Anticoagulant Use in Patients With Nonvalvular Atrial Fibrillation: Has Prescribing Improved?

Abstract

Discordance between international guideline recommendations and anticoagulant prescribing patterns among patients with nonvalvular atrial fibrillation (NVAF) has been frequently reported. This study was designed to compare the anticoagulant utilization pattern to earlier data in the same population and identify predictors of anticoagulant prescribing among patients with NVAF. We reviewed patients with NVAF admitted to Tasmania’s 3 major hospitals between January 2011 and June 2012 and compared the anticoagulant utilization pattern to earlier data. Patients were excluded if they had only 1 episode of NVAF that reverted spontaneously or upon cardioversion. Multivariate logistic regression analysis was used to identify predictors of anticoagulant prescribing. Overall, 53.8% of patients received anticoagulant treatment compared to 40.4% 15 years ago. Among eligible patients at high-risk of stroke, 52.5% were receiving anticoagulant therapy (vs 42.1% 15 years ago). Approximately 10% of patients with a CHADS₂ score ≥2 were not receiving any antithrombotic treatment, reduced from 18.2% in the earlier cohort, whereas anticoagulant use increased among those at low risk (score 0) to 48.5% from 14.2%. Younger age (odds ratio [OR] 0.99, 95% confidence interval [CI] 0.97-1.0; P = .04); CHADS₂ = 1, relative to 0 (OR 1.68, 95% CI 1.07-2.63; P = .02); CHF (OR 1.56, 95% CI 1.12-2.15; P = .008); and embolic disease history (OR 1.77, 95% CI 1.09-2.86; P = .02) were significant predictors of anticoagulant prescribing. While there has been improvement over the past 15 years, suboptimal use of anticoagulant therapy among high-risk patients with NVAF remains common. There is significant potential for improvement in the quality of stroke prophylaxis in patients with NVAF.

Keywords

atrial fibrillation anticoagulants stroke cardiac failure embolism

Introduction

Long-term use of anticoagulant therapy is an important strategy in the management of nonvalvular atrial fibrillation (NVAF), and its efficacy for prevention of stroke in this condition is well established.¹ Vitamin K antagonists (VKAs) and non-VKA oral anticoagulants (NOACs) can be used for thromboprophylaxis. In comparison, antiplatelet therapy has limited utility for preventing stroke in atrial fibrillation (AF).^2,3 Despite the proven benefits of anticoagulant therapy in AF, there have been frequent reports of discordance between guideline recommendations and anticoagulant prescribing patterns.^4,5 Early discontinuation and underuse of anticoagulants in real-world practice have been reported.^5

–8 A retrospective study conducted by Jackson et al 15 years ago (1997-1999) in Tasmania revealed that anticoagulant therapy was underused in high-risk patients with AF.⁹ This study provided local data on stroke prophylaxis measures in AF at a particular point of time and provided the basis for an intervention to improve anticoagulant prescribing. In fact, a community-based educational intervention carried out among general practitioners resulted in a significant increase in the prescribing of warfarin for stroke prevention in Tasmanian patients with AF.¹⁰

Contemporary guidelines recommend prophylaxis with antithrombotic agents in people with AF and at least 1 other risk factor for stroke.^3,11 The guidelines recommend using risk stratification schemes, such as CHADS₂ (congestive heart failure (CHF), hypertension (HTN), age≥75 years, diabetes and previous stroke or transient ischemic attack (TIA))¹² or CHA₂DS₂-VASc (CHF, HTN, diabetes, vascular disease (prior myocardial infarction, peripheral artery disease or aortic plaque), age 65-74 years, female gender, age ≥75 years, and previous stroke or TIA, or thromboembolism (TE)),¹³ to assess the risk of stroke in patients with AF prior to commencing antithrombotic treatment. Despite these guidelines, high risk of stroke has not been found to be an independent predictor of anticoagulant prescribing in several observational studies.^14,15

Data regarding predictors of anticoagulant prescribing among patients with AF are limited in the Australian context. A prospective study conducted across a local health district in Sydney¹⁶ revealed that the likelihood of receiving anticoagulant therapy among patients with NVAF increased by being classified at high risk of stroke. Another study, involving a single hospital site, revealed that patients aged 80 years or older were less likely to be prescribed anticoagulation compared to antiplatelet therapy.¹⁷ The need to bridge the gap in translation of evidence into clinical practice was deemed necessary in both of these studies. We thus designed this study to determine (1) the pattern of anticoagulant utilization, compared to guideline recommendations,^3,11 and predictors of anticoagulant prescribing in patients with NVAF and (2) whether stroke prevention in the management of patients with NVAF had improved over the past 15 years in Tasmania.

