Anticoagulation Outcomes in Nursing Home Residents With Atrial Fibrillation and ADL Disability

Qiaoxi Chen; Daniel Singer; Chun-Ting Yang; Kevin T Pritchard; Dae Hyun Kim; Kueiyu Joshua Lin

doi:10.1001/jamanetworkopen.2026.37533

. 2026 Oct 5;9(10):e2637533. doi: 10.1001/jamanetworkopen.2026.37533

Anticoagulation Outcomes in Nursing Home Residents With Atrial Fibrillation and ADL Disability

Qiaoxi Chen ^1,^✉, Daniel Singer ^2,³, Chun-Ting Yang ¹, Kevin T Pritchard ¹, Dae Hyun Kim ⁴, Kueiyu Joshua Lin ^1,^2,^✉

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

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

³Department of Medicine, Harvard Medical School, Boston, Massachusetts

⁴The Hinda and Arthur Marcus Institute for Aging Research, Hebrew SeniorLife, Harvard Medical School, Boston, Massachusetts

Accepted for Publication: August 11, 2026.

Published: October 5, 2026. doi:10.1001/jamanetworkopen.2026.37533

Open Access: This is an open access article distributed under the terms of the CC-BY-NC-ND License, which does not permit alteration or commercial use, including those for text and data mining, AI training, and similar technologies. © 2026 Chen Q et al. JAMA Network Open.

^✉

Corresponding Authors: Qiaoxi Chen (qchen11@bwh.harvard.edu), and Kueiyu Joshua Lin (jklin@bwh.harvard.edu), Division of Pharmacoepidemiology and Pharmacoeconomics, Department of Medicine, Brigham and Women’s Hospital, Harvard Medical School, 399 Revolution Dr, Somerville, MA 02145.

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

Concept and design: Chen, Yang, Kim, Lin.

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

Drafting of the manuscript: Chen, Lin.

Critical review of the manuscript for important intellectual content: Singer, Yang, Pritchard, Kim, Lin.

Statistical analysis: Chen, Yang, Pritchard, Lin.

Obtained funding: Lin.

Administrative, technical, or material support: Chen, Pritchard, Lin.

Supervision: Lin.

Conflict of Interest Disclosures: Dr Singer reported receiving grants from Bristol Myers Squibb and personal fees from Bayer Healthcare outside the submitted work. Dr Kim reported receiving personal fees from Pfizer and grants from the National Institutes of health outside the submitted work. No other disclosures were reported.

Funding/Support: This work was supported by grants R01AG075335 and R01AG081412 from the National Institute on Aging of the National Institutes of Health.

Role of the Funder/Sponsor: The funder 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: PMC13639757 PMID: 42832198

Abstract

Importance

Long-term nursing home residents with atrial fibrillation (AF) and activities of daily living (ADL) disability are at high risk of bleeding and death, making the net benefit of anticoagulation uncertain.

Objective

To compare the association of oral anticoagulant (OAC) use vs nonuse with ischemic stroke and major bleeding in long-term nursing home residents with AF and ADL disability.

Design, Setting, and Participants

This cohort study used US Medicare claims linked with the Minimum Data Set (January 1, 2013, to December 31, 2022). Nursing home residents with nonvalvular AF and impairment in 3 or more ADLs were included. Three cohorts comparing use of apixaban, rivaroxaban, or warfarin with nonuse were constructed using incidence-density sampling.

Exposures

Initiation of apixaban, rivaroxaban, or warfarin use.

Main Outcomes and Measures

Primary outcomes were ischemic stroke, major bleeding, and their composite. Propensity score (PS) matching balanced 111 covariates. Adjusted rate ratios (ARRs) with 95% CIs were estimated. Tipping-point analyses assessed the relative weighting ischemic stroke vs major bleeding required to favor anticoagulation. Direct active-comparator analyses compared apixaban with rivaroxaban and warfarin.

Results

The study comprised 26 940 nursing home residents (median [IQR] age, 82 [74,88] years; 17 504 [65.0%] female), including PS-matched cohorts of 7600 residents receiving apixaban, 2654 receiving rivaroxaban, and 3216 receiving warfarin. Compared with nonuse, use of all OACs was associated with lower rates of ischemic stroke (apixaban: ARR, 0.53; 95% CI, 0.38-0.73; rivaroxaban: ARR, 0.48; 95% CI, 0.27-0.87; warfarin: ARR, 0.47; 95% CI, 0.25-0.88) and higher rates of major bleeding (apixaban: ARR, 1.62; 95% CI, 1.29-2.03; rivaroxaban: ARR, 2.68; 95% CI, 1.84-3.93; warfarin: ARR, 3.02; 95% CI, 2.15-4.25). Only apixaban was not associated with a higher composite outcome (ARR, 1.08; 95% CI, 0.90-1.30), whereas rivaroxaban (ARR, 1.62; 95% CI, 1.20-2.18) and warfarin (ARR, 1.99; 95% CI, 1.50-2.63) were associated with higher rates. Findings were consistent across frailty strata, although stroke reduction with apixaban was attenuated among frail individuals. Tipping-point analyses indicated a more favorable benefit-risk balance for apixaban than for rivaroxaban or warfarin.

