, 95% CI = 0.81-1.08).⁴⁹ Lip et al reported similar results in a propensity score-matched cohort study using US claims data. Rivaroxaban showed a significantly higher risk of major bleeding (HR = 1.82, 95% CI = 1.36-2.43) than apixaban.⁵⁰ In another cohort study using nationwide registries in Norway, dabigatran and apixaban showed a significantly lower risk of major bleeding than rivaroxaban.³⁴ However, we did not find any statistically significant differences in NOAC utilization between high-risk bleeding (ATRIA bleeding score > 4) and low-risk bleeding (ATRIA bleeding score ≤ 4) groups.

The results of this study indicated that patients with persistent AF had lower odds of using dabigatran, relative to rivaroxaban, than patients with paroxysmal AF. Although anticoagulant therapy is recommended regardless of AF type, the current guidelines do not recommend one NOAC over another.

Currently all 4 NOACs are listed in the Korean national formulary with the same levels of copayments and we do not have a Preferred Drug List (PDL) for NOACs. Therefore, we have not considered this as an influencing factor.

For South Korean patients who were covered by NHI, the cost of warfarin was approximately $0.6 per month; whereas, that of NOACs was approximately $20 per month, which was quite different. However, between NOACs, the cost differences were insignificant; moreover, the insurance coverage criteria were the same as those approved when administered to high-risk (CHA₂DS₂-VASc score ≥ 2) NVAF patients, including the first-line therapy. Therefore, it was unlikely that cost influenced NOAC selection; thus, it was excluded from analysis. The results of this study may not be applicable to healthcare systems in which patient out-of-pocket costs are higher or different between NOACs.

This study had some limitations. First, since the claims data we used were collected for reimbursement, the diagnosis information could be susceptible to up-coding by providers for the purpose of increasing the reimbursement rate.⁵¹ Second, these data did not involve clinical data such as laboratory data or disease severity.³⁸ Third, socioeconomic factors such as education and household net worth were not reflected. Fourth, as the analysis was conducted with several factor and outcome variables, the model fit was poor as demonstrated by the very low pseudo R². Fifth, the study drugs identified on the index date were not necessarily the initial therapeutic choice, because the study period was short due to the limitation of the 1-year data. Finally, the factors influencing NOAC selection observed in our study do not suggest that the selected NOACs are more beneficial to patients with such characteristics. Even if the results were comparable to those of other observational studies, this may be due to consistent confounding variables.

Despite these limitations, this study had several important implications. Since most of the Korean population is covered by the national insurance, the research results can be generalized,³⁸ and our analysis of NOAC selection among HIRA-APS may represent treatment decisions for all elderly patients with AF in Korea. We conducted the analysis very recently, at a time when NOACs, including edoxaban, became more widely used as OACs. We analyzed factors influencing NOAC selection, not against warfarin use, but in comparison to each other. This is a study of the recent trends in factors influencing NOAC selection in a real-world setting, and it confirms how consistent these are with recent recommendations. We hope that this study will be helpful in looking back at the NOAC prescription patterns and identifying areas for improvement in selecting NOACs for patients with AF.

Conclusion

This study found that one-third of NOAC users with NVAF were taking rivaroxaban on the index date, while in tertiary hospitals and the capital city, apixaban took the lead. Among NOAC users, the OR of apixaban prescription increased in patients aged 75 years or older, females, and patients with renal disease. These results generally fit well with recent recommendations.

Supplemental Material

Supplemental Material, sj-docx-1-cpt-10.1177_10742484211049919 - Factors Influencing the Selection of Non-Vitamin K Antagonist Oral Anticoagulants for Stroke Prevention in Patients With Non-Valvular Atrial Fibrillation

Supplemental Material, sj-docx-1-cpt-10.1177_10742484211049919 for Factors Influencing the Selection of Non-Vitamin K Antagonist Oral Anticoagulants for Stroke Prevention in Patients With Non-Valvular Atrial Fibrillation by Susin Park and Nam Kyung Je in Journal of Cardiovascular Pharmacology and Therapeutics

Footnotes

Authors' Note

We used the Aged Population Sample data that were collected by the Korea Health Insurance Review and Assessment Service (HIRA-APS-2018-0039); however, the results were not related to the Ministry of Health and Welfare or HIRA.