Materials and Methods

Study Design

The Tasmanian atrial fibrillation (TAF) study is an ongoing retrospective study that enrolls patients from 3 different hospitals in Tasmania, Australia: the Royal Hobart Hospital (RHH), Launceston General Hospital (LGH), and North West Regional Hospital (NWRH). We identified patients from the Medical Record Departments of these hospitals using Australian Refined Diagnosis Related Groups (AR-DRGs) codes. Medical records of 2502 patients admitted between January 1, 2011, and June 30, 2012, and diagnosed with valvular or NVAF at discharge (AR-DRG code 148: atrial fibrillation or flutter) were reviewed for this study. Patients diagnosed with AF as their primary (ie, AF was the presenting complaint) or secondary condition (ie, AF was listed as a current illness in the medical history or discharge summary) were included. Patients were excluded if they had only 1 episode of AF that reverted spontaneously or upon cardioversion without any documented recurrences, as stroke prophylaxis may not be warranted in these patients. Of 2502 medical records reviewed, 1469 patients were included (RHH: 777, NWRH: 289, and LGH: 403) and 1033 were excluded (episode of AF that reverted spontaneously or upon cardioversion: 590, developed AF as a short-term complication: 288, and no documented AF [coding error]: 155). Seventy-eight (5.3%) index admissions were excluded due to death of the patient or unavailability of their records, leaving 1391 patients. Of these, 1261 patients were eligible for anticoagulant therapy (ie, did not possess contraindications). After excluding 144 (11.4%) patients with valvular AF, 1117 patients with NVAF were included in this analysis.

Data regarding patient demographics, medications on admission, history of comorbid conditions, and discharge antithrombotic medications were entered into an electronic database. The Charlson Comorbidity Index¹⁸ was used as a measure of comorbidity. We evaluated CHADS₂ ¹² and CHA₂DS₂-VASc¹³ for assessing stroke risk and the HAS-BLED (HTN, abnormal renal function, abnormal liver function, bleeding predisposition, age >65 years, the use of drugs predisposing patients to bleeding (non-steroidal anti-inflammatory drugs), alcohol use (>8 drinks per week), previous stroke and labile international normalized ratios (INRs; if documented))¹⁹ score for assessing bleeding risk. We defined labile INRs as unstable/high INRs as recorded in the medical record of our patients. A patient’s first admission to a hospital with the diagnosis of AF within our study period was defined as the index admission. Contraindications to anticoagulant therapy included patients with a documented history of dementia, labile INRs, bleeding diseases, allergies to anticoagulant therapy, and pregnancy.

Our definition of anticoagulant therapy for this study comprised anticoagulant therapy with or without concomitant antiplatelet therapy. We utilized the recommendations from the European Society of Cardiology 2010 guidelines and American College of Chest Physicians, 9th Edition,^3,11 to examine the appropriateness of anticoagulant prescribing at discharge of the index admission. We considered underutilization as no prescribing of anticoagulant therapy to patients without contraindications to anticoagulant therapy and a CHADS₂≥2 per these guidelines. The CHA₂DS₂-VASc scores were not available to clinicians to guide treatment during our data collection period. We also compared the changes in anticoagulant prescribing patterns referring to a local study conducted by Jackson et al (1997-1999), 15 years ago in Tasmania.⁹ This study was a retrospective review of 505 patients with NVAF admitted to RHH. The appropriateness of antithrombotic therapy was assessed both on admission and on discharge from the hospital care, and sufficient data were available to calculate the CHADS₂ scores retrospectively. The study also determined the clinical outcomes (ischemic stroke, systemic embolism, or bleeding complications with warfarin or aspirin use) in the total population.

