Comparative Effectiveness and Safety of Oral Anticoagulants by Dementia Status in Older Patients With Atrial Fibrillation

Kueiyu Joshua Lin; Daniel E Singer; Katsiaryna Bykov; Lily G Bessette; Julianna M Mastrorilli; Alexander Cervone; Dae Hyun Kim

doi:10.1001/jamanetworkopen.2023.4086

. 2023 Mar 28;6(3):e234086. doi: 10.1001/jamanetworkopen.2023.4086

Comparative Effectiveness and Safety of Oral Anticoagulants by Dementia Status in Older Patients With Atrial Fibrillation

Kueiyu Joshua Lin ^1,^2,^✉, Daniel E Singer ², Katsiaryna Bykov ¹, Lily G Bessette ¹, Julianna M Mastrorilli ¹, Alexander Cervone ¹, Dae Hyun Kim ^1,^3,⁴

¹Division of Pharmacoepidemiology and Pharmacoeconomics, Department of Medicine, Brigham and Women’s Hospital and Harvard Medical School, Boston, Massachusetts

²Division of General Internal Medicine, Department of Medicine, Massachusetts General Hospital, Boston

³Hinda and Arthur Marcus Institute for Aging Research, Hebrew SeniorLife, Boston, Massachusetts

⁴Division of Gerontology, Department of Medicine, Beth Israel Deaconess Medical Center, Boston, Massachusetts

Accepted for Publication: February 4, 2023.

Published: March 28, 2023. doi:10.1001/jamanetworkopen.2023.4086

^✉

Corresponding Author: Kueiyu Joshua Lin, MD, ScD, MPH, Division of Pharmacoepidemiology and Pharmacoeconomics, Department of Medicine, Brigham and Women’s Hospital, Harvard Medical School, 1620 Tremont St, Ste 3030, Boston, MA 02120 (jklin@bwh.harvard.edu).

Author Contributions: Dr Lin had full access to all of the data in the study and takes responsibility for the integrity of the data and the accuracy of the data analysis.

Concept and design: Lin, Bessette, Kim.

Acquisition, analysis, or interpretation of data: All authors.

Drafting of the manuscript: Lin, Mastrorilli, Cervone.

Critical revision of the manuscript for important intellectual content: Lin, Singer, Bykov, Bessette, Kim.

Statistical analysis: Lin, Bessette, Mastrorilli, Cervone.

Obtained funding: Lin, Kim.

Administrative, technical, or material support: Lin, Bykov, Bessette, Mastrorilli, Cervone.

Supervision: Lin, Bessette, Kim.

Conflict of Interest Disclosures: Dr Singer reported receiving grants from Bristol-Myers Squibb and personal fees from Bristol-Myers Squibb, Fitbit, Medtronic, and Pfizer outside the submitted work. Dr Bykov reported receiving personal fees from Alosa Health outside the submitted work. Dr Kim reported receiving personal fees from Alosa Health and VillageMD outside the submitted work. No other disclosures were reported.

Funding/Support: This study was supported by grants 1RF1AG063381-01 (Dr Lin) and 1R01AG062713-01 (Dr Kim) and career development award K01AG068365 (Dr Bykov) from the National Institute on Aging.

Role of the Funder/Sponsor: The funding organization had no role in the design and conduct of the study; collection, management, analysis, and interpretation of the data; preparation, review, or approval of the manuscript; and decision to submit the manuscript for publication.

Data Sharing Statement: See Supplement 2.

^✉

Corresponding author.

PMCID: PMC10051113 PMID: 36976562

Key Points

Question

What are the comparative effectiveness and safety of oral anticoagulants (OACs) by dementia status in older patients with atrial fibrillation (AF)?

Findings

In this comparative effectiveness study of 1 160 462 patients with AF with and without dementia, apixaban was associated with net clinical benefits compared with other OACs and a greater absolute risk reduction in patients with vs without dementia. The rate difference of the composite outcome of ischemic stroke or major bleeding events per 1000 person-years over 6 months was 29.8 vs 16.0 for warfarin vs apixaban, 29.6 vs 5.8 for dabigatran vs apixaban, and 20.5 vs 15.9 for rivaroxaban vs apixaban in patients with vs without dementia.

Meaning

These results support the use of apixaban for OAC therapy in persons living with dementia who have AF.

Abstract

Importance

The development of an optimal stroke prevention strategy, including the use of oral anticoagulant (OAC) therapy, is particularly important for patients with atrial fibrillation (AF) who are living with dementia, a condition that increases the risk of adverse outcomes. However, data on the role of dementia in the safety and effectiveness of OACs are limited.

Objective

To assess the comparative safety and effectiveness of specific OACs by dementia status among older patients with AF.

Design, Setting, and Participants

This retrospective comparative effectiveness study used 1:1 propensity score matching among 1 160 462 patients 65 years or older with AF. Data were obtained from the Optum Clinformatics Data Mart (January 1, 2013, to June 30, 2021), IBM MarketScan Research Database (January 1, 2013, to December 31, 2020), and Medicare claims databases maintained by the Centers for Medicare & Medicaid Services (inpatient, outpatient, and pharmacy; January 1, 2013, to December 31, 2017). Data analysis was performed from September 1, 2021, to May 24, 2022.

Exposures

Apixaban, dabigatran, rivaroxaban, or warfarin.

Main Outcomes and Measures

Composite end point of ischemic stroke or major bleeding events over the 6-month period after OAC initiation, pooled across databases using random-effects meta-analyses.

Results

Among 1 160 462 patients with AF, the mean (SD) age was 77.4 (7.2) years; 50.2% were male, 80.5% were White, and 7.9% had dementia. Three comparative new-user cohorts were established: warfarin vs apixaban (501 990 patients; mean [SD] age, 78.1 [7.4] years; 50.2% female), dabigatran vs apixaban (126 718 patients; mean [SD] age, 76.5 [7.1] years; 52.0% male), and rivaroxaban vs apixaban (531 754 patients; mean [SD] age, 76.9 [7.2] years; 50.2% male). Among patients with dementia, compared with apixaban users, a higher rate of the composite end point was observed in warfarin users (95.7 events per 1000 person-years [PYs] vs 64.2 events per 1000 PYs; adjusted hazard ratio [aHR], 1.5; 95% CI, 1.3-1.7), dabigatran users (84.5 events per 1000 PYs vs 54.9 events per 1000 PYs; aHR, 1.5; 95% CI, 1.2-2.0), and rivaroxaban users (87.4 events per 1000 PYs vs 68.5 events per 1000 PYs; aHR, 1.3; 95% CI, 1.1-1.5). In all 3 comparisons, the magnitude of the benefits associated with apixaban was similar regardless of dementia diagnosis on the HR scale but differed substantially on the rate difference (RD) scale. The adjusted RD of the composite outcome per 1000 PYs for warfarin vs apixaban users was 29.8 (95% CI, 18.4-41.1) events in patients with dementia vs 16.0 (95% CI, 13.6-18.4) events in patients without dementia. The corresponding adjusted RD estimates of the composite outcome were 29.6 (95% CI, 11.6-47.6) events per 1000 PYs in patients with dementia vs 5.8 (95% CI, 1.1-10.4) events per 1000 PYs in patients without dementia for dabigatran vs apixaban users and 20.5 (95% CI, 9.9-31.1) events per 1000 PYs in patients with dementia vs 15.9 (95% CI, 11.4-20.3) events per 1000 PYs in patients without dementia for rivaroxaban vs apixaban users. The pattern was more distinct for major bleeding than for ischemic stroke.

Conclusions and Relevance

In this comparative effectiveness study, apixaban was associated with lower rates of major bleeding and ischemic stroke compared with other OACs. The increased absolute risks associated with other OACs compared with apixaban were greater among patients with dementia than those without dementia, particularly for major bleeding. These findings support the use of apixaban for anticoagulation therapy in patients living with dementia who have AF.

This comparative effectiveness study uses data from 3 US claims databases to assess the comparative safety and effectiveness of specific oral anticoagulant therapies by dementia status among older patients with atrial fibrillation.

Introduction

Nonvalvular atrial fibrillation (AF) is highly prevalent in older adults¹ and is associated with a 5-fold increased risk of embolic stroke.² Anticoagulation can reduce this risk by approximately 70%, and clinical guidelines^3,4 recommend that patients with AF who have an estimated annual stroke risk of 2.2% or greater (based on patient characteristics) receive anticoagulation therapy,^5,6 which means the majority of older adults with AF should receive an oral anticoagulant (OAC).^7,8 However, the prescribing decisions for OACs need to be individualized according to comorbidities and frailty because the risk-benefit profiles of OACs can vary substantially in different risk groups.⁹ The presence of dementia often complicates such a decision because persons living with dementia are at higher risk of falls, traumatic intracranial bleeding, medication errors, and low adherence to treatment regimens.^10,11 Because dementia affects more than 6 million persons in the US¹² and approximately 11% of older adults with AF,^13,14 it is important to have an optimal stroke prevention strategy for persons living with dementia who have AF.

While the efficacy and safety of direct oral anticoagulants (DOACs) vs warfarin in patients with AF has been well established in randomized clinical trials (RCTs),^15,16,17,18 persons living with dementia and older adults with frailty were substantially underrepresented in these trials. The small number of persons living with dementia who can consent and adhere to the study protocol of an RCT is unlikely to be representative of those living with dementia who are receiving routine care in terms of severity of cognitive and mobility impairment.^19,20 In the absence of RCT evidence, analyses of routinely collected clinical data must be relied on to inform optimal choices for stroke prevention strategies among persons living with dementia who have AF. However, there are few RCT data on the role of dementia in the safety and effectiveness of OACs.⁹ We aimed to examine the comparative safety and effectiveness of specific OACs by assessing the risks of ischemic stroke and major bleeding events by dementia status among older patients with AF.