Author Contributions

SP and NKJ contributed to the conception and design of the study, data analysis and interpretation of the data. Both authors participated in the writing of the manuscript.

Declaration of Conflicting Interests

The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The author(s) received no financial support for the research, authorship, and/or publication of this article.

ORCID iD

Nam Kyung Je

Supplemental Material

Supplemental material for this article is available online.

References

Baetz

Spinler

. Dabigatran etexilate: an oral direct thrombin inhibitor for prophylaxis and treatment of thromboembolic diseases. Pharmacotherapy. 2008;28(11):1354-1373.

Eriksson

Smith

Yasothan

, et al. Dabigatran etexilate. Nat Rev Drug Discov. 2008;7(7):557-558.

AGENCY EM. Pradaxa EPAR. European Medicines Agency. Pradaxa EPAR 2010. Accessed September 21, 2021. https://www.ema.europa.eu/en/medicines/human/EPAR/pradaxa#authorisation-details-section

Connolly

Ezekowitz

Yusuf

, et al. Dabigatran versus warfarin in patients with atrial fibrillation. New Engl J Med. 2009;361(12):1139-1151.

Paikin

Haroun

Eikelboom

. Dabigatran for stroke prevention in atrial fibrillation: the RE-LY trial. Expert Rev Cardiovasc Ther. 2011;9(3):279-286.

Administration FaD. Drug Approval Package: Xarelto (rivaroxaban). 2011. wwwaccessdatafdagov

Administration FaD. Drug Approval Package: Eliquis (apixaban). 2012. wwwaccessdatafdagov

Administration FaD. Drug Approval Package: Savaysa (edoxaban). 2015. wwwaccessdatafdagov

Wolf

Abbott

Kannel

. Atrial fibrillation as an independent risk factor for stroke: the Framingham Study. Stroke. 1991;22(8):983-988.

10.

Wolf

Dawber

Thomas

Jr , et al. Epidemiologic assessment of chronic atrial fibrillation and risk of stroke: the Framingham study. Neurology. 1978;28(10):973-977.

11.

Stewart

Hart

Hole

, et al. A population-based study of the long-term risks associated with atrial fibrillation: 20-year follow-up of the Renfrew/Paisley study. Am J Med. 2002;113(5):359-364.

12.

Hart

Pearce

Aguilar

. Meta-analysis: antithrombotic therapy to prevent stroke in patients who have nonvalvular atrial fibrillation. Ann Intern Med. 2007;146(12):857-867.

13.

Ansell

Hirsh

Hylek

, et al. Pharmacology and management of the vitamin K antagonists: American College of Chest Physicians Evidence-Based Clinical Practice Guidelines (8th Edition). CHEST. 2008;133(6 suppl):160s-198s.

14.

Hankey

Eikelboom

. Dabigatran etexilate. Circulation. 2011;123(13):1436-1450.

15.

Choi

Lee

. Inadequate stroke prevention in Korean atrial fibrillation patients in the post-warfarin era. Int J Cardiol. 2016;220:647-652.

16.

Lee

Choi

Han

, et al. Temporal trends of antithrombotic therapy for stroke prevention in Korean patients with non-valvular atrial fibrillation in the era of non-vitamin K antagonist oral anticoagulants: a nationwide population-based study. PLoS One. 2017;12(12):e0189495.

17.