Conclusions and Relevance

Among nursing home residents with AF and ADL disability, OAC use was associated with lower ischemic stroke but higher major bleeding rates than nonuse. Apixaban demonstrated the most favorable benefit-risk profile among the OACs evaluated. These findings may inform individualized anticoagulation decisions.

This cohort study of nursing home residents with atrial fibrillation (AF) and activities of daily living (ADL) disability evaluates the association of oral anticoagulant use and clinical outcomes.

Key Points

Question

What are the comparative effectiveness and safety outcomes associated with oral anticoagulant use compared with nonuse among long-term nursing home residents with atrial fibrillation and activities of daily living disability?

Findings

In this cohort study of 26 940 nursing home residents, use of oral anticoagulants was associated with lower ischemic stroke rates but higher major bleeding rates than nonuse. Apixaban showed the most favorable benefit-risk profile and, in direct active-comparator analyses, was associated with less major bleeding than rivaroxaban and warfarin while maintaining similar ischemic stroke rates.

Meaning

Anticoagulation decisions among nursing home residents with atrial fibrillation and activities of daily living disability should be individualized, and apixaban may offer the most favorable balance between stroke prevention and bleeding risk when anticoagulation is appropriate.

Introduction

Atrial fibrillation (AF) is the most common substantial cardiac arrhythmia, and its prevalence increases with age.¹ The number of nursing home residents with AF is projected to exceed 300 000 by 2030.² AF is substantially more prevalent among nursing home residents (10%-17%) than in the general population (1%-2%).^3,4,5 Although clinical guidelines recommend oral anticoagulants (OACs) to reduce stroke risk, OACs are often underused or discontinued in older adults, particularly in nursing homes because of concerns about bleeding risk and falls.^6,7

In older adults, AF is associated with greater functional dependence, in part because AF-related strokes often result in life-altering disability, and coexists with conditions, such as falls and cognitive impairment, that further contribute to functional decline and reduced quality of life.^{8,9,10,11,12,13} Approximately 58% of nursing home residents with AF have moderate or severe activities of daily living (ADL) disability, underscoring the challenges of OAC decision-making in this at-risk population.¹⁴ ADL disability may further influence the decision by reflecting greater frailty and limited life expectancy, which can shift anticoagulation decisions toward avoiding treatment.^15,16,17

The decreased use of OACs in this population may have important consequences because prior studies^18,19,20 have shown that individuals with ADL disability have a significantly elevated risk of stroke. Conversely, older adults with AF at high risk of falls experience substantially higher rates of intracranial hemorrhage (ICH),^21,22,23 the category of anticoagulant-related bleeding at least as severe as ischemic stroke. Despite the high burden of both stroke and bleeding risk in this population, there is a lack of direct evidence evaluating the safety and effectiveness of OACs. To address this gap, we evaluated the association between OAC use and clinical outcomes in a US national cohort of long-term nursing home residents with AF and ADL disability.

Methods

Study Design and Data Sources

This retrospective cohort study used a target trial emulation framework^24,25 (eTable 1 in Supplement 1) to emulate 3 hypothetical trials comparing use vs nonuse of 3 OACs (ie, apixaban, rivaroxaban, and warfarin) among long-term nursing home residents with AF and ADL disability. The study used US Medicare claims data with the Minimum Data Set (MDS), version 3.0.²⁶ The analyses used deidentified linked Medicare claims and MDS 3.0 data obtained under a data use agreement with the Centers for Medicare & Medicaid Services. Medicare claims provide longitudinal information on demographics, health care utilization, diagnoses, procedures, and prescription dispensing, whereas the MDS 3.0 contains standardized assessments of cognition and functional status among US nursing home residents. The study was approved by the Mass General Brigham Institutional Review Board with a waiver of informed consent because data were deidentified. This study was reported in accordance with the Reporting of Studies Conducted Using Observational Routinely Collected Health Data for Pharmacoepidemiology (RECORD-PE) guideline, an extension of the Strengthening the Reporting of Observational Studies in Epidemiology (STROBE) reporting guideline for cohort studies.

Study Population

Using data from January 1, 2013, and December 31, 2022, we included individuals who met the following cohort eligibility criteria: (1) 1 or more inpatient or 2 or more outpatient claims for AF; (2) impairment in 3 or more ADL items (scores of 2-4 indicating dependence) to limit to patients with advanced ADL disability^27,28; (3) resident in a long-term nursing home; (4) CHA₂DS₂-VASc (congestive heart failure; hypertension; age ≥75 years; diabetes; prior stroke, transient ischemic attack, or thromboembolism; vascular disease; age 65-74 years; and sex category) score of 2 or higher for men or 3 or higher for women; (5) age of 65 years or older; and (6) continuous enrollment in Medicare Parts A (hospital), B (outpatient), and D (prescription drug coverage) during the 365 days before cohort entry date (CED). We excluded individuals with the following characteristics during the 365 days before the CED: (1) valvular AF (ie, rheumatic mitral valve disease or presence of a mechanical heart valve); (2) prior left atrial appendage occlusion (Watchman device or surgical); (3) venous thromboembolism; (4) joint replacement surgery; (5) hospice or palliative care; or (6) major bleeding or ischemic stroke within 60 days before the CED (eTables 2 and 7 and eFigure 1in Supplement 1).