Methods

Study Population Based on OAC Exposure

We conducted a comparative effectiveness study using data from 3 large US claims databases: Optum Clinformatics Data Mart (January 1, 2013, to June 30, 2021), IBM MarketScan Research Database (January 1, 2013, to December 31, 2020), and Medicare fee-for-service Part A (inpatient claims), Part B (outpatient claims), and Part D (pharmacy claims; January 1, 2013, to December 31, 2017) databases maintained by the Centers for Medicare & Medicaid Services. Among adults 65 years or older, we established 3 comparative cohorts of new users of warfarin vs apixaban, dabigatran vs apixaban, and rivaroxaban vs apixaban who filled a prescription for the specific OAC without use of any OAC in the preceding 183 days from January 1, 2013, through June 30, 2021. Data analysis was performed from September 1, 2021, to May 24, 2022. This study was approved by the institutional review board of Brigham and Women’s Hospital in Boston, Massachusetts. A waiver of informed consent was granted because the study was a secondary analysis of routinely collected data that were unidentified. This study followed the Strengthening the Reporting of Observational Studies in Epidemiology (STROBE) reporting guideline for cohort studies.²¹

We used apixaban as the referent group because this medication has become the preferred OAC in older adults.^22,23,24 The cohort entry (index) date was the OAC dispensing date, and the 365-day period before the index date was the baseline assessment period. We further applied the following inclusion criteria: (1) 1 or more diagnosis of AF in the baseline assessment period and (2) enrollment in medical and pharmacy coverage for at least 365 days. We excluded those with the following conditions in the baseline assessment period: (1) received hospice care, (2) had other indications for OAC (venous thromboembolism or joint replacement), (3) had major bleeding within the past 30 days, and (4) had valvular heart diseases or end-stage kidney or liver disease in the previous year (definitions are provided in eMethods in Supplement 1).

Measurement of Baseline Covariates

We measured 79 covariates, including patient demographic characteristics, comorbidities, prescription drug use, and health care use, during the baseline assessment period. Because absence of code is assumed to indicate absence of condition, missing data (eg, missing a definitive diagnosis or procedure code) would have resulted in misclassification of the covariates. Therefore, we included only participants with baseline enrollment for at least 365 days before study entry to reduce this type of misclassification.²⁵ Race and ethnicity were included as covariates because they could have been potential confounders of the analysis; a missing indicator was used for those missing race and ethnicity information. We computed the CHA₂DS₂-VASc (congestive heart failure, hypertension, age ≥75 years, diabetes, stroke or transient ischemic attack, vascular disease, age 65-74 years, and sex category) score^5,6 for ischemic stroke risk and the modified HAS-BLED (hypertension, abnormal liver and/or kidney function, stroke history, bleeding history or predisposition, labile international normalized ratio, elderly, drug and alcohol use concomitantly) score for major bleeding risk.^6,26,27

We used validated algorithms with positive predictive values of 78% to 92% to define dementia.²⁸ Using data from the 365 days before cohort entry, we identified patients through International Classification of Diseases, Ninth Revision (ICD-9) and International Classification of Diseases, Tenth Revision (ICD-10) codes for dementia validated by Taylor et al,²⁹ which has also been adopted as the Chronic Conditions Data Warehouse definition by the Centers for Medicare & Medicaid Services. In a sensitivity analysis, we further subdivided patients with a dementia diagnosis into those using dementia drugs (ie, donepezil, rivastigmine, galantamine, and memantine) vs those not using dementia drugs because persons living with dementia who are receiving these medications are more likely to have more clinically severe dementia.^30,31

The burden of comorbidity was quantified using a combined comorbidity score based on 20 chronic medical conditions; this combined score has outperformed 2 widely used comorbidity scores, the Charlson Comorbidity Index and the Elixhauser Comorbidity Index, in estimating 1-year mortality.³² Frailty was measured using a claims-based frailty index (CFI) that was validated against clinical measures of frailty.^33,34,35,36 The CFI has been validated against gait speed,³³ grip strength,³³ frailty phenotype,³⁵ the deficit-accumulation frailty index,³⁵ and severe disability.³⁵ Using accepted cutoffs,^37,38,39 we defined nonfrailty as a CFI score of lower than 0.15, prefrailty as a CFI score of 0.15 to 0.24, and frailty as a CFI score of 0.25 or higher.

Outcomes and Follow-up

The primary outcome was a composite end point of ischemic stroke or major bleeding. Secondary outcomes included individual components of the primary outcome (ie, ischemic stroke and components of the major bleeding events, including gastrointestinal [GI] and other extracranial bleeding and intracranial hemorrhage [ICH]) and all-cause mortality. Because death ascertainment in the Optum and MarketScan databases was inadequate (data are only based on discharge status), the death outcome was only assessed in the Medicare population (which also contains death information from Social Security files). Clinical events were defined using validated algorithms with high performance (positive predictive values of 85%-90% for ischemic stroke^40,41 and 86%-96%⁴² for major bleeding outcomes) (definitions are available in eMethods in Supplement 1). Follow-up began 1 day after the index date and continued for a maximum of 183 days or the earliest of the following events: occurrence of the study outcome, disenrollment from insurance, or end of data availability. We used intention-to-treat analysis in the primary analysis and as-treated analysis in the sensitivity analysis. In the sensitivity analysis, study participants were also censored at the date on which they discontinued use of the index drug or switched to another OAC. To assess the robustness of findings for discontinuation of the index drug, 3 definitions of discontinuation were used based on gaps in OAC dispensing of more than 7 days, more than 14 days, and more than 30 days.

Statistical Analysis

We compared clinical characteristics between participants using apixaban vs participants using each of the other OACs (dabigatran, rivaroxaban, or warfarin). We used 1:1 propensity score (PS) matching to adjust for confounding. We estimated the PS using a logistic regression analysis that modeled the probability of initiating apixaban vs another AOC as a function of the 79 covariates, including calendar time by quarters (PS variables are shown in eTables 1-3 in Supplement 1). The PS model development and PS matching were conducted within each database and dementia subgroup (dementia or no dementia). In the matched cohort, we estimated hazard ratios (HRs) and 95% CIs using Cox proportional hazards models. We checked the proportionality assumption by inspecting the Kaplan-Meier curves; this assumption was satisfied in our analyses. We estimated rate differences (RDs) using generalized linear models, assuming Poisson distributions. We used random effects meta-analysis to pool database-specific estimates. We also conducted a meta-analysis of the dementia subgroup–specific HR and RD estimates and examined heterogeneity across subgroups.⁴³

We conducted several sensitivity analyses. First, to account for unmeasured confounding, we also used the high-dimensional propensity score (HDPS),⁴⁴ which semiautomatically identifies, prioritizes, and adjusts for thousands of codes empirically associated with the exposure and the outcome.^{44,45,46,47,48} Second, we excluded those with ischemic stroke in the 60 days before OAC initiation (to avoid misclassifying diagnoses of previous stroke as incident events during follow-up). Third, we excluded those admitted to a short-term skilled nursing facility (SNF) at baseline to avoid misclassification of OAC exposure given that Medicare Part D does not cover medications during short-term SNF stays. Fourth, to account for variations in clinical practice, we adjusted the PS model for the 50 US states. Fifth, we excluded those receiving reduced doses of a DOAC (75.0 mg of dabigatran, 15.0 mg of rivaroxaban, or 2.5 mg of apixaban) because the reasons underlying the reduced dosing may be underrecorded in the claims data. Sixth, to assess the potential impact of competing risks due to death, we assessed a composite outcome of ischemic stroke, major bleeding, and death. Analyses were conducted through the Aetion Evidence Generation Platform (Aetion) using R software, version 3.4.2 (R Foundation for Statistical Computing).^49,50,51 The significance threshold was 2-tailed P = .05.

Results

Characteristics of Study Populations

Among 1 160 462 patients with AF, the mean (SD) age was 77.4 (7.2) years; 49.8% were female, 50.2% were male, 4.3% were Black, 80.5% were White, and 61% were of other races and/or ethnicities (including Asian, Hispanic, North American Native, other, unknown, and missing). In total, 7.9% of patients had dementia and 92.1% did not. We established 3 new-user cohorts comprising patients using warfarin vs apixaban (501 990 total patients from the 3 databases; mean [SD] age, 78.1 [7.4] years; 50.2% female) (eFigure 1 in Supplement 1), dabigatran vs apixaban (126 718 total patients from the 3 databases; mean [SD] age, 76.5 [7.1] years; 48.0% female) (eFigure 2 in Supplement 1), and rivaroxaban vs apixaban (531 754 total patients from the 3 databases; mean [SD] age, 76.9 [7.2] years; 49.8% female) (eFigure 3 in Supplement 1). The pooled estimates of patient characteristics based on the 3 databases combined (after PS matching) are shown in the Table (database-specific estimates are shown in eTables 4-15 in Supplement 1). Before PS matching, compared with apixaban users, dabigatran and rivaroxaban users were younger (dabigatran vs apixaban group: mean [SD] age, 76.4 [7.1] years vs 78.1 [7.4] years; rivaroxaban vs apixaban group: mean [SD] age, 76.7 [7.1] years vs 77.8. [7.4] years). Dabigatran and rivaroxaban users were also healthier than apixaban users, as observed in lower CHA₂DS₂-VASc scores (mean [SD], 4.40 [1.69] among dabigatran users and 4.41 [1.70] among rivaroxaban users vs 4.63 [1.70] among apixaban users), lower comorbidity scores (mean [SD], 2.85 [2.58] among dabigatran users and 3.02 [2.66] among rivaroxaban users vs 3.50 [2.90] among apixaban users), and lower prevalence of comorbidities (eg, hypertension: 84.6% of dabigatran users and 84.9% of rivaroxaban users vs 87.1% of apixaban users) (eTables 8-15 in Supplement 1). In contrast, warfarin vs apixaban users had similar ages (mean [SD], 78.2 [7.3] years vs 77.8 [7.4] years), CHA₂DS₂-VASc scores (mean [SD], 4.80 [1.70] vs 4.60 [1.70]), and comorbidity scores (mean [SD], 3.50 [2.80] vs 3.50 [2.90]) (eTables 4-7 in Supplement 1). After PS matching, all characteristics were adequately balanced between exposure groups in all 3 cohorts (Table) and within dementia status (eTables 16-27 in Supplement 1). Compared with patients without dementia, those living with dementia were older, more likely to be female, and more likely to have higher CFI scores, higher comorbidity scores, higher CHA₂DS₂-VASc scores, higher HAS-BLED scores, and greater health care use (eTables 16-27 in Supplement 1).