January

Wann

Calkins

, et al. 2019 AHA/ACC/HRS Focused Update of the 2014 AHA/ACC/HRS Guideline for the management of patients with atrial fibrillation: a report of the American College of Cardiology/American Heart Association Task Force on Clinical Practice Guidelines and the Heart Rhythm Society. J Am Coll Cardiol. 2019;74(1):104-132.

18.

Lip

GYH

Banerjee

Boriani

, et al. Antithrombotic therapy for atrial fibrillation: CHEST guideline and expert panel report. Chest. 2018;154(5):1121-1201.

19.

Lee

Joung

Cha

, et al. Erratum: 2018 KHRS Guidelines for stroke prevention therapy in Korean patients with nonvalvular atrial fibrillation. Korean J Med. 2018;93(3):311-312.

20.

Brieger

Amerena

Attia

, et al. National Heart Foundation of Australia and Cardiac Society of Australia and New Zealand: Australian clinical guidelines for the diagnosis and management of atrial fibrillation 2018. Med J Australia. 2018;209(8):356-362.

21.

Kirchhof

Benussi

Kotecha

, et al. 2016 ESC Guidelines for the management of atrial fibrillation developed in collaboration With EACTS. Revista Española de Cardiología (English Edition). 2017;70(1):50.

22.

Loo

Dell’Aniello

Huiart

, et al. Trends in the prescription of novel oral anticoagulants in UK primary care. Brit J Clin Pharmaco. 2017;83(9):2096-2106.

23.

Staerk

Fosbøl

Gadsbøll

, et al. Non-vitamin K antagonist oral anticoagulation usage according to age among patients with atrial fibrillation: temporal trends 2011-2015 in Denmark. Sci Rep. 2016;6:31477.

24.

Zhu

Alexander

Nazarian

, et al. Trends and variation in oral anticoagulant choice in patients with atrial fibrillation, 2010-2017. Pharmacotherapy. 2018;38(9):907-920.

25.

Brownell

Sezer

Jain

, et al. Trends and predictors of NOAC versus warfarin use for stroke prevention in patients with non-valvular atrial fibrillation. J Am Coll Cardiol. 2020;75(11_supplement_1):518-518.

26.

Patel

Mahaffey

Garg

, et al. Rivaroxaban versus warfarin in nonvalvular atrial fibrillation. New Engl J Med. 2011;365(10):883-891.

27.

Granger

Alexander

McMurray

, et al. Apixaban versus warfarin in patients with atrial fibrillation. New Engl J Med. 2011;365(11):981-992.

28.

Giugliano

Ruff

Braunwald

, et al. Edoxaban versus warfarin in patients with atrial fibrillation. New Engl J Med. 2013;369(22):2093-2104.

29.

Steffel

Verhamme

Potpara

, et al. The 2018 European Heart Rhythm Association Practical Guide on the use of non-vitamin K antagonist oral anticoagulants in patients with atrial fibrillation. Eur Heart J. 2018;39(16):1330-1393.

30.

January

Wann

Calkins

31.

Ruff

Giugliano

Braunwald

, 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. 2014;383(9921):955-962.

32.

Bai

Shi

, et al. Meta-analysis of effectiveness and safety of oral anticoagulants in atrial fibrillation with focus on apixaban. Am J Cardiol. 2017;120(9):1689-1695.

33.

Lee

Choi

Kwon

, et al. Effectiveness and safety of contemporary oral anticoagulants among Asians with nonvalvular atrial fibrillation. Stroke. 2019;50(8):2245-2249.

34.

Rutherford

O-CW

Jonasson

Ghanima

, et al. Comparison of dabigatran, rivaroxaban, and apixaban for effectiveness and safety in atrial fibrillation: a nationwide cohort study. Eur Heart J—Cardiovascular Pharmacotherapy. 2020;6(2):75-85.

35.

Gundlund

Staerk

Fosbøl

, et al.

Initiation of anticoagulation in atrial fibrillation: which factors are associated with choice of anticoagulant?

J Intern Med. 2017;282(2):164-174.