Use and Nonuse of OACs

New OAC use was considered the initiation of apixaban, rivaroxaban, or warfarin without prior use in the preceding 365 days. Nonuse was no OAC use on the sampling date and in the prior 365 days. Using the incidence-density sampling method,²⁹ we matched each patient receiving an OAC to a patient not receiving an OAC on time since AF diagnosis within a caliper of 30 days, CHA₂DS₂-VASc score, recent hospitalization or skilled nursing facility stay (within 30 days), and prior major bleeding (within 90 days). The CED was defined as the dispensing date for those receiving OACs and the sampling date for those not receiving OACs. Patients entered the study cohort only once, at their first qualifying CED. Both patients receiving and not taking OACs were required to have an AF diagnosis in the 180 days before CED.

Outcomes and Follow-Up

Primary outcomes were ischemic stroke, major bleeding, and their composite. Secondary outcomes included gastrointestinal bleeding, ICH, and mortality. Outcomes were identified using validated claims-based algorithms.^30,31,32 Follow-up began 1 day after CED and continued until the earliest of the following: (1) outcome occurrence, (2) Medicare disenrollment, (3) death, (4) maximum follow-up of 365 days, (5) end of data availability, or (6) change in exposure status (ie, initiation of OACs in those not previously taking OACs or initiation of a different OAC or discontinuation of OACs among those previously taking OACs, defined as a >30-day gap in supply among users). This corresponds to an as-treated approach to assess the effects while following the index treatment strategy.

Baseline Covariates

Baseline covariates were assessed during the 365-day baseline assessment period before and including the CED. We assessed 111 covariates, including sociodemographic characteristics, comorbidities, medication use, and health care utilization. Race was included as a baseline covariate because it may be associated with differences in OAC treatment patterns and clinical outcomes. These covariates were used to estimate the CHA₂DS₂-VASc score,^33,34 HAS-BLED (hypertension, abnormal kidney or liver function, stroke, bleeding, elderly, and drugs or excess alcohol use) score,^35,36 Gagne Comorbidity Index,³⁷ and a claim-based frailty index.^38,39,40,41 For the comparison of those receiving warfarin vs those not receiving any OAC, we additionally incorporated a validated claims-based prediction of time in therapeutic range (TTR) as a proxy for anticoagulation control.⁴²

Statistical Analysis

Main Analysis

Baseline characteristics were compared between those receiving and those not receiving OACs within each cohort. Propensity scores (PSs) were estimated using logistic regression models in all 3 OAC cohorts vs the nonuse cohort, including 111 baseline covariates. We performed 1:1 nearest neighbor matching with a caliper of 0.01. Covariate balance was assessed using standardized mean differences (SMDs), with an SMD less than 0.1 indicating adequate balance.

The primary analyses estimated the comparative effectiveness and safety associated with apixaban, rivaroxaban, or warfarin use compared with nonuse using an as-treated approach. In the matched cohorts, adjusted rate ratios (ARRs) were estimated using Poisson regression models, whereas adjusted rate differences (ARDs) were estimated using modified generalized linear models with a normal distribution and identity link⁴³ (eMethods in Supplement 1).

As a complementary analysis, we performed an 180-day intention-to-treat (ITT) analysis in which participants were followed for up to 180 days regardless of subsequent treatment changes. This analysis complements the primary as-treated analysis by estimating the effect associated with the initial treatment strategy at cohort entry.

To further complement the primary analyses comparing OAC use with nonuse, we performed direct active-comparator analyses comparing those receiving apixaban with those receiving rivaroxaban and warfarin. Separate active-comparator cohorts were constructed from eligible patients receiving OACs, and 1:1 PS matching was performed using the same baseline covariates and analytic approach as in the primary analyses. We also conducted a tipping-point analysis using the ARDs for ischemic stroke and major bleeding to identify the relative weighting of ischemic strokes vs major bleeding at which the overall benefit-risk balance favored anticoagulation (eMethods in Supplement 1).