Statistical Analysis

Data were analyzed using SPSS version 21.0 (Prentice Hall, New Jersey). Continuous variables were expressed as median (interquartile range). Categorical variables were expressed as frequencies and percentages. Chi-square tests were used to compare categorical variables, while Mann-Whitney U tests were used for continuous variables. Multivariate logistic regression modeling was used to identify the independent risk factors associated with anticoagulant prescribing in eligible patients. Variables that had P ≤ .2 in univariate analyses were combined in a multivariate logistic regression model (enter method). P < .05 was considered as statistically significant for all analyses.

Ethics approval for the project was obtained from Tasmanian Health and Medical Human Research Ethics Committee. As the study was a retrospective prescribing audit, it was not deemed necessary to ask for informed consent from the study participants.

Results

Clinical Characteristics

The overall rate of anticoagulant use at discharge from the index admission among eligible patients was 53.8%. Characteristics of the patients anticoagulated and not anticoagulated at discharge are shown in Table 1. Patients treated with anticoagulants were significantly younger, inclined to be male, and more likely to have CHF and a history of embolic events.

Table 1.

Baseline Characteristics of Patients With NVAF Anticoagulated and Not Anticoagulated at Discharge From Index Admission (n = 1117).

Variable	Anticoagulated (n = 601)	Not Anticoagulated (n = 516)	P Value
Age, years, mean (SD)	74.4 (± 12.0)	76.1 (± 12.6)	.002^a
Gender, n (%)
Male	358 (59.6)	273 (52.9)	.03^a
Medical history, n (%)
HTN	404 (67.2)	336 (65.1)	.46
CHF	154 (25.6)	96 (18.6)	.005^a
Diabetes	139 (23.1)	97 (18.8)	.08
Cerebrovascular disease	108 (18.0)	100 (19.4)	.55
MI	108 (18.0)	79 (15.3)	.24
Embolic events (DVT, pulmonary embolism, and so on)	55 (9.2)	28 (5.4)	.02^a
Peripheral vascular disease	41 (6.8)	29 (5.6)	.41
CHADS₂, median (IQR)	2.0 (1-3)	2.0 (1-3)	.96
Low (score 0)	63 (10.5)	67 (13.0)	.06
Intermediate (score 1)	165 (27.5)	112 (21.7)
High (score 2-6)	373 (62.1)	337 (65.3)
CHA₂DS₂-VASc, median (IQR)	4 (2-5)	4 (2-5)	.99
Low (score 0)	16 (2.7)	20 (3.9)	.20
Intermediate (score 1)	53 (8.8)	54 (10.5)
High (score 2-6)	532 (88.5)	442 (85.7)
HAS-BLED, median (IQR)	2.0 (1-2)	2.0 (1-2)	.21
Low (score 0)	45 (7.5)	38 (7.4)	.60
Intermediate (1-2)	451 (75.0)	376 (72.9)
High (≥3)	105 (17.5)	102 (19.8)
CCI, median (IQR)	5.0 (4-7)	5.0 (4-6)	.43
0	50 (8.3)	46 (8.9)	.43
1-2	26 (4.3)	16 (3.1)
3-4	134 (22.3)	132 (25.6)
≥5	391 (65.1)	322 (62.4)
AF duration, n (%)
New onset	153 (26.2)	159 (31.5)	.06
Old	430 (73.8)	346 (68.5)

Abbreviations: NVAF, nonvalvular atrial fibrillation; SD, standard deviation; HTN, hypertension; CHF, congestive heart failure; MI, myocardial infarction; DVT, deep vein thrombosis; CHADS₂, CHF, HTN, age ≥75 years, diabetes and previous stroke or transient ischemic attack; IQR, interquartile range; CHA₂DS₂-VASc, CHF, HTN, diabetes, vascular disease (prior MI, peripheral artery disease or aortic plaque), age 65 to 74 years, female gender, age ≥75 years, and previous stroke or transient ischemic attack, or thromboembolism; HAS-BLED, HTN, abnormal renal function, abnormal liver function, bleeding predisposition, age >65 years, the use of drugs predisposing patients to bleeding nonsteroidal anti-inflammatory drugs , alcohol use (>8 drinks per week), previous stroke and labile international normalized ratios (if documented); CCI, Charlson Comorbidity Index; AF, atrial fibrillation.

^aP < .05.