Table. Participant Characteristics^a.

Characteristic	Participants, No. (%)
	Warfarin cohort		Dabigatran cohort		Rivaroxaban cohort
	Apixaban (n = 250 995)	Warfarin (n = 250 995)	Apixaban (n = 63 359)	Dabigatran (n = 63 359)	Apixaban (n = 265 877)	Rivaroxaban (n = 265 877)
Age, mean (SD), y	78.1 (7.4)	78.1 (7.3)	76.5 (7.1)	76.6 (7.1)	76.8 (7.2)	76.9 (7.2)
Sex
Female	125 863 (50.1)	126 130 (50.3)	30 448 (48.1)	30 261 (47.8)	132 320 (49.8)	132 423 (49.8)
Male	125 132 (49.9)	124 865 (49.7)	32 911 (51.9)	33 098 (52.2)	133 557 (50.2)	133 454 (50.2)
Race and ethnicity
Black	10 827 (4.7)	10 818 (4.7)	2487 (4.4)	2428 (4.3)	11 828 (4.9)	11 853 (4.9)
White	206 224 (89.5)	206 187 (89.4)	50 350 (88.8)	50 444 (89.0)	210 296 (87.5)	210 296 (87.5)
Other^b	13 487 (5.9)	13 533 (5.9)	3834 (6.8)	3799 (6.7)	18 084 (7.5)	18 059 (7.5)
Dual insurance status^c	11 430 (5.9)	11 457 (5.9)	2633 (5.3)	2583 (5.2)	10 882 (5.6)	10 704 (5.6)
CHA₂DS₂-VASc score, mean (SD)	4.71 (1.70)	4.71 (1.66)	4.42 (1.69)	4.41 (1.70)	4.43 (1.69)	4.42 (1.70)
HAS-BLED score, mean (SD)	2.30 (0.70)	2.30 (0.74)	2.22 (0.72)	2.22 (0.70)	2.23 (0.71)	2.23 (0.70)
CCI score, mean (SD)	3.48 (2.90)	3.48 (2.83)	2.90 (2.60)	2.88 (2.60)	3.11 (2.71)	3.08 (2.70)
CFI score, mean (SD)	0.21 (0.10)	0.21 (0.07)	0.20 (0.07)	0.20 (0.10)	0.20 (0.07)	0.20 (0.10)
Cardiovascular condition
Acute myocardial infarction	17 199 (6.9)	17 277 (6.9)	2898 (4.6)	2866 (4.5)	14 572 (5.5)	14 427 (5.4)
Cardioablation	1286 (0.5)	1260 (0.5)	411 (0.6)	403 (0.6)	1708 (0.6)	1699 (0.6)
Cardioversion	10 679 (4.3)	10 414 (4.1)	3998 (6.3)	4021 (6.3)	17 755 (6.7)	17 691 (6.7)
Cerebrovascular disease	70 312 (28.0)	70 459 (28.1)	16 593 (26.2)	16 594 (26.2)	65 070 (24.5)	64 393 (24.2)
Congestive heart failure
Inpatient	51 717 (20.6)	51 651 (20.6)	9880 (15.6)	9577 (15.1)	45 065 (16.9)	44 516 (16.7)
Outpatient	86 644 (34.5)	86 440 (34.4)	18 812 (29.7)	18 789 (29.7)	77 540 (29.2)	77 249 (29.1)
Coronary revascularization	11 564 (4.6)	11 485 (4.6)	1622 (2.6)	1589 (2.5)	7461 (2.8)	7433 (2.8)
Hypertension	215 906 (86.0)	215 775 (86.0)	53 643 (84.7)	53 642 (84.7)	226 620 (85.2)	226 481 (85.2)
Ischemic heart disease	117 173 (46.7)	116 858 (46.6)	27 410 (43.3)	27 616 (43.6)	113 388 (42.6)	113 118 (42.5)
PVD or PVD surgical procedure	37 345 (14.9)	37 351 (14.9)	8182 (12.9)	8111 (12.8)	35 389 (13.3)	35 243 (13.3)
Stroke
Inpatient	22 655 (9.0)	22 863 (9.1)	4879 (7.7)	4803 (7.6)	18 749 (7.1)	18 282 (6.9)
Outpatient	31 251 (12.5)	31 324 (12.5)	6890 (10.9)	6915 (10.9)	29 610 (11.1)	29 254 (11.0)
Syncope	24 758 (9.9)	24 920 (9.9)	6197 (9.8)	6290 (9.9)	27 617 (10.4)	27 595 (10.4)
Noncardiovascular condition
Acute kidney failure	36 212 (14.4)	36 240 (14.4)	5743 (9.1)	5665 (8.9)	28 822 (10.8)	28 373 (10.7)
Alcohol misuse or dependence	4783 (1.9)	4799 (1.9)	1179 (1.9)	1196 (1.9)	5635 (2.1)	5642 (2.1)
Anemia	72 201 (28.8)	72 184 (28.8)	15 198 (24.0)	15 251 (24.1)	65 428 (24.6)	64 976 (24.4)
CKD stage 3, 4, or unspecified	47 757 (19.0)	47 525 (18.9)	7757 (12.2)	7676 (12.1)	37 598 (14.1)	37 044 (13.9)
COPD	57 603 (22.9)	57 315 (22.8)	12 994 (20.5)	12 753 (20.1)	56 712 (21.3)	56 494 (21.2)
Dementia	21 079 (8.4)	21 190 (8.4)	4338 (6.8)	4239 (6.7)	20 229 (7.6)	20 073 (7.5)
Diabetes	94 441 (37.6)	94 494 (37.6)	22 987 (36.3)	23 017 (36.3)	93 060 (35.0)	93 275 (35.1)
Endoscopy	5630 (2.2)	5754 (2.3)	1057 (1.7)	1069 (1.7)	4505 (1.7)	4486 (1.7)
Falls	15 256 (6.1)	15 324 (6.1)	2830 (4.5)	2836 (4.5)	14 772 (5.6)	14 569 (5.5)
Fractures	22 711 (9.0)	22 817 (9.1)	4946 (7.8)	5016 (7.9)	22 651 (8.5)	22 484 (8.5)
GI bleeding
Inpatient	14 792 (5.9)	14 750 (5.9)	2923 (4.6)	2848 (4.5)	11 940 (4.5)	11 913 (4.5)
Outpatient	33 501 (13.3)	33 629 (13.4)	8001 (12.6)	8082 (12.8)	33 578 (12.6)	33 615 (12.6)
Liver disease	15 573 (6.2)	15 508 (6.2)	3738 (5.9)	3644 (5.8)	16 460 (6.2)	16 377 (6.2)
Cancer	45 657 (18.2)	45 764 (18.2)	11 225 (17.7)	11 228 (17.7)	47 014 (17.7)	46 769 (17.6)
Obesity	48 879 (19.5)	48 953 (19.5)	12 222 (19.3)	12 129 (19.1)	53 387 (20.1)	53 565 (20.1)
Peptic ulcer	5622 (2.2)	5637 (2.2)	1077 (1.7)	1087 (1.7)	5312 (2.0)	5214 (2.0)
Ever smoked	77 467 (30.9)	77 226 (30.8)	17 619 (27.8)	17 632 (27.8)	81537 (30.7)	81 186 (30.5)
Cardiovascular medication
ACE inhibitor	72 535 (28.9)	72 450 (28.9)	16 933 (26.7)	16 840 (26.6)	71 166 (26.8)	71 131 (26.8)
Angiotensin 2 receptor blocker	13 661 (5.4)	13 623 (5.4)	4085 (6.4)	4141 (6.5)	15 985 (6.0)	15 891 (6.0)
Antiarrhythmic agent	47 005 (18.7)	46 726 (18.6)	13 731 (21.7)	13 684 (21.6)	51 923 (19.5)	51 877 (19.5)
Injectable anticoagulant	2673 (1.1)	2733 (1.1)	465 (0.7)	467 (0.7)	1836 (0.7)	1857 (0.7)
Antiplatelet agent	37 072 (14.8)	36 842 (14.7)	8726 (13.8)	8863 (14.0)	39 437 (14.8)	39 282 (14.8)
β-blocker	171 519 (68.3)	171 238 (68.2)	41 245 (65.1)	41 157 (65.0)	175 911 (66.2)	175 740 (66.1)
Calcium channel blocker	4418 (1.8)	4425 (1.8)	1073 (1.7)	1049 (1.7)	4273 (1.6)	4294 (1.6)
Diuretic	143 919 (57.3)	143 827 (57.3)	33 426 (52.8)	33 405 (52.7)	138 452 (52.1)	138 451 (52.1)
Fibrate	11 428 (4.6)	11 430 (4.6)	2930 (4.6)	2935 (4.6)	11 527 (4.3)	11 445 (4.3)
Nitrate	33 127 (13.2)	32 813 (13.1)	7245 (11.4)	7375 (11.6)	29 109 (10.9)	28 907 (10.9)
Statin	164 076 (65.4)	164 017 (65.3)	40 400 (63.8)	40 472 (63.9)	168 276 (63.3)	168 050 (63.2)
Other medication
Anticonvulsant	41 224 (16.4)	41 414 (16.5)	9738 (15.4)	9800 (15.5)	42 094 (15.8)	42 098 (15.8)
Antidepressant
SSRI or SNRI	50 898 (20.3)	50 832 (20.3)	12 783 (20.2)	12 915 (20.4)	53 510 (20.1)	53 296 (20.0)
Tricyclic	6709 (2.7)	6750 (2.7)	1627 (2.6)	1644 (2.6)	6952 (2.6)	6968 (2.6)
Other	21 171 (8.4)	21 200 (8.4)	4951 (7.8)	4878 (7.7)	22 242 (8.4)	22 214 (8.4)
Antipsychotic agent	7120 (2.8)	7289 (2.9)	1791 (2.8)	1783 (2.8)	8012 (3.0)	7930 (3.0)
Anxiolytic (except benzodiazepine)	2819 (1.1)	2843 (1.1)	699 (1.1)	685 (1.1)	3295 (1.2)	3400 (1.3)
Anxiolytic (benzodiazepine)	43 070 (17.2)	42 997 (17.1)	11 225 (17.7)	11 345 (17.9)	47 560 (17.9)	47 574 (17.9)
Bronchodilator	47 843 (19.1)	47 725 (19.0)	11 914 (18.8)	11 917 (18.8)	51 917 (19.5)	51 838 (19.5)
Corticosteroid
Inhaled	52 557 (20.9)	52 626 (21.0)	13 923 (22.0)	14 178 (22.4)	59 412 (22.3)	59 287 (22.3)
Oral	77 799 (31.0)	77 813 (31.0)	19 970 (31.5)	19 909 (31.4)	85 193 (32.0)	85 226 (32.1)
Dementia medication	12 672 (5.0)	12 798 (5.1)	3152 (5.0)	3086 (4.9)	13 218 (5.0)	13 082 (4.9)
Diabetes medication
Insulin	20 364 (8.1)	20 365 (8.1)	4616 (7.3)	4569 (7.2)	18 254 (6.9)	18 217 (6.9)
Metformin	40 724 (16.2)	40 688 (16.2)	10 687 (16.9)	10 686 (16.9)	43 459 (16.3)	43 851 (16.5)
Other	13 439 (5.4)	13 486 (5.4)	3576 (5.6)	3481 (5.5)	14 335 (5.4)	14 327 (5.4)
Sulfonylurea	25 654 (10.2)	25 591 (10.2)	6071 (9.6)	6049 (9.5)	23 584 (8.9)	23 533 (8.9)
Estrogen	6624 (2.6)	6730 (2.7)	2086 (3.3)	2036 (3.2)	8580 (3.2)	8536 (3.2)
GI
Histamine type 2 receptor blocker	19 015 (7.6)	19 093 (7.6)	4499 (7.1)	4543 (7.2)	19 827 (7.5)	19 636 (7.4)
Proton pump inhibitor	80 061 (31.9)	79 915 (31.8)	19 592 (30.9)	19 832 (31.3)	83 026 (31.2)	82 760 (31.1)
Sucralfate	4669 (1.9)	4588 (1.8)	1203 (1.9)	1229 (1.9)	5027 (1.9)	4954 (1.9)
Hypnotic	19 257 (7.7)	19 167 (7.6)	5563 (8.8)	5552 (8.8)	22 039 (8.3)	22 182 (8.3)
NSAID	34 955 (13.9)	34 885 (13.9)	10 330 (16.3)	10 263 (16.2)	45 815 (17.2)	45 816 (17.2)
Opioid	96 939 (38.6)	96 727 (38.5)	24 184 (38.2)	24 135 (38.1)	99 554 (37.4)	99 953 (37.6)
Parkinsonism medication	9566 (3.8)	9687 (3.9)	2443 (3.9)	2444 (3.9)	9900 (3.7)	9921 (3.7)
Thyroid hormone replacement	56 383 (22.5)	56 525 (22.5)	13 489 (21.3)	13 353 (21.1)	56 783 (21.4)	56 666 (21.3)
Health care use
ED visit	104 777 (41.7)	105 108 (41.9)	24 831 (39.2)	24 792 (39.1)	113 964 (42.9)	113 434 (42.7)
Home health services (in days)	7119 (3.7)	7239 (3.7)	2738 (5.5)	2720 (5.5)	6756 (3.5)	6829 (3.5)
Home oxygen use	8294 (3.3)	8362 (3.3)	1708 (2.7)	1653 (2.6)	8029 (3.0)	7947 (3.0)
Hospitalization	123 433 (49.2)	123 123 (49.1)	27 250 (43.0)	26 701 (42.1)	123 353 (46.4)	122 468 (46.1)
Geographic region
Northeast	50 722 (20.2)	51 118 (20.4)	13 778 (21.7)	13 909 (22.0)	50 882 (19.1)	50 711 (19.1)
Midwest	66 024 (26.3)	65 599 (26.1)	13 688 (21.6)	13 398 (21.1)	60 070 (22.6)	60 053 (22.6)
South	87 165 (34.7)	87 114 (34.7)	24 577 (38.8)	24 779 (39.1)	103 420 (38.9)	103 250 (38.8)
West	46 822 (18.7)	46 892 (18.7)	11 236 (17.7)	11 190 (17.7)	51 227 (19.3)	51 567 (19.4)
Other	262 (0.1)	272 (0.1)	80 (0.1)	83 (0.1)	278 (0.1)	296 (0.1)