Subgroup and Sensitivity Analyses

Subgroup analyses were conducted according to a claims-based frailty index (<0.35 vs ≥0.35). Patients were classified into a moderate to severe frail subgroup (claims-based frailty index ≥0.35) and a nonfrail or mildly frail subgroup (frail index <0.35). PS estimation and matching were conducted within frailty subgroups. Heterogeneity across subgroups was assessed using inverse variance-weighted z test (P for heterogeneity ≤ .05 was considered statistically significant).⁴⁴

Sensitivity analyses included (1) an additional 365-day ITT analysis; (2) an as-treated analysis allowing a longer permissible gap in OAC exposure (60 days); (3) a high-dimensional PS adjustment to account for residual confounding by automatically identifying and incorporating additional proxy variables^45,46,47,48; (4) an inverse probability of censoring weighting (IPCW) analysis; (5) a Fine and Gray competing-risk model to account for the competing risk of death; (6) an analysis requiring at least 1 health care encounter within 14 days before CED, which represented a recent opportunity for anticoagulation initiation; and (7) a dose-stratified analysis according to the prescribed OAC dose at treatment initiation (apixaban: 5 vs 2.5 mg; rivaroxaban: 20 vs 15 mg). Analyses were conducted using the Aetion Evidence Generation Platform, version 5.18.0 (Aetion) and SAS software, version 9.4 (SAS Institute Inc).^49,50,51

Results

Overall Baseline Characteristics

After PS matching, the study comprised 26 940 nursing home residents (median [IQR] age, 82 [74-88] years; 17 504 [65.0%] female and 9436 [35.0%] male; 3057 [11.3%] Black, 22 419 [83.2%] White, and 1464 [5.4%] unknown or other, including Asian, Hispanic, North American Native, and missing race categories), including PS-matched cohorts of 7600 residents receiving apixaban, 2654 receiving rivaroxaban, and 3216 receiving warfarin (Table 1). Before PS matching, those not receiving OACs were older (mean [SD] age, 83.0 [8.6] vs 81.1 [8.4] years), had lower prevalence of prior ischemic stroke, had higher prevalence of intracranial or major gastrointestinal bleeding, and had greater ADL disability than those receiving OACs. The entire population had a high burden of comorbidity, particularly heart failure and diabetes, and 16 250 (39.5%) were moderately to severely frail (eTable 3 in Supplement 1). After PS matching, the baseline characteristics were well balanced across all matched cohorts (all SMDs <0.1) (Table 1).

Table 1. Characteristics of Long-Term Nursing Home Residents With Atrial Fibrillation and ADL Disability Receiving Oral Anticoagulants.