Anticoagulant Prescribing Pattern Among Patients With NVAF at Discharge

Among those anticoagulated in the total population (n = 601 of 1117), the majority of the patients were taking warfarin (61.7%) followed by a combination of warfarin and antiplatelet (24.8%), dabigatran (2.5%), and dabigatran plus antiplatelet (0.7%). The proportion of eligible patients treated with anticoagulant alone, combination therapy, antiplatelet alone, or no treatment at various levels of stroke risk (CHADS₂) is shown in Figure 1. Of patients with a high risk of stroke (CHADS₂ ≥ 2), 52.5% (n = 373 of 710) were prescribed anticoagulant therapy. Two hundred and sixty four (37.2%) of these patients were prescribed antiplatelet therapy, and 73 (10.3%) patients were not on any antithrombotic treatment. Physician’s decision was the most commonly documented reason for not providing anticoagulant treatment at discharge to the eligible patients at high risk of stroke (n = 44; 6.2%), followed by “patient refusal” (n = 22; 3.1%), “risk of falls” (n = 22; 3.1%), “nonadherence” (n = 5; 0.7%), and “adverse drug reaction” (n = 3; 0.4%). Among those with a high risk of stroke (CHADS₂ ≥ 2), only 3.2% (n = 23) had a HAS-BLED score that exceeded their CHADS₂ score. Among those at lower risk of stroke (CHADS₂ = 0) and receiving anticoagulant therapy, only 6.2% (n = 8 of 130) had a history of embolic diseases other than AF where anticoagulation may have been indicated.

Figure 1.

Antithrombotic prescribing pattern among patients with NVAF based on CHADS₂ score. NVAF indicates nonvalvular atrial fibrillation; CHADS₂, congestive heart failure (CHF), hypertension, age ≥75 years, diabetes and previous stroke or transient ischemic attack.

When we compared the rate of anticoagulant prescribing among high-risk patients to that 15 years ago in Tasmania, there appeared to be an increase in the rate (52.5% vs 42.1%) of prescribing. Nearly 10% of the high-risk patients did not receive any therapy (18.2%, 15 years ago) and 48.5% of low-risk patients were anticoagulated (14.2% in the earlier cohort).

Factors Associated With Anticoagulant Prescribing at Discharge

We considered patients with NVAF and without contraindications as eligible for anticoagulant therapy. Multivariate logistic regression included 7 variables: age, gender, stratified CHADS₂ (with CHADS₂ = 0 as reference category), history of diabetes, CHF, embolic disease, and preadmission AF. Younger age (odds ratio [OR] 0.99, 95% confidence interval [CI] 0.97-1.0; P = .04), CHADS₂ = 1 (OR 1.68, 95% CI 1.07-2.63; P = .02), CHF (OR 1.56, 95% CI 1.12-2.15; P = .008), and embolic disease history (OR 1.77, 95% CI 1.09-2.86; P = .02) were independently associated with anticoagulant prescribing (Table 2).

Table 2.

Univariate and Multivariate Logistic Regression of Risk Factors for Anticoagulant Prescribing Among Eligible Patients With NVAF.

Odds of Anticoagulant Prescribing	Univariate OR (95% CI)	P Value	Multivariate OR (95% CI)	P Value
Age	0.99 (0.98-1.0)	.01^a	0.99 (0.97-1.0)	.04^a
Sex
Female	0.76 (0.60-0.97)	.03^a	0.84 (0.65-1.08)	.18
CHADS₂
0 (ref)
1	1.57 (1.03-2.38)	.04^a	1.68 (1.07-2.63)	.02^a
2	1.08 (0.72-1.60)	.72	1.13 (0.70-1.83)	.62
≥3	1.30 (0.87-1.95)	.20	1.19 (0.69-2.03)	.53
Comorbidities
CHF	1.51 (1.13-2.01)	.005^a	1.56 (1.12-2.15)	.008^a
HTN	1.10 (0.86-1.41)	.46
Diabetes	1.30 (0.97-1.74)	.08	1.25 (0.90-1.75)	.19
CVD	0.91 (0.67-1.23)	.55
Embolic history	1.76 (1.10-2.81)	.02^a	1.77 (1.09-2.86)	.02^a
AF onset
New (ref)
Old	1.29 (0.99-1.68)	.06	1.18 (0.90-1.55)	.25

Abbreviations: NVAF, nonvalvular atrial fibrillation; OR, odds ratio; CI, confidence interval; CHADS₂, congestive heart failure (CHF), hypertension (HTN), age ≥75 years, diabetes and previous stroke or transient ischemic attack; CVD, cerebrovascular disease; AF, atrial fibrillation.