Open in a new tab

Abbreviations: ACE, angiotensin-converting enzyme; CCI, combined comorbidity index; CFI, claims-based frailty index; CHA₂DS₂-VASc, congestive heart failure, hypertension, age ≥75 years, diabetes, stroke or transient ischemic attack, vascular disease, age 65-74 years, and sex category; CKD, chronic kidney disease; COPD, chronic obstructive pulmonary disease; ED, emergency department; GI, gastrointestinal; HAS-BLED, hypertension, abnormal liver and/or kidney function, stroke history, bleeding history or predisposition, labile international normalized ratio, elderly, drug and alcohol use concomitantly; NSAID, nonsteroidal anti-inflammatory drug; PVD, peripheral vascular disease; SNRI, serotonin-norepinephrine reuptake inhibitor; SSRI, selective serotonin reuptake inhibitor.

^{^a}

Among a propensity score–matched population with atrial fibrillation who received treatment with oral anticoagulants, pooled across Medicare, Optum Clinformatics Data Mart, and IBM MarketScan Research Database populations.

^{^b}

Other races and ethnicities include Asian, Hispanic, North American Native, other, unknown, and missing categories in the Medicare database; Asian, Hispanic, unknown, and missing categories in the Optum Clinformatics Data Mart; and unavailable category in the IBM MarketScan Research Database.

^{^c}

With both Medicare and Medicaid enrollment eligibility.

Clinical Outcomes Associated With Warfarin vs Apixaban by Dementia Status

Over a mean (SD) follow-up of 165.8 (43.0) days, the pooled rate of the composite end point per 1000 person-years (PYs) from the 3 databases combined was 53.2 events for warfarin users and 38.2 events for apixaban users (adjusted HR [aHR], 1.4 [95% CI, 1.4-1.5]; RD per 1000 PYs, 16.1 [95% CI, 12.3-20.0] events). Compared with apixaban users, warfarin users had higher rates of major bleeding (aHR, 1.6 [95% CI, 1.5-1.7]; RD per 1000 PYs, 13.5 [95% CI, 9.8-17.1] events), ischemic stroke (aHR, 1.3 [95% CI, 1.2-1.4]; RD per 1000 PYs, 3.3 [95% CI, 2.3-4.2] events), and death (aHR, 1.1 [95% CI, 1.1-1.1]; RD per 1000 PYs, 10.9 [95% CI, 7.6-14.1] events). The magnitude of the benefit associated with apixaban was consistent across dementia status on the HR scale (1.5 [95% CI, 1.3-1.7] for those with dementia vs 1.5 [95% CI, 1.4-1.5] for those without dementia; P = .92 for heterogeneity) but greater for those with vs without dementia on the RD scale (RD per 1000 PYs, 29.8 [95% CI, 18.4-41.1] events vs 16.0 [95% CI, 13.6-18.4] events; P = .02 for heterogeneity). The treatment effect heterogeneity in RD by dementia groups was evident for death and major bleeding, including both intracranial and extracranial bleeding, but not for ischemic stroke (pooled results are shown in Figure 1 and eTable 28 in Supplement 1, and database-specific estimates are shown in eTables 29-31 in Supplement 1).

Figure 1. — Among a propensity score–matched US Medicare population with atrial fibrillation pooled across Medicare, Optum Clinformatics Data Mart, and IBM MarketScan Research Database populations. The estimates are pooled results across the 3 databases and were calculated using a random effects meta-analysis of the weighted mean of database-specific estimates. Mortality results were based on the Medicare population due to incomplete death ascertainment in the other 2 databases. IR indicates incidence rate.

Clinical Outcomes Associated With Dabigatran vs Apixaban by Dementia Status

Over a mean (SD) follow-up of 169.5 (38.5) days, the pooled rate of the composite end point per 1000 PYs was 37.5 events for dabigatran users and 33.1 events for apixaban users (aHR, 1.2 [95% CI, 1.1-1.3]; RD per 1000 PYs, 6.3 [95% CI, 2.6-10.0] events). Compared with apixaban users, dabigatran users had higher rates of ischemic stroke (aHR, 1.2 [95% CI, 1.0-1.4]; RD per 1000 PYs, 1.8 [95% CI, 0.1-3.5] events). The magnitude of the benefit associated with apixaban was consistent across dementia subgroups on the HR scale (1.5 [95% CI, 1.2-2.0] for those with dementia vs 1.2 [95% CI, 1.1-1.4] for those without dementia; P = .08 for heterogeneity) but greater on the RD scale for those with vs without dementia (29.6 [95% CI, 11.6-47.6] events per 1000 PYs vs 5.8 [95% CI, 1.1-10.4] events per 1000 PYs; P = .001 for heterogeneity). The hemorrhagic risk reduction associated with apixaban was particularly distinct for GI bleeding but not evident for ICH, a pattern consistent across dementia subgroups (pooled results are shown in Figure 2 and eTable 32 in Supplement 1, and database-specific estimates are shown in eTables 33-35 in Supplement 1).