Characteristic	Apixaban cohort			Rivaroxaban cohort			Warfarin cohort
Characteristic	Apixaban group, No. (%)^a (n = 7600)	Matched group, No. (%)^a (n = 7600)	SMD	Rivaroxaban group, No. (%)^a (n = 2654)	Matched group, No. (%)^a (n = 2654)	SMD	Warfarin group, No. (%)^a (n = 3216)	Matched group, No. (%)^a (n = 3216)	SMD
Sociodemographic
Age, mean (SD), y	82.1 (8.3)	82.2 (8.7)	0.00	81.9 (8.2)	81.8 (8.7)	0.01	81.9 (8.1)	81.7 (8.4)	0.02
Sex
Female	5074 (66.8)	5058 (66.6)	0.00	1788 (67.4)	1785 (67.3)	0.00	2154 (67.0)	2158 (67.1)	0.00
Male	2526 (33.2)	2542 (33.4)	0.01	866 (32.6)	869 (32.7)	0.03	1062 (33.0)	1058 (32.9)	0.01
Race^b
Black	898 (11.8)	853 (11.2)	0.02	262 (9.9)	270 (10.2)	0.01	376 (11.7)	398 (12.4)	0.02
White	6279 (82.6)	6330 (83.3)	0.02	2241 (84.4)	2234 (84.2)	0.01	2678 (83.3)	2657 (82.6)	0.02
Other or missing	423 (5.6)	417 (5.5)	0.00	151 (5.7)	150 (5.7)	0.00	162 (5.0)	161 (5.0)	0.00
Dual status (yes or no)	3587 (47.2)	3598 (47.3)	0.00	1290 (48.6)	1308 (49.3)	0.01	1562 (48.6)	1605 (49.9)	0.03
Risk scores, mean (SD)
CHA₂DS₂-VASc	5.8 (1.4)	5.8 (1.4)	0.00	5.8 (1.4)	5.8 (1.4)	0.00	5.9 (1.4)	5.9 (1.4)	0.01
HAS-BLED	2.9 (0.7)	2.9 (0.7)	0.01	2.9 (0.7)	2.9 (0.7)	0.02	2.9 (0.7)	2.9 (0.7)	0.01
Mean CCI	7.7 (3.0)	7.7 (3.0)	0.00	7.2 (3.0)	7.2 (3.0)	0.00	7.5 (3.1)	7.5 (3.1)	0.00
Mean CFI	0.3 (0.1)	0.3 (0.1)	0.01	0.3 (0.1)	0.3 (0.1)	0.03	0.3 (0.1)	0.3 (0.1)	0.00
Cardiovascular conditions
Acute myocardial infarction	917 (12.1)	901 (11.9)	0.01	280 (10.6)	280 (10.6)	0.00	397 (12.3)	408 (12.7)	0.01
Congestive heart failure	5206 (68.5)	5210 (68.6)	0.00	1848 (69.6)	1849 (69.7)	0.00	2383 (74.1)	2386 (74.2)	0.00
Hypertension	7268 (95.6)	7264 (95.6)	0.00	2535 (95.5)	2532 (95.4)	0.01	3093 (96.2)	3092 (96.1)	0.00
Ischemic heart disease	4539 (59.7)	4546 (59.8)	0.00	1553 (58.5)	1532 (57.7)	0.02	1958 (60.9)	1965 (61.1)	0.00
Ischemic stroke	1393 (18.3)	1355 (17.8)	0.01	317 (11.9)	335 (12.6)	0.02	371 (11.5)	383 (11.9)	0.01
PVD or PVD surgery	2648 (34.8)	2673 (35.2)	0.01	987 (37.2)	968 (36.5)	0.02	1277 (39.7)	1251 (38.9)	0.02
Syncope	1187 (15.6)	1197 (15.8)	0.00	389 (14.7)	382 (14.4)	0.01	473 (14.7)	483 (15.0)	0.01
Noncardiovascular conditions
Acute kidney failure	3515 (46.3)	3464 (45.6)	0.01	1075 (40.5)	1053 (39.7)	0.02	1321 (41.1)	1332 (41.4)	0.01
COPD	3420 (45.0)	3436 (45.2)	0.00	1212 (45.7)	1190 (44.8)	0.02	1442 (44.8)	1446 (45.0)	0.00
Diabetes	4326 (56.9)	4321 (56.9)	0.00	1507 (56.8)	1473 (55.5)	0.03	1877 (58.4)	1876 (58.3)	0.00
Falls	1702 (22.4)	1700 (22.4)	0.00	548 (20.6)	541 (20.4)	0.01	671 (20.9)	677 (21.1)	0.01
Fractures	1745 (23.0)	1777 (23.4)	0.01	635 (23.9)	626 (23.6)	0.01	746 (23.2)	758 (23.6)	0.01
Intracranial bleed	251 (3.3)	239 (3.1)	0.01	80 (3.0)	71 (2.7)	0.02	103 (3.2)	106 (3.3)	0.01
Major gastrointestinal bleeding	1476 (19.4)	1451 (19.1)	0.01	514 (19.4)	514 (19.4)	0.00	709 (22.0)	729 (22.7)	0.02
Other bleeding event^c	1579 (20.8)	1585 (20.9)	0.00	497 (18.7)	506 (19.1)	0.01	658 (20.5)	668 (20.8)	0.01
Obesity	1894 (24.9)	1860 (24.5)	0.01	676 (25.5)	642 (24.2)	0.03	816 (25.4)	855 (26.6)	0.03
Liver disease	925 (12.2)	929 (12.2)	0.02	294 (11.1)	311 (11.7)	0.02	368 (11.4)	362 (11.3)	0.01
Malignant tumor	870 (11.4)	847 (11.1)	0.01	308 (11.6)	322 (12.1)	0.02	424 (13.2)	404 (12.6)	0.02
Peptic ulcer	472 (6.2)	492 (6.5)	0.01	140 (5.3)	139 (5.2)	0.00	140 (4.4)	157 (4.9)	0.03