^aP < .05.

Discussion

Nearly two-third of our patient cohort was at high risk of stroke. We observed underutilization of anticoagulant therapy among these eligible high-risk patients. Our data suggest that, despite an improvement in prescribing of anticoagulant therapy over the past 15 years, there remains significant potential for reduction in stroke outcomes with improved use of antithrombotic prophylaxis in the TAF population. Despite guideline recommendations, we did not observe CHADS₂ ≥ 2 as a significant predictor of anticoagulant prescribing in our population.

Underuse of anticoagulant therapy among high-risk groups has been frequently reported.^5,16,20,21 In our study, only half of the patients in the high-risk group were receiving anticoagulant treatment. There was an apparent preference for lone antiplatelet therapy, with nearly one-third of our high-risk population receiving this therapy.

The observed improvements in anticoagulant prescribing among high-risk patients with AF in Tasmania most likely reflect increased focus on the importance of effective thromboprophylaxis as recommended in international guidelines as well as results of previous local intervention studies.^3,11,22 In contrast, we observed that the proportion of low-risk patients receiving anticoagulant therapy was higher now than 15 years ago. This appears to reflect current trends, however, as use of anticoagulant therapy in this group was similar to the Global Anticoagulation Registry in the FIELD (GARFIELD) study.⁵ Embolic diseases other than AF may have resulted in some patients at low risk of stroke receiving anticoagulant therapy in our study, as embolic history was an independent predictor of anticoagulation. Alternatively, physicians’ clinical judgement of stroke risk may have compelled them to consider factors beyond those included in CHADS₂.⁵

Several observational studies have reported on factors influencing anticoagulant prescribing patterns.^23,24 Although prescribing guidelines for patients with intermediate risk of stroke were unclear at the time of our study (antiplatelet or anticoagulant therapy was indicated),^3,11 the association between intermediate stroke risk (CHADS₂ score of 1) and anticoagulation use, instead of high risk (CHADS₂ ≥ 2), relative to a CHADS₂ score of 0 was surprising. Existing guidelines clearly recommended prescription of anticoagulant therapy among those with a CHADS₂ ≥ 2,^3,11 but in alignment with previous studies conducted in China and Turkey,^14,15 we did not find CHADS₂ ≥ 2 as a predictor of anticoagulant prescribing. We hypothesized that this finding may have been due to high bleeding risk among these patients (given the overlap between stroke and bleeding risk factors), but only 3.2% of patients within this group had an HAS-BLED score higher than their CHADS₂ score. We identified physicians’ decision to be the most common reason for not prescribing anticoagulant treatment to high-risk patients without any contraindications. Bleeding is one of the most feared complications of anticoagulant use among treating physicians²⁵; this may have influenced physicians’ prescribing decisions in our study. This apparent discrepancy may further highlight the previously documented issue of quantifiable versus perceived bleeding risk by clinicians, and this issue requires further study if we are to make meaningful improvements in underprescribing of anticoagulants in patients at high risk of stroke.

A history of CHF and embolic disease and younger age were the only other independent predictors of anticoagulant prescribing in our population. Congestive heart failure has been previously associated with a higher rate of oral anticoagulation use,²¹ and previous TE is now clearly acknowledged as a risk factor for stroke through its inclusion in the CHA₂DS₂-VASc scoring system.¹³ We observed older age as a negative predictor for anticoagulant prescribing. Studies have identified underuse of anticoagulant therapy among elderly people due to the fear of bleeding.^26,27 In contrast, benefits of anticoagulant therapy have been proven in clinical trials conducted among elderly patients,^28,29 and hence their use should not be discounted among elderly high-risk groups.

Limitations

The retrospective nature of our registry needs to be acknowledged. Our study is subject to limitations such as missing or incomplete documentation.