Figure 2. — Among a propensity score–matched US Medicare population with atrial fibrillation pooled across Medicare, Optum Clinformatics Data Mart, and IBM MarketScan Research Database populations. The estimates are pooled results across the 3 databases and were calculated using a random effects meta-analysis of the weighted mean of database-specific estimates. Mortality results were based on the Medicare population due to incomplete death ascertainment in the other 2 databases. IR indicates incidence rate.

Clinical Outcomes Associated With Rivaroxaban vs Apixaban by Dementia Status

Over a mean (SD) follow-up of 164.0 (44.7) days, the rate of the composite end point per 1000 PYs was 47.5 events for rivaroxaban users and 33.0 events for apixaban users (aHR, 1.5 [95% CI, 1.3-1.6]; RD per 1000 PYs, 15.1 [95% CI, 10.0-20.1] events). Compared with apixaban users, rivaroxaban users had higher rates of major bleeding (aHR, 1.7 [95% CI, 1.5-1.8]; RD per 1000 PYs, 13.2 [95% CI, 7.9-18.5] events), ischemic stroke (aHR, 1.2 [95% CI, 1.1-1.3]; RD per 1000 PYs, 1.8 [95% CI, 1.0-2.6] events), and death (aHR, 1.1 [95% CI, 1.1-1.2]; RD per 1000 PYs, 10.8 [95% CI, 7.8-13.8] events). The absolute mortality reduction associated with apixaban was greater in persons living with dementia vs those without dementia (RD per 1000 PYs, 44.4 [95% CI, 24.1-64.6] events vs 6.3 [95% CI, 3.6-9.1] events; P < .001 for heterogeneity). The risk reduction in major bleeding associated with apixaban was particularly distinct for GI bleeding but not evident for ICH, a pattern consistent across dementia subgroups (pooled results are shown in Figure 3 and eTable 36 in Supplement 1, and database-specific estimates are shown in eTables 37-39 in Supplement 1).

Figure 3. — Among a propensity score–matched US Medicare population with atrial fibrillation pooled across Medicare, Optum Clinformatics Data Mart, and IBM MarketScan Research Database populations. The estimates are pooled results across the 3 databases and were calculated using a random effects meta-analysis of the weighted mean of database-specific estimates. Mortality results were based on the Medicare population due to incomplete death ascertainment in the other 2 databases. IR indicates incidence rate.

Sensitivity Analyses

In all of the prespecified sensitivity analyses, we observed a consistent pattern; apixaban users had lower rates of the composite outcomes than users of other OACs, which was primarily due to fewer major bleeding outcomes. The magnitude of the benefit associated with apixaban was consistent across dementia status on the HR scale but substantially greater on the RD scale in those with vs without dementia. This pattern was observed when patients with dementia were further subdivided by use vs nonuse of dementia medications (eFigures 4-6 in Supplement 1), when death was included in the composite outcome, when HDPS was used for confounding adjustment, when as-treated analyses were conducted using various gaps between prescriptions to define discontinuation, when those with recent stroke were excluded, when those with a short-term SNF stay at baseline were excluded, when those receiving a reduced dose of a DOAC were excluded, and when the model was adjusted for US states (eTable 40 in Supplement 1).

Discussion

In this comparative effectiveness study of adults 65 years or older with AF who were identified in 3 US nationwide health insurance databases, we observed that apixaban users had lower rates of ischemic strokes and intracranial and extracranial major bleeding compared with users of other OACs in the general population. The magnitude of the benefit associated with apixaban was similar regardless of dementia status on the relative scale but greater in terms of absolute rate reduction in patients with vs without dementia when compared with specific OACs for certain outcomes (particularly warfarin in relation to major bleeding). These increases in the absolute benefits of apixaban among patients with dementia reflect the substantially higher rates of poor outcomes in these patients.

Routine-care database studies of anticoagulation therapy among patients with AF using Medicare^{52,53,54,55,56} and commercial insurance claims^{56,57,58,59,60,61} have reported results that are generally consistent with those of RCTs.^15,16,17 However, there are limited data comparing the benefits of OACs among persons living with dementia. None of the RCTs evaluated subgroup effects among persons living with dementia. One small observational study⁶² based on 2399 patients in the UK reported that DOAC use was associated with 2-fold increases in the risk of GI bleeding and overall death compared with warfarin. In this UK study,⁶² DOAC users were noted to be substantially older with more severe illness at baseline, and residual confounding was acknowledged as a possible explanation for the findings. Given the limited sample size, the previous study⁶² was not able to compare clinical outcomes by specific DOACs, which is suboptimal because patient characteristics and safety and effectiveness outcome incidence were all substantially different across DOACs in our analysis. We found that dabigatran and rivaroxaban users were younger and had a smaller comorbidity burden compared with apixaban users, who had baseline demographic and comorbidity profiles similar to warfarin users. In contrast to results from the UK study,⁶² our findings suggest apixaban is associated with lower major bleeding risks compared with all other OACs, which is consistent with findings from an RCT¹⁷ and other routine care database analyses^{52,53,54,55,56,57,58,59,60,61} conducted in general populations.

Examining the large number of major bleeding events associated with other OACs vs apixaban among persons living with dementia, the majority of major bleeding events involved extracranial bleeding, although intracranial bleeding is associated with a substantially higher risk (or incidence) of death and disability.⁶³ It has been postulated that persons living with dementia could have an increased risk of cerebral amyloid angiopathy–mediated intracerebral hemorrhage.^64,65 Yet, the ways in which different OACs may interact with this type of vulnerability is unclear. Among persons living with dementia, compared with apixaban users, warfarin users had a higher risk of ICH, but this higher risk was not observed in dabigatran or rivaroxaban users. In contrast, the risk reduction in major bleeding associated with apixaban was observed when compared with both warfarin and other DOACs (eTables 28-39 in Supplement 1). This pattern was similar across dementia subgroups.

Our results among persons living with dementia have provided important data about this vulnerable population of patients who have been substantially underrepresented in clinical trials. The association of OACs with significantly higher risks of major bleeding and ischemic stroke on the RD scale was proportional to the absolute event rates across dementia subgroups. Despite a substantially higher baseline rate of all outcomes among persons living with dementia, apixaban use was associated with a similar relative rate reduction regardless of dementia diagnosis, yielding a substantially higher absolute rate reduction compared with the other OACs. The fact that the absolute net clinical benefits associated with apixaban in comparison with other OACs were greater among persons living with dementia than those without dementia supports the use of apixaban in highly vulnerable patients with frailty.

Limitations

This study has several strengths, including the use of large nationally representative databases of older adults in the US to achieve sufficient power that enabled detailed subgroup analysis by dementia status. We implemented a rigorous analytical framework with a new-user active-comparator design, which has been reported to successfully replicate RCT findings.⁶⁶ In addition, we conducted a variety of sensitivity analyses and observed consistent results, which suggests our findings are robust.

This study also has limitations. First, the dementia subgroups could have been misclassified. We conducted a sensitivity analysis assessing the treatment effects of OACs among persons living with dementia who were using dementia medications and found consistent results. In addition, because this study was based on claims data, we were unable to measure clinical variables (eg, laboratory results) that may have implications for the effectiveness and safety of OACs or may confound associations. Therefore, unmeasured confounding remains possible. To mitigate unmeasured confounding, we applied a validated proxy confounding adjustment algorithm, HDPS,⁴⁴ and found similar results. We also applied a validated CFI to account for frailty in our analysis. Another possible source of bias is misclassification of the OAC exposure during follow-up. Using a fixed gap between prescriptions to define treatment discontinuation can produce differential misclassification of the drug exposure given that the prescribing routine may be different for DOACs vs warfarin.⁶⁷ Therefore, we conducted as-treated analyses varying the allowable gaps between prescriptions as sensitivity analyses, which yielded consistent results.

Conclusions

In this comparative effectiveness study of OAC use among 3 US nationwide cohorts of older adults with AF, apixaban was consistently associated with lower rates of major bleeding and ischemic stroke compared with other OACs. The substantial risks associated with other OACs compared with apixaban were greater among patients with dementia than those without dementia on the absolute scale, particularly for major bleeding events. These results support the use of apixaban for anticoagulation therapy in vulnerable patients with frailty, particularly those living with dementia.