Smoking	3120 (41.1)	3116 (41.0)	0.00	952 (35.9)	940 (35.4)	0.01	1139 (35.4)	1157 (36.0)	0.01
Cardiovascular medications
ACEIs	2118 (27.9)	2149 (28.3)	0.01	854 (32.2)	835 (31.5)	0.02	1037 (32.2)	1052 (32.7)	0.01
Angiotensin II receptor blockers	262 (3.4)	274 (3.6)	0.01	103 (3.9)	108 (4.1)	0.01	140 (4.4)	126 (3.9)	0.02
Antiarrhythmic agents	1046 (13.8)	1069 (14.1)	0.01	351 (13.2)	348 (13.1)	0.00	390 (12.1)	397 (12.3)	0.01
Antiplatelet agent	1563 (20.6)	1599 (21.0)	0.01	544 (20.5)	525 (19.8)	0.02	685 (21.3)	691 (21.5)	0.01
β-Blockers	5024 (66.1)	5066 (66.7)	0.01	1705 (64.2)	1689 (63.6)	0.01	2138 (66.5)	2122 (66.0)	0.01
Calcium channel blockers	219 (2.9)	213 (2.8)	0.01	76 (2.9)	73 (2.8)	0.01	90 (2.8)	92 (2.9)	0.00
Diuretics	4834 (63.6)	4825 (63.5)	0.00	1738 (65.5)	1728 (65.1)	0.01	2146 (66.7)	2148 (66.8)	0.00
Statins	5209 (68.5)	5169 (68.0)	0.01	1731 (65.2)	1743 (65.7)	0.01	2090 (65.0)	2095 (65.1)	0.00
Health care utilization
No. of office visits, mean (SD)	5.9 (7.1)	5.8 (7.4)	0.00	6.0 (7.8)	5.9 (7.6)	0.02	6.2 (7.5)	6.2 (7.4)	0.00
No. of ED visits, mean (SD)	0.2 (0.5)	0.2 (0.5)	0.01	0.2 (0.5)	0.2 (0.5)	0.02	0.2 (0.5)	0.2 (0.5)	0.01
No. of hospital days. mean (SD)	4.3 (5.7)	4.3 (5.7)	0.00	3.9 (5.7)	3.9 (5.5)	0.00	4.2 (5.8)	4.4 (5.8)	0.03
No. of SNF days, mean (SD)	12.0 (12.7)	12.1 (12.7)	0.01	11.7 (12.6)	11.9 (12.9)	0.02	12.0 (12.8)	11.8 (12.7)	0.02
Home health stay	2276 (29.9)	2278 (26.0)	0.00	677 (25.5)	659 (24.8)	0.02	796 (24.8)	810 (25.2)	0.01
Geographic region
North	1617 (21.3)	1618 (21.3)	0.00	540 (20.3)	526 (19.8)	0.01	769 (23.9)	744 (23.1)	0.02
Midwest	1865 (24.5)	1871 (24.6)	0.00	661 (24.9)	641 (24.2)	0.02	911 (28.3)	941 (29.3)	0.02
South and others	3147 (41.4)	3125 (41.1)	0.01	1091 (41.1)	1106 (41.7)	0.01	1130 (35.1)	1149 (35.7)	0.01
West	971 (12.8)	986 (13.0)	0.01	362 (13.6)	381 (14.4)	0.02	406 (12.6)	382 (11.9)	0.02
ADL impairment
Bathing	7322 (96.3)	7330 (96.4)	0.01	2558 (96.4)	2556 (96.3)	0.00	3121 (97.0)	3115 (96.9)	0.01
Bed mobility	6457 (85.0)	6456 (84.9)	0.00	2281 (85.9)	2281 (85.9)	0.00	2719 (84.5)	2714 (84.4)	0.00
Dressing	6897 (90.8)	6910 (90.9)	0.01	2424 (91.3)	2429 (91.5)	0.01	2933 (91.2)	2933 (91.2)	0.00
Feeding	2271 (29.9)	2276 (29.9)	0.00	815 (30.7)	820 (30.9)	0.00	995 (30.9)	987 (30.7)	0.01
Grooming	6592 (86.7)	6617 (87.1)	0.01	2305 (86.9)	2311 (87.1)	0.01	2813 (87.5)	2791 (86.8)	0.02
Locomotive	6650 (87.5)	6640 (87.4)	0.00	2308 (87.0)	2318 (87.3)	0.01	2818 (87.6)	2809 (87.3)	0.01
Toileting	6784 (89.3)	6787 (89.3)	0.00	2386 (89.9)	2378 (89.6)	0.01	2871 (89.3)	2870 (89.2)	0.00
Transferring	6531 (85.9)	6530 (85.9)	0.00	2302 (86.7)	2306 (86.9)	0.00	2774 (86.3)	2775 (86.3)	0.00
No. of ADL impairment, median (IQR)	7 (7-8)	7 (7-8)	0.00	7 (7-8)	7 (7-8)	0.05	7 (7-8)	7 (7-8)	0.00
Behavioral and cognitive impairment
Behavioral problem	712 (9.4)	700 (9.2)	0.01	248 (9.3)	250 (9.4)	0.00	297 (9.2)	299 (9.3)	0.00
Delusion	266 (3.5)	251 (3.3)	0.01	107 (4.0)	105 (4.0)	0.00	112 (3.5)	106 (3.3)	0.01
Depression	631 (8.3)	634 (8.3)	0.00	231 (8.7)	226 (8.5)	0.01	260 (8.1)	286 (8.9)	0.03
Hallucination	109 (1.4)	103 (1.4)	0.01	45 (1.7)	47 (1.8)	0.01	49 (1.5)	55 (1.7)	0.02
Severe cognitive impairment	2280 (30.0)	2280 (30.0)	0.00	781 (29.4)	774 (29.2)	0.01	906 (28.2)	901 (28.0)	0.00