36.

Schaefer

Sood

Haymart

, et al. Sociodemographic factors in patients continuing warfarin vs those transitioning to direct oral anticoagulants. Blood Adv. 2017;1(26):2536-2540.

37.

Balsam

Ozieranski

Tyminska

, et al. Comparison of clinical characteristics of real-life atrial fibrillation patients treated with vitamin K antagonists, dabigatran, and rivaroxaban: results from the CRAFT study. Kardiol Pol. 2018;76(5):889-898.

38.

Kim

M-S

You

S-H

, et al. Conducting and reporting a clinical research using Korean Healthcare Claims Database. Korean J Fam Med. 2020;41(3):146-152.

39.

Service

NHI

. National health insurance system of Korea. 2015. Accessed July 9, 2021. www.kobia.kr/skin/bbs/downloads_e2/download.php?tbl=policy_report&no=401

40.

van Hove

Leng

. Trends in anticoagulant prescribing: a review of local policies in English primary care. BMC Health Serv Res. 2020;20(1):279.

41.

Kim

Park

, et al. Impact of the health insurance coverage policy on oral anticoagulant prescription among patients with atrial fibrillation in Korea from 2014 to 2016. J Korean Med Sci. 2018;33(23):e163.

42.

Yang

P-S

Hwang

, et al. Social inequalities of oral anticoagulation after the introduction of non-vitamin K antagonists in patients with atrial fibrillation. Korean Circ J. 2020;50(3):267-277.

43.

Diener

Aisenberg

Ansell

, et al. Choosing a particular oral anticoagulant and dose for stroke prevention in individual patients with non-valvular atrial fibrillation: part 2. Eur Heart J. 2017;38(12):860-868.

44.

Lee

Joung

Lee

S-R

, et al. 2018 KHRS expert consensus recommendation for oral anticoagulants choice and appropriate doses: specific situation and high risk patients. Korean J Med. 2018;93(2):110-132.

45.

Hammersley

Signy

. Navigating the choice of oral anticoagulation therapy for atrial fibrillation in the NOAC era. Ther Adv Chronic Dis. 2017;8(12):165-176.

46.

Halvorsen

Atar

Yang

, et al. Efficacy and safety of apixaban compared with warfarin according to age for stroke prevention in atrial fibrillation: observations from the ARISTOTLE trial. Eur Heart J. 2014;35(28):1864-1872.

47.

Park

J-K

Kim

Baek

, et al. 2018 KHRS practical guide on the use of non-vitamin K antagonist oral anticoagulants in Korean patients with atrial fibrillation: how to initiate and organize the follow-up. Korean J Med. 2019;94(1):17-39.

48.

Lip

GYH

Mitchell

Liu

, et al. Relative efficacy and safety of non-Vitamin K oral anticoagulants for non-valvular atrial fibrillation: network meta-analysis comparing apixaban, dabigatran, rivaroxaban and edoxaban in three patient subgroups. Int J Cardiol. 2016;204:88-94.

49.

Ganse

Danchin

Mahé

, et al. Comparative safety and effectiveness of oral anticoagulants in nonvalvular atrial fibrillation. Stroke. 2020;51(7):2066-2075.

50.

Lip

Keshishian

Kamble

, 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 Haemostasis. 2016;116(5):975-986.

51.

Kim

J-A

Kim

. A guide for the utilization of Health Insurance Review and Assessment Service National Patient Samples. Epidemiology and Health. 2014;36:e2014008.

Supplementary Material

Please find the following supplemental material available below.

For Open Access articles published under a Creative Commons License, all supplemental material carries the same license as the article it is associated with.

For non-Open Access articles published, all supplemental material carries a non-exclusive license, and permission requests for re-use of supplemental material or any part of supplemental material shall be sent directly to the copyright owner as specified in the copyright notice associated with the article.

0.00 MB

0.57 MB

0.02 MB