Supplement 1.

eMethods. Definitions of Exposures, Outcomes, Inclusions and Exclusions, Covariates, and Comorbidity Scores

eFigure 1. Selection of Study Population for the Warfarin vs Apixaban Cohort Pooled Across Medicare, Optum, and MarketScan Populations

eFigure 2. Selection of Study Population for the Dabigatran vs Apixaban Cohort Pooled Across Medicare, Optum, and MarketScan Populations

eFigure 3. Selection of Study Population for the Rivaroxaban vs Apixaban Cohort Pooled Across Medicare, Optum, and MarketScan Populations

eFigure 4. Sensitivity Analyses for Warfarin vs Apixaban in Older Adults With Atrial Fibrillation in the Medicare Population

eFigure 5. Sensitivity Analyses for Dabigatran vs Apixaban in Older Adults With Atrial Fibrillation In the Medicare Population

eFigure 6. Sensitivity Analyses for Rivaroxaban vs Apixaban in Older Adults With Atrial Fibrillation In the Medicare Population

eTable 1. Propensity Score Models for Each Oral Anticoagulant vs Apixaban in the Medicare Population

eTable 2. Propensity Score Models for Each Oral Anticoagulant vs Apixaban in the Optum Population

eTable 3. Propensity Score Models for Each Oral Anticoagulant vs Apixaban in the MarketScan Population

eTable 4. Study Population With Atrial Fibrillation Treated With Warfarin vs Apixaban Before and After 1:1 Propensity Score Matching Pooled Across Medicare, Optum, and MarketScan Populations

eTable 5. Study Population With Atrial Fibrillation Treated With Warfarin vs Apixaban Before and After 1:1 Propensity Score Matching in the Medicare Population

eTable 6. Study Population With Atrial Fibrillation Treated With Warfarin vs Apixaban Before and After 1:1 Propensity Score Matching in the Optum Population

eTable 7. Study Population With Atrial Fibrillation Treated With Warfarin vs Apixaban Before and After 1:1 Propensity Score Matching in the MarketScan Population

eTable 8. Study Population With Atrial Fibrillation Treated With Dabigatran vs Apixaban Before and After 1:1 Propensity Score Matching Pooled Across Medicare, Optum, and MarketScan Populations

eTable 9. Study Population With Atrial Fibrillation Treated With Dabigatran vs Apixaban Before and After 1:1 Propensity Score Matching in the Medicare Population

eTable 10. Study Population With Atrial Fibrillation Treated With Dabigatran vs Apixaban Before and After 1:1 Propensity Score Matching in the Optum Population

eTable 11. Study Population With Atrial Fibrillation Treated With Dabigatran vs Apixaban Before and After 1:1 Propensity Score Matching in the MarketScan Population

eTable 12. Study Population With Atrial Fibrillation Treated With Rivaroxaban vs Apixaban Before and After 1:1 Propensity Score Matching Pooled Across Medicare, Optum, and MarketScan Populations

eTable 13. Study Population With Atrial Fibrillation Treated With Rivaroxaban vs Apixaban Before and After 1:1 Propensity Score Matching in the Medicare Population

eTable 14. Study Population With Atrial Fibrillation Treated With Rivaroxaban vs Apixaban Before and After 1:1 Propensity Score Matching in the Optum Population

eTable 15. Study Population With Atrial Fibrillation Treated With Rivaroxaban vs Apixaban Before and After 1:1 Propensity Score Matching in the MarketScan Population

eTable 16. Selected Characteristics of Propensity Score-Matched Population With Atrial Fibrillation Treated With Warfarin vs Apixaban by Dementia Diagnosis Pooled Across Medicare, Optum, and MarketScan Populations

eTable 17. Selected Characteristics of Propensity Score–Matched Population With Atrial Fibrillation Treated With Warfarin vs Apixaban by Dementia Diagnosis in the Medicare Population

eTable 18. Selected Characteristics of Propensity Score–Matched Population With Atrial Fibrillation Treated With Warfarin vs Apixaban by Dementia Diagnosis in the Optum Population

eTable 19. Selected Characteristics of Propensity Score–Matched Population With Atrial Fibrillation Treated With Warfarin vs Apixaban by Dementia Diagnosis in the MarketScan Population

eTable 20. Selected Characteristics of Propensity Score–Matched Population With Atrial Fibrillation Treated With Dabigatran vs Apixaban by Dementia Diagnosis Pooled Across Medicare, Optum, and MarketScan Populations

eTable 21. Selected Characteristics of Propensity Score–Matched Population With Atrial Fibrillation Treated With Dabigatran vs Apixaban by Dementia Diagnosis in the Medicare Population

eTable 22. Selected Characteristics of Propensity Score–Matched Population With Atrial Fibrillation Treated With Dabigatran vs Apixaban by Dementia Diagnosis in the Optum Population

eTable 23. Selected Characteristics of Propensity Score–Matched Population With Atrial Fibrillation Treated With Dabigatran vs Apixaban by Dementia Diagnosis in the MarketScan Population

eTable 24. Selected Characteristics of Propensity Score–Matched Population With Atrial Fibrillation Treated With Rivaroxaban vs Apixaban by Dementia Diagnosis Pooled Across Medicare, Optum, and MarketScan Populations

eTable 25. Selected Characteristics of Propensity Score–Matched Population With Atrial Fibrillation Treated With Rivaroxaban vs Apixaban by Dementia Diagnosis in the Medicare Population

eTable 26. Selected Characteristics of Propensity Score–Matched Population With Atrial Fibrillation Treated With Rivaroxaban vs Apixaban by Dementia Diagnosis in the Optum Population

eTable 27. Selected Characteristics of Propensity Score–Matched Population With Atrial Fibrillation Treated With Rivaroxaban vs Apixaban by Dementia Diagnosis in the MarketScan Population

eTable 28. Dementia and Adverse Outcomes in the US Medicare Population With Atrial Fibrillation Treated With Warfarin vs Apixaban After 1:1 Propensity Score Matching Pooled Across Medicare, Optum, and MarketScan Populations

eTable 29. Dementia and Adverse Outcomes in the US Medicare Population With Atrial Fibrillation Treated With Warfarin vs Apixaban After 1:1 Propensity Score Matching in the Medicare Population

eTable 30. Dementia and Adverse Outcomes in the US Medicare Population With Atrial Fibrillation Treated With Warfarin vs Apixaban After 1:1 Propensity Score Matching in the Optum Population

eTable 31. Dementia and Adverse Outcomes in the US Medicare Population With Atrial Fibrillation Treated With Warfarin vs Apixaban After 1:1 Propensity Score Matching in the MarketScan Population

eTable 32. Dementia and Adverse Outcomes in the US Medicare Population With Atrial Fibrillation Treated With Warfarin vs Dabigatran After 1:1 Propensity Score Matching Pooled Across Medicare, Optum, and MarketScan Population

eTable 33. Dementia and Adverse Outcomes in the US Medicare Population With Atrial Fibrillation Treated With Dabigatran vs Apixaban After 1:1 Propensity Score Matching in the Medicare Population

eTable 34. Dementia and Adverse Outcomes in the US Medicare Population With Atrial Fibrillation Treated With Dabigatran vs Apixaban After 1:1 Propensity Score Matching in the Optum Population

eTable 35. Dementia and Adverse Outcomes in the US Medicare Population With Atrial Fibrillation Treated With Dabigatran vs Apixaban After 1:1 Propensity Score Matching in the MarketScan Population

eTable 36. Dementia and Adverse Outcomes in the US Medicare Population With Atrial Fibrillation Treated With Rivaroxaban vs Dabigatran After 1:1 Propensity Score Matching Pooled Across Medicare, Optum, and MarketScan Population

eTable 37. Dementia and Adverse Outcomes in the US Medicare Population With Atrial Fibrillation Treated With Rivaroxaban vs Apixaban After 1:1 Propensity Score Matching in the Medicare Population

eTable 38. Dementia and Adverse Outcomes in the US Medicare Population With Atrial Fibrillation Treated With Rivaroxaban vs Apixaban After 1:1 Propensity Score Matching in the Optum Population

eTable 39. Dementia and Adverse Outcomes in the US Medicare Population With Atrial Fibrillation Treated With Rivaroxaban vs Apixaban After 1:1 Propensity Score Matching in the MarketScan Population

eTable 40. Sensitivity Analyses for the Primary Outcome in Older Adults With Atrial Fibrillation in the Medicare Population

Click here for additional data file.^{(4MB, pdf)}

Supplement 2.