Open in a new tab

Abbreviations: ACEIs, angiotensin-converting enzyme inhibitors; ADL, activities of daily living; CCI, Charlson Comorbidity Index; CFI, Claims-Based Frailty Index; CHA₂DS₂-VASc, congestive heart failure; hypertension; age 75 years or older; diabetes; prior stroke, transient ischemic attack, or thromboembolism; vascular disease; age 65 to 74 years; and sex category; COPD, chronic obstructive pulmonary disease; ED, emergency department; HAS-BLED, hypertension, abnormal kidney or liver function, stroke, bleeding, elderly, and drugs or excess alcohol use; PVD, pulmonary valvular diseases; SMD, standardized mean difference; SNF, skilled nursing facility.

^{^a}

Unless otherwise indicated.

^{^b}

Other race category includes Asian, Hispanic, North American Native, unknown, other, and missing race categories in Medicare.

^{^c}

Other bleeding event was defined as not gastrointestinal and not intracranial bleeding.

Clinical Outcomes for Apixaban Use vs Nonuse

During a median (IQR) follow-up of 149 (50-365) days, apixaban use was associated with a lower rate of ischemic stroke compared with nonuse (14.6 vs 27.6 per 1000 person-years [PYs]; ARR, 0.53; 95% CI, 0.38-0.73; ARD, −12.96; 95% CI, −19.44 to −6.49). Major bleeding rates were higher in those receiving OACs (50.1 vs 31.0 per 1000 PYs; ARR, 1.62; 95% CI, 1.29-2.03; ARD, 19.13; 95% CI, 10.05-28.21). These opposing effects resulted in no significant difference in the composite outcome (ARR, 1.08; 95% CI, 0.90-1.30; ARD, 4.86; 95% CI, −6.31 to 16.03) (Table 2; eTable 4 in Supplement 1).

Table 2. Clinical Outcomes for Oral Anticoagulant Use vs Nonuse in Long-Term Nursing Home Residents With Atrial Fibrillation and Activities of Daily Living Disability.

Outcome	No. of events (IR per 1000 PY)
	Apixaban cohort (n = 7600 pairs)				Rivaroxaban cohort (n = 2654 pairs)				Warfarin cohort (n = 3216 pairs)
	Apixaban group	Matched group	ARD (95% CI)	ARR (95% CI)	Rivaroxaban group	Matched group	ARD (95% CI)	ARR (95% CI)	Warfarin group	Matched group	ARD (95% CI)	ARR (95% CI)
Ischemic stroke
Total	54 (14.6)	109 (27.6)	−12.96 (−19.44 to −6.49)	0.53 (0.38 to 0.73)	16 (12.9)	35 (26.9)	−13.96 (−24.88 to −3.04)	0.48 (0.27 to 0.87)	14 (9.8)	30 (20.8)	−11.06 (−20.09 to −2.02)	0.47 (0.25 to 0.88)
Frail patients	25 (18.1)	30 (19.6)	−1.51 (−11.49 to 8.46)	0.92 (0.54 to 1.57)	<10 (11.6)	10 (20.3)	−8.74 (−24.89 to 7.40)	0.57 (0.19 to 1.67)	<10 (9.4)	12 (21.9)	−12.42 (−27.29 to 2.45)	0.43 (0.15 to 1.22)
Nonfrai or pre-frail patients	33 (14.6)	66 (28.0)	−13.43 (−21.81 to −5.05)	0.52 (0.34 to 0.79)	12 (15.6)	17 (22.1)	−6.48 (−20.21 to 7.26)	0.71 (0.34 to 1.48)	6 (7.2)	17 (20.0)	−12.83 (−23.93 to −1.73)	0.36 (0.14 to 0.91)
P value for heterogeneity by frailty	NA	NA	.08	.10	NA	NA	.83	.66	NA	NA	.97	.80
Major bleeding^a
Total	185 (50.1)	122 (31.0)	19.13 (10.05 to 28.21)	1.62 (1.29 to 2.03)	94 (76.5)	37 (28.5)	48.01 (30.02 to 65.99)	2.68 (1.84 to 3.93)	132 (93.0)	44 (30.8)	62.22 (43.94 to 80.51)	3.02 (2.15 to 4.25)
Frail patients	72 (52.2)	53 (34.9)	17.32 (2.04 to 32.61)	1.50 (1.05 to 2.13)	35 (81.7)	17 (34.8)	46.96 (15.23 to 78.68)	2.35 (1.32 to 4.19)	51 (97.2)	22 (40.4)	56.81 (25.25 to 88.38)	2.41 (1.46 to 3.97)
Nonfrail or prefrail patients	110 (48.7)	64 (27.2)	21.50 (10.22 to 32.79)	1.79 (1.32 to 2.44)	53 (69.7)	18 (23.5)	46.22 (24.54 to 67.89)	2.97 (1.74 to 5.07)	79 (95.3)	23 (27.2)	68.03 (44.25 to 91.80)	3.50 (2.20 to 5.56)
P value for heterogeneity by frailty	NA	NA	.67	.46	NA	NA	.97	.56	NA	NA	.58	.28
Composite of major bleeding or ischemic stroke
Total	235 (63.9)	231 (59.0)	4.86 (−6.31 to 16.03)	1.08 (0.90 to 1.30)	109 (88.8)	71 (54.9)	33.91 (12.90 to 54.92)	1.62 (1.20 to 2.18)	146 (103.0)	74 (51.9)	51.17 (30.71 to 71.64)	1.99 (1.50 to 2.63)
Frail patients	96 (69.9)	83 (54.9)	15.01 (−3.28 to 33.30)	1.27 (0.95 to 1.71)	40 (93.5)	27 (55.4)	38.12 (2.41 to 73.83)	1.69 (1.04 to 2.75)	56 (107.0)	34 (62.5)	44.45 (9.42 to 79.47)	1.71 (1.12 to 2.62)
Nonfrail or pre-frail patients	140 (62.2)	130 (55.6)	6.54 (−7.51 to 20.59)	1.12 (0.88 to 1.42)	64 (84.4)	34 (44.5)	39.83 (14.31 to 65.34)	1.89 (1.25 to 2.87)	85 (102.6)	40 (47.5)	55.10 (28.78 to 81.42)	2.16 (1.48 to 3.15)
P value for heterogeneity by frailty	NA	NA	.47	.52	NA	NA	.94	.73	NA	NA	.42	.63