Conclusion

Our results suggest that there has been some improvement in the use of anticoagulant therapy among high-risk patients with AF over the past 15 years. Despite this improvement, further gains are required, and improved adherence to stroke risk stratification schemes for antithrombotic prophylaxis in AF could potentially reduce stroke outcomes in our population.

Footnotes

Authors’ Note

DB contributed to design of the work; acquisition, analysis, or interpretation of data; literature search; and manuscript writing. LC contributed to the conception of the work, manuscript drafting, critical revising, and final approval of the version to be published. GP contributed to the conception of the work, critical manuscript review, and final approval of the version to be published. LB contributed to the conception of the work, critical manuscript revising, and final approval of the version to be published.

Acknowledgments

We would like to acknowledge the Royal Hobart Hospital Research Foundation (RHHRF) for providing funding (C0020642) to conduct our study. We would also like to thank Mr Peter Gee for his support in designing the electronic TAF database and in the data extraction process. We would also like to thank Mr Chris Ballard, Dr Carley Bagley, Dr Georgia McGuinness, and Mr Cameron Wright for their help in data collection.

Declaration of Conflicting Interests

The author(s) declared the following potential conflicts of interest with respect to the research, authorship, and/or publication of this article: Leanne Chalmers, Luke RE Bereznicki, and Gregory M. Peterson have received consultancy funding from Aspen Pharmacare Australia and Boehringer Ingelheim. Luke RE Bereznicki has also received consultancy funding from Roche Diagnostics Australia and Sanofi Aventis and speaker honorarium payments from Roche Diagnostics Australia and Boehringer Ingelheim.

Funding

The author(s) disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: The author(s) received financial support from the Royal Hobart Hospital Research Foundation for the research (C0020642).

References

Fuster

Rydén

Cannom

. 2011 ACCF/AHA/HRS focused updates incorporated into the ACC/AHA/ESC 2006 Guidelines for the management of patients with atrial fibrillation: a report of the American College of Cardiology Foundation/American Heart Association Task Force on Practice Guidelines developed in partnership with the European Society of Cardiology and in collaboration with the European Heart Rhythm Association and the Heart Rhythm Society. J Am Coll Cardiol. 2011;57(11):e101–e98.

Lip

. The role of aspirin for stroke prevention in atrial fibrillation. Nat Rev Cardiol. 2011;8(10):602–606.

Camm

Lip

De Caterina

. 2012 focused update of the ESC Guidelines for the management of atrial fibrillation: An update of the 2010 ESC Guidelines for the management of atrial fibrillation developed with the special contribution of the European Heart Rhythm Association. Europace. 2012;14(10):1385–1413.

Sabouret

Bricard

Hermann

Cotté

Deret-Bixio

Rushton-Smith

. Discrepancy between guidelines for stroke prevention in atrial fibrillation and practice patterns in primary care. The nationwide French AFIGP survey. Arch Cardiovasc Dis. 2015;108(11)544–553.

Kakkar

Mueller

Bassand

. Risk profiles and antithrombotic treatment of patients newly diagnosed with atrial fibrillation at risk of stroke: perspectives from the international, observational, prospective GARFIELD registry. PLoS One. 2013;8(5):e63479.

Spivey

Qiao

Liu

. Discontinuation/Interruption of Warfarin Therapy in Patients with Nonvalvular Atrial Fibrillation. J Manag Care Spec Pharm. 2015;21(7):596–606.

Pugh

Mead

. Attitudes of physicians regarding anticoagulation for atrial fibrillation: a systematic review. Age Ageing. 2011;40(6):675–683.

Holbrook

Pereira

Labiris

. Systematic overview of warfarin and its drug and food interactions. Arch Intern Med. 2005;165(10):1095–1106.

Jackson

Peterson

Vial

Daud

Ang

. Outcomes in the management of atrial fibrillation: clinical trial results can apply in practice. Intern Med J. 2001;31(6):329–336.

10.

Jackson

Peterson

Vial

. A community-based educational intervention to improve antithrombotic drug use in atrial fibrillation. Ann Pharmacother. 2004;38(11):1794–1799.

11.

You

Singer

Howard

. Antithrombotic therapy for atrial fibrillation: Antithrombotic Therapy and Prevention of Thrombosis, 9th ed: American College of Chest Physicians Evidence-Based Clinical Practice Guidelines. Chest. 2012;141(2 suppl):e531S–e575S.