Data Sharing Statement

Click here for additional data file.^{(13.6KB, pdf)}

References

1.Nantsupawat T, Nugent K, Phrommintikul A. Atrial fibrillation in the elderly. Drugs Aging. 2013;30(8):593-601. doi: 10.1007/s40266-013-0094-8 [DOI] [PubMed] [Google Scholar]
2.Wolf PA, Abbott RD, Kannel WB. Atrial fibrillation as an independent risk factor for stroke: the Framingham Study. Stroke. 1991;22(8):983-988. doi: 10.1161/01.STR.22.8.983 [DOI] [PubMed] [Google Scholar]
3.January CT, Wann LS, Alpert JS, et al. ; ACC/AHA Task Force Members . 2014 AHA/ACC/HRS guideline for the management of patients with atrial fibrillation: executive summary: a report of the American College of Cardiology/American Heart Association Task Force on Practice Guidelines and the Heart Rhythm Society. Circulation. 2014;130(23):2071-2104. doi: 10.1161/CIR.0000000000000040 [DOI] [PubMed] [Google Scholar]
4.Kirchhof P, Benussi S, Kotecha D, et al. 2016 ESC guidelines for the management of atrial fibrillation developed in collaboration with EACTS. Europace. 2016;18(11):1609-1678. doi: 10.1093/europace/euw295 [DOI] [PubMed] [Google Scholar]
5.Lip GYH, Nieuwlaat R, Pisters R, Lane DA, Crijns HJGM. 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. doi: 10.1378/chest.09-1584 [DOI] [PubMed] [Google Scholar]
6.Friberg L, Rosenqvist M, Lip GYH. Evaluation of risk stratification schemes for ischaemic stroke and bleeding in 182 678 patients with atrial fibrillation: the Swedish Atrial Fibrillation cohort study. Eur Heart J. 2012;33(12):1500-1510. doi: 10.1093/eurheartj/ehr488 [DOI] [PubMed] [Google Scholar]
7.Singer DE, Chang Y, Fang MC, et al. The net clinical benefit of warfarin anticoagulation in atrial fibrillation. Ann Intern Med. 2009;151(5):297-305. doi: 10.7326/0003-4819-151-5-200909010-00003 [DOI] [PMC free article] [PubMed] [Google Scholar]
8.Banerjee A, Lane DA, Torp-Pedersen C, Lip GYH. Net clinical benefit of new oral anticoagulants (dabigatran, rivaroxaban, apixaban) versus no treatment in a ‘real world’ atrial fibrillation population: a modelling analysis based on a nationwide cohort study. Thromb Haemost. 2012;107(3):584-589. doi: 10.1160/TH11-11-0784 [DOI] [PubMed] [Google Scholar]
9.Kim DH, Pawar A, Gagne JJ, et al. Frailty and clinical outcomes of direct oral anticoagulants versus warfarin in older adults with atrial fibrillation: a cohort study. Ann Intern Med. 2021;174(9):1214-1223. doi: 10.7326/M20-7141 [DOI] [PMC free article] [PubMed] [Google Scholar]
10.Gumbinger C, Holstein T, Stock C, Rizos T, Horstmann S, Veltkamp R. Reasons underlying non-adherence to and discontinuation of anticoagulation in secondary stroke prevention among patients with atrial fibrillation. Eur Neurol. 2015;73(3-4):184-191. doi: 10.1159/000371574 [DOI] [PubMed] [Google Scholar]
11.van Deelen BAJ, van den Bemt PMLS, Egberts TCG, van ’t Hoff A, Maas HAAM. Cognitive impairment as determinant for sub-optimal control of oral anticoagulation treatment in elderly patients with atrial fibrillation. Drugs Aging. 2005;22(4):353-360. doi: 10.2165/00002512-200522040-00007 [DOI] [PubMed] [Google Scholar]
12.2021 Alzheimer’s disease facts and figures. Alzheimers Dement. 2021;17(3):327-406. doi: 10.1002/alz.12328 [DOI] [PubMed] [Google Scholar]
13.Alonso A, Knopman DS, Gottesman RF, et al. Correlates of dementia and mild cognitive impairment in patients with atrial fibrillation: the Atherosclerosis Risk in Communities Neurocognitive Study (ARIC-NCS). J Am Heart Assoc. 2017;6(7):e006014. doi: 10.1161/JAHA.117.006014 [DOI] [PMC free article] [PubMed] [Google Scholar]
14.Ott A, Breteler MM, de Bruyne MC, van Harskamp F, Grobbee DE, Hofman A. Atrial fibrillation and dementia in a population-based study. the Rotterdam Study. Stroke. 1997;28(2):316-321. doi: 10.1161/01.STR.28.2.316 [DOI] [PubMed] [Google Scholar]
15.Connolly SJ, Ezekowitz MD, Yusuf S, et al. ; RE-LY Steering Committee and Investigators . Dabigatran versus warfarin in patients with atrial fibrillation. N Engl J Med. 2009;361(12):1139-1151. doi: 10.1056/NEJMoa0905561 [DOI] [PubMed] [Google Scholar]
16.Patel MR, Mahaffey KW, Garg J, et al. ROCKET AF Investigators. Rivaroxaban versus warfarin in nonvalvular atrial fibrillation. N Engl J Med. 2011;365(10):883-891. doi: 10.1056/NEJMoa1009638 [DOI] [PubMed] [Google Scholar]
17.Granger CB, Alexander JH, McMurray JJV, et al. ; ARISTOTLE Committees and Investigators . Apixaban versus warfarin in patients with atrial fibrillation. N Engl J Med. 2011;365(11):981-992. doi: 10.1056/NEJMoa1107039 [DOI] [PubMed] [Google Scholar]
18.Giugliano RP, Ruff CT, Braunwald E, et al. ; ENGAGE AF-TIMI 48 Investigators . Edoxaban versus warfarin in patients with atrial fibrillation. N Engl J Med. 2013;369(22):2093-2104. doi: 10.1056/NEJMoa1310907 [DOI] [PubMed] [Google Scholar]
19.Mody L, Miller DK, McGloin JM, et al. Recruitment and retention of older adults in aging research. J Am Geriatr Soc. 2008;56(12):2340-2348. doi: 10.1111/j.1532-5415.2008.02015.x [DOI] [PMC free article] [PubMed] [Google Scholar]
20.Kennedy-Martin T, Curtis S, Faries D, Robinson S, Johnston J. A literature review on the representativeness of randomized controlled trial samples and implications for the external validity of trial results. Trials. 2015;16:495. doi: 10.1186/s13063-015-1023-4 [DOI] [PMC free article] [PubMed] [Google Scholar]
21.von Elm E, Altman DG, Egger M, Pocock SJ, Gøtzsche PC, Vandenbroucke JP. STROBE Initiative. The Strengthening the Reporting of Observational Studies in Epidemiology (STROBE) statement: guidelines for reporting observational studies. J Clin Epidemiol. 2008;61(4):344-349. doi: 10.1016/j.jclinepi.2007.11.008 [DOI] [PubMed] [Google Scholar]
22.Andrade JG, Krahn AD, Skanes AC, Purdham D, Ciaccia A, Connors S. Values and preferences of physicians and patients with nonvalvular atrial fibrillation who receive oral anticoagulation therapy for stroke prevention. Can J Cardiol. 2016;32(6):747-753. doi: 10.1016/j.cjca.2015.09.023 [DOI] [PubMed] [Google Scholar]
23.Urbaniak AM, Strøm BO, Krontveit R, Svanqvist KH. Prescription patterns of non–vitamin K oral anticoagulants across indications and factors associated with their increased prescribing in atrial fibrillation between 2012-2015: a study from the Norwegian Prescription Database. Drugs Aging. 2017;34(8):635-645. doi: 10.1007/s40266-017-0476-4 [DOI] [PubMed] [Google Scholar]
24.Disdier Moulder MPA, Hendricks AK, Ou NN. Towards appropriate polypharmacy in older cardiovascular patients: how many medications do I have to take? Clin Cardiol. 2020;43(2):137-144. doi: 10.1002/clc.23304 [DOI] [PMC free article] [PubMed] [Google Scholar]
25.Taylor DH Jr, Østbye T, Langa KM, Weir D, Plassman BL. The accuracy of Medicare claims as an epidemiological tool: the case of dementia revisited. J Alzheimers Dis. 2009;17(4):807-815. doi: 10.3233/JAD-2009-1099 [DOI] [PMC free article] [PubMed] [Google Scholar]
26.Pisters R, Lane DA, Nieuwlaat R, de Vos CB, Crijns HJGM, Lip GYH. A novel user-friendly score (HAS-BLED) to assess 1-year risk of major bleeding in patients with atrial fibrillation: the Euro Heart Survey. Chest. 2010;138(5):1093-1100. doi: 10.1378/chest.10-0134 [DOI] [PubMed] [Google Scholar]
27.Lip GYH, Frison L, Halperin JL, Lane DA. Comparative validation of a novel risk score for predicting bleeding risk in anticoagulated patients with atrial fibrillation: the HAS-BLED (hypertension, abnormal renal/liver function, stroke, bleeding history or predisposition, labile INR, elderly, drugs/alcohol concomitantly) score. J Am Coll Cardiol. 2011;57(2):173-180. doi: 10.1016/j.jacc.2010.09.024 [DOI] [PubMed] [Google Scholar]
28.Wilkinson T, Ly A, Schnier C, et al. ; UK Biobank Neurodegenerative Outcomes Group and Dementias Platform UK . Identifying dementia cases with routinely collected health data: a systematic review. Alzheimers Dement. 2018;14(8):1038-1051. doi: 10.1016/j.jalz.2018.02.016 [DOI] [PMC free article] [PubMed] [Google Scholar]
29.Condition categories. Chronic Conditions Data Warehouse, Centers for Medicare & Medicaid Services. Accessed November 29, 2017. https://www.ccwdata.org/web/guest/condition-categories
30.O’Brien JT, Holmes C, Jones M, et al. Clinical practice with anti-dementia drugs: a revised (third) consensus statement from the British Association for Psychopharmacology. J Psychopharmacol. 2017;31(2):147-168. doi: 10.1177/0269881116680924 [DOI] [PubMed] [Google Scholar]
31.Chu LW. Alzheimer’s disease: early diagnosis and treatment. Hong Kong Med J. 2012;18(3):228-237. [PubMed] [Google Scholar]
32.Gagne JJ, Glynn RJ, Avorn J, Levin R, Schneeweiss S. A combined comorbidity score predicted mortality in elderly patients better than existing scores. J Clin Epidemiol. 2011;64(7):749-759. doi: 10.1016/j.jclinepi.2010.10.004 [DOI] [PMC free article] [PubMed] [Google Scholar]
33.Kim DH, Glynn RJ, Avorn J, et al. Validation of a claims-based frailty index against physical performance and adverse health outcomes in the Health and Retirement Study. J Gerontol A Biol Sci Med Sci. 2019;74(8):1271-1276. doi: 10.1093/gerona/gly197 [DOI] [PMC free article] [PubMed] [Google Scholar]