12.

Gage

Waterman

Shannon

Boechler

Rich

Radford

. Validation of clinical classification schemes for predicting stroke: results from the National Registry of Atrial Fibrillation. JAMA. 2001;285(22):2864–2870.

13.

Lip

Nieuwlaat

Pisters

Lane

Crijns

. Refining clinical risk stratification for predicting stroke and thromboembolism in atrial fibrillation using a novel risk factor-based approach: the euro heart survey on atrial fibrillation. Chest. 2010;137(2):263–272.

14.

Liu

Xuan

. Treatment patterns associated with stroke prevention in patients with atrial fibrillation in three major cities in the People’s Republic of china. Int J Gen Med. 2014;7:29–35.

15.

Ertaş

Kaya

Atılgan

. Predictors of warfarin use in patients with non-valvular atrial fibrillation who presented to the cardiology outpatient clinic of a tertiary hospital in Turkey: an observational study. Turk J Med Sci. 2012;42(suppl 1):1172–1179.

16.

Gallagher

Roach

Sadler

. Who gets stroke prevention? Stroke prevention in atrial fibrillation patients in the inpatient setting. Heart Lung Circ. 2015;24(5):488–494.

17.

Bajorek

Ren

. Utilisation of antithrombotic therapy for stroke prevention in atrial fibrillation in a Sydney hospital: then and now. Int J Clin Pharm. 2012;34(1):88–97.

18.

Charlson

Pompei

Ales

MacKenzie

. A new method of classifying prognostic comorbidity in longitudinal studies: development and validation. J Chronic Dis. 1987;40(5):373–383.

19.

Pisters

Lane

Nieuwlaat

de Vos

Crijns

Lip

. A Novel user-friendly score (HAS-BLED) to assess 1-Year risk of major bleeding in patients with atrial fibrillation: the Euro Head Survey. Chest. 2010;138(5):1093–1100.

20.

Ogilvie

Newton

Welner

Cowell

Lip

. Underuse of oral anticoagulants in atrial fibrillation: a systematic review. Am J Med. 2010;123(7):638–645.

21.

Agarwal

Bennett

Smith

. Predictors of warfarin use in atrial fibrillation patients in the inpatient setting. Am J Cardiovasc Drugs. 2010;10(1):37–48.

22.

Jackson

Peterson

. Stroke risk assessment for atrial fibrillation: hospital-based stroke risk assessment and intervention program. J Clin Pharm Ther. 2011;36(1):71–79.

23.

Bahri

Roca

Lechani

. Underuse of oral anticoagulation for individuals with atrial fibrillation in a nursing home setting in France: comparisons of resident characteristics and physician attitude. J Am Geriatr Soc. 2015;63(1):71–76.

24.

Coll-Vinent

Martin

Malagon

. Stroke prophylaxis in atrial fibrillation: searching for management improvement opportunities in the emergency department: the HERMES-AF study. Ann Emerg Med. 2015;65(1):1–12.

25.

Man-Son-Hing

Laupacis

. Anticoagulant-related bleeding in older persons with atrial fibrillation: physicians’ fears often unfounded. Arch Intern Med. 2003;163(13):1580–1586.

26.

Lip

Laroche

Dan

G-A

. ‘Real-world’antithrombotic treatment in atrial fibrillation: the EORP-AF pilot survey. Am J Med. 2014;127(6):519–529.

27.

Gamra

Murin

Chiang

. Use of antithrombotics in atrial fibrillation in Africa, Europe, Asia and South America: Insights from the International RealiseAF Survey. Arch Cardiovasc Dis. 2014;107(2):77–87.

28.

Mant

Hobbs

Fletcher

. Warfarin versus aspirin for stroke prevention in an elderly community population with atrial fibrillation (the Birmingham Atrial Fibrillation Treatment of the Aged Study, BAFTA): a randomised controlled trial. Lancet. 2007;370(9586):493–503.

29.

Rash

Downes

Portner

. A randomised controlled trial of warfarin versus aspirin for stroke prevention in octogenarians with atrial fibrillation (WASPO). Age Ageing. 2007;36(2):151–156.