34.Kim DH, Schneeweiss S, Glynn RJ, Lipsitz LA, Rockwood K, Avorn J. Measuring frailty in Medicare data: development and validation of a claims-based frailty index. J Gerontol A Biol Sci Med Sci. 2018;73(7):980-987. doi: 10.1093/gerona/glx229 [DOI] [PMC free article] [PubMed] [Google Scholar]
35.Kim DH, Patorno E, Pawar A, Lee H, Schneeweiss S, Glynn RJ. Measuring frailty in administrative claims data: comparative performance of four claims-based frailty measures in the U.S. Medicare data. J Gerontol A Biol Sci Med Sci. 2020;75(6):1120-1125. doi: 10.1093/gerona/glz224 [DOI] [PMC free article] [PubMed] [Google Scholar]
36.Gautam N, Bessette L, Pawar A, Levin R, Kim DH. Updating International Classification of Diseases Ninth Revision to Tenth Revision of a claims-based frailty index. J Gerontol A Biol Sci Med Sci. 2021;76(7):1316-1317. doi: 10.1093/gerona/glaa150 [DOI] [PMC free article] [PubMed] [Google Scholar]
37.Rockwood K, Mitnitski A. Frailty in relation to the accumulation of deficits. J Gerontol A Biol Sci Med Sci. 2007;62(7):722-727. doi: 10.1093/gerona/62.7.722 [DOI] [PubMed] [Google Scholar]
38.Shi SM, McCarthy EP, Mitchell S, Kim DH. Changes in predictive performance of a frailty index with availability of clinical domains. J Am Geriatr Soc. 2020;68(8):1771-1777. doi: 10.1111/jgs.16436 [DOI] [PMC free article] [PubMed] [Google Scholar]
39.Shi SM, McCarthy EP, Mitchell SL, Kim DH. Predicting mortality and adverse outcomes: comparing the frailty index to general prognostic indices. J Gen Intern Med. 2020;35(5):1516-1522. doi: 10.1007/s11606-020-05700-w [DOI] [PMC free article] [PubMed] [Google Scholar]
40.Tirschwell DL, Longstreth WT Jr. Validating administrative data in stroke research. Stroke. 2002;33(10):2465-2470. doi: 10.1161/01.STR.0000032240.28636.BD [DOI] [PubMed] [Google Scholar]
41.Roumie CL, Mitchel E, Gideon PS, Varas-Lorenzo C, Castellsague J, Griffin MR. Validation of ICD-9 codes with a high positive predictive value for incident strokes resulting in hospitalization using Medicaid health data. Pharmacoepidemiol Drug Saf. 2008;17(1):20-26. doi: 10.1002/pds.1518 [DOI] [PubMed] [Google Scholar]
42.Cunningham A, Stein CM, Chung CP, Daugherty JR, Smalley WE, Ray WA. An automated database case definition for serious bleeding related to oral anticoagulant use. Pharmacoepidemiol Drug Saf. 2011;20(6):560-566. doi: 10.1002/pds.2109 [DOI] [PMC free article] [PubMed] [Google Scholar]
43.Fleiss JL. The statistical basis of meta-analysis. Stat Methods Med Res. 1993;2(2):121-145. doi: 10.1177/096228029300200202 [DOI] [PubMed] [Google Scholar]
44.Schneeweiss S, Rassen JA, Glynn RJ, Avorn J, Mogun H, Brookhart MA. High-dimensional propensity score adjustment in studies of treatment effects using health care claims data. Epidemiology. 2009;20(4):512-522. doi: 10.1097/EDE.0b013e3181a663cc [DOI] [PMC free article] [PubMed] [Google Scholar]
45.Rassen JA, Choudhry NK, Avorn J, Schneeweiss S. Cardiovascular outcomes and mortality in patients using clopidogrel with proton pump inhibitors after percutaneous coronary intervention or acute coronary syndrome. Circulation. 2009;120(23):2322-2329. doi: 10.1161/CIRCULATIONAHA.109.873497 [DOI] [PMC free article] [PubMed] [Google Scholar]
46.Garbe E, Kloss S, Suling M, Pigeot I, Schneeweiss S. High-dimensional versus conventional propensity scores in a comparative effectiveness study of coxibs and reduced upper gastrointestinal complications. Eur J Clin Pharmacol. 2013;69(3):549-557. doi: 10.1007/s00228-012-1334-2 [DOI] [PubMed] [Google Scholar]
47.Le HV, Poole C, Brookhart MA, et al. Effects of aggregation of drug and diagnostic codes on the performance of the high-dimensional propensity score algorithm: an empirical example. BMC Med Res Methodol. 2013;13:142. doi: 10.1186/1471-2288-13-142 [DOI] [PMC free article] [PubMed] [Google Scholar]
48.Hallas J, Pottegård A. Performance of the high-dimensional propensity score in a Nordic healthcare model. Basic Clin Pharmacol Toxicol. 2017;120(3):312-317. doi: 10.1111/bcpt.12716 [DOI] [PubMed] [Google Scholar]
49.Wang SV, Verpillat P, Rassen JA, Patrick A, Garry EM, Bartels DB. Transparency and reproducibility of observational cohort studies using large healthcare databases. Clin Pharmacol Ther. 2016;99(3):325-332. doi: 10.1002/cpt.329 [DOI] [PMC free article] [PubMed] [Google Scholar]
50.Fralick M, Kesselheim AS, Avorn J, Schneeweiss S. Use of health care databases to support supplemental indications of approved medications. JAMA Intern Med. 2018;178(1):55-63. doi: 10.1001/jamainternmed.2017.3919 [DOI] [PMC free article] [PubMed] [Google Scholar]
51.Kim SC, Solomon DH, Rogers JR, et al. Cardiovascular safety of tocilizumab versus tumor necrosis factor inhibitors in patients with rheumatoid arthritis: a multi-database cohort study. Arthritis Rheumatol. 2017;69(6):1154-1164. doi: 10.1002/art.40084 [DOI] [PMC free article] [PubMed] [Google Scholar]
52.Hernandez I, Baik SH, Piñera A, Zhang Y. Risk of bleeding with dabigatran in atrial fibrillation. JAMA Intern Med. 2015;175(1):18-24. doi: 10.1001/jamainternmed.2014.5398 [DOI] [PMC free article] [PubMed] [Google Scholar]
53.Graham DJ, Reichman ME, Wernecke M, et al. Cardiovascular, bleeding, and mortality risks in elderly Medicare patients treated with dabigatran or warfarin for nonvalvular atrial fibrillation. Circulation. 2015;131(2):157-164. doi: 10.1161/CIRCULATIONAHA.114.012061 [DOI] [PubMed] [Google Scholar]
54.Graham DJ, Reichman ME, Wernecke M, et al. Stroke, bleeding, and mortality risks in elderly Medicare beneficiaries treated with dabigatran or rivaroxaban for nonvalvular atrial fibrillation. JAMA Intern Med. 2016;176(11):1662-1671. doi: 10.1001/jamainternmed.2016.5954 [DOI] [PubMed] [Google Scholar]
55.Amin A, Keshishian A, Trocio J, et al. Risk of stroke/systemic embolism, major bleeding and associated costs in non-valvular atrial fibrillation patients who initiated apixaban, dabigatran or rivaroxaban compared with warfarin in the United States Medicare population. Curr Med Res Opin. 2017;33(9):1595-1604. doi: 10.1080/03007995.2017.1345729 [DOI] [PubMed] [Google Scholar]
56.Lip GYH, Keshishian A, Li X, et al. Effectiveness and safety of oral anticoagulants among nonvalvular atrial fibrillation patients. Stroke. 2018;49(12):2933-2944. doi: 10.1161/STROKEAHA.118.020232 [DOI] [PMC free article] [PubMed] [Google Scholar]
57.Lauffenburger JC, Farley JF, Gehi AK, Rhoney DH, Brookhart MA, Fang G. Effectiveness and safety of dabigatran and warfarin in real-world US patients with non-valvular atrial fibrillation: a retrospective cohort study. J Am Heart Assoc. 2015;4(4):e001798. doi: 10.1161/JAHA.115.001798 [DOI] [PMC free article] [PubMed] [Google Scholar]
58.Abraham NS, Noseworthy PA, Yao X, Sangaralingham LR, Shah ND. Gastrointestinal safety of direct oral anticoagulants: a large population-based study. Gastroenterology. 2017;152(5):1014-1022. doi: 10.1053/j.gastro.2016.12.018 [DOI] [PubMed] [Google Scholar]
59.Norby FL, Alonso A. Comparative effectiveness of rivaroxaban in the treatment of nonvalvular atrial fibrillation. J Comp Eff Res. 2017;6(6):549-560. doi: 10.2217/cer-2017-0025 [DOI] [PMC free article] [PubMed] [Google Scholar]
60.Go AS, Singer DE, Toh S, et al. Outcomes of dabigatran and warfarin for atrial fibrillation in contemporary practice: a retrospective cohort study. Ann Intern Med. 2017;167(12):845-854. doi: 10.7326/M16-1157 [DOI] [PubMed] [Google Scholar]
61.Seeger JD, Bykov K, Bartels DB, Huybrechts K, Zint K, Schneeweiss S. Safety and effectiveness of dabigatran and warfarin in routine care of patients with atrial fibrillation. Thromb Haemost. 2015;114(6):1277-1289. [DOI] [PubMed] [Google Scholar]
62.Fanning L, Lau WCY, Mongkhon P, et al. Safety and effectiveness of direct oral anticoagulants vs warfarin in people with atrial fibrillation and dementia. J Am Med Dir Assoc. 2020;21(8):1058-1064. doi: 10.1016/j.jamda.2019.11.022 [DOI] [PubMed] [Google Scholar]
63.Shah SJ, Singer DE, Fang MC, Reynolds K, Go AS, Eckman MH. Net clinical benefit of oral anticoagulation among older adults with atrial fibrillation. Circ Cardiovasc Qual Outcomes. 2019;12(11):e006212. doi: 10.1161/CIRCOUTCOMES.119.006212 [DOI] [PMC free article] [PubMed] [Google Scholar]
64.Zhou J, Yu JT, Wang HF, et al. Association between stroke and Alzheimer’s disease: systematic review and meta-analysis. J Alzheimers Dis. 2015;43(2):479-489. doi: 10.3233/JAD-140666 [DOI] [PubMed] [Google Scholar]
65.Olichney JM, Hansen LA, Hofstetter CR, Grundman M, Katzman R, Thal LJ. Cerebral infarction in Alzheimer’s disease is associated with severe amyloid angiopathy and hypertension. Arch Neurol. 1995;52(7):702-708. doi: 10.1001/archneur.1995.00540310076019 [DOI] [PubMed] [Google Scholar]
66.Franklin JM, Patorno E, Desai RJ, et al. Emulating randomized clinical trials with nonrandomized real-world evidence studies: first results from the RCT DUPLICATE Initiative. Circulation. 2021;143(10):1002-1013. doi: 10.1161/CIRCULATIONAHA.120.051718 [DOI] [PMC free article] [PubMed] [Google Scholar]
67.Lin KJ, Schneeweiss S, Pawar A, Singer DE, Liu J, Gagne JJ. Using a simple prescription gap to determine warfarin discontinuation can lead to substantial misclassification. Thromb Haemost. 2022;122(3):386-393. doi: 10.1055/a-1508-8187 [DOI] [PubMed] [Google Scholar]

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Supplementary Materials