Open in a new tab

Abbreviations: ARD, adjusted rate difference; ARR, adjusted rate ratio; IR, incidence rate; NA, not applicable; PY, person-years.

^{^a}

Major bleeding was defined as any intracranial bleeding, major gastrointestinal bleeding, or other extracranial bleeding requiring hospitalization.

Those receiving apixaban had an increased rate of major gastrointestinal bleeding (ARR, 1.61; 95% CI, 1.19-2.17; ARD, 10.97; 95% CI, 4.08-17.86) but no statistically significant increase in intracranial bleeding (ARR, 1.52; 95% CI, 0.96-2.40; ARD, 4.05; 95% CI, −0.41 to 8.50). Apixaban use was associated with lower all-cause mortality rates (ARR, 0.90; 95% CI, 0.85-0.96; ARD, −57.74; 95% CI, −91.17 to −24.30) (Figure 1). In subgroup analyses, the association between apixaban use and lower ischemic stroke rates appeared attenuated among frail individuals (ARR, 0.92; 95% CI, 0.54-1.57; ARD, −1.51; 95% CI, −11.49 to 8.46), although evidence of heterogeneity was limited (P for heterogeneity = .08) (Table 2).

Data figure with three panels of A R R dot plots for bleeding and death outcomes. Three stacked panels labeled A, B, and C, each combining a left table and a right forest style dot plot. Panel A title at upper left: Apixaban use vs nonuse. The table columns read: Outcomes; No. of events (I R per 1000 P Y) with subcolumns Apixaban group and Matched group; and A R R (95% C I). Row 1 Major gastrointestinal bleeding: 107 (28.9), 71 (18.0), A R R 1.61 (1.19-2.17). Row 2 Intracranial bleeding: 44 (11.9), 31 (7.8), A R R 1.52 (0.96-2.40). Row 3 Death: 1971 (530.9), 2341 (588.6), A R R 0.90 (0.85-0.96). To the right, a horizontal axis labeled A R R (95% C I) with tick marks at 0, 1, 2, 3 and a vertical dashed reference line at 1.0. Headings above the plot read Favors apixaban on the left of the dashed line and Favors nonuse on the right. Three dark square markers with horizontal confidence interval lines align to the three outcomes; the death marker lies left of 1, the two bleeding markers lie right of 1, with the intracranial interval crossing 1. Panel B title: Rivaroxaban use vs nonuse. Table rows: Major gastrointestinal bleeding 65 (52.7), 19 (14.6), A R R 3.62 (2.17-6.03); Intracranial bleeding 15 (12.1), 13 (10.0), A R R 1.21 (0.58-2.55); Death 593 (477.8), 692 (528.9), A R R 0.90 (0.81-1.01). The right plot repeats the same axis and dashed line; the major gastrointestinal bleeding confidence interval extends beyond the right edge and ends with a right pointing arrow. Panel C title: Warfarin use vs nonuse. Table rows: Major gastrointestinal bleeding 82 (57.6), 28 (19.5), A R R 2.95 (1.92-4.53); Intracranial bleeding 35 (24.4), 11 (7.6), A R R 3.20 (1.62-6.30); Death 786 (547.2), 679 (470.2), A R R 1.16 (1.05-1.29). The right plot repeats the same formatting; the intracranial bleeding confidence interval extends beyond the right edge with a right pointing arrow, and the death marker lies just right of 1. — Secondary outcomes included major gastrointestinal bleeding, intracranial bleeding, and all-cause mortality. The vertical dashed line indicates a rate ratio of 1.0. IR indicates incidence rate; PY, person-years.

Clinical Outcomes for Rivaroxaban Use vs Nonuse

During a median (IQR) follow-up of 126 (48-365) days, those receiving rivaroxaban had a lower rate of ischemic stroke than those not receiving rivaroxaban (12.9 vs 26.9 per 1000 PYs; ARR, 0.48; 95% CI, 0.27-0.87; ARD, −13.96; 95% CI, −24.88 to −3.04). The rate of major bleeding was higher among those receiving rivaroxaban vs not (76.5 vs 28.5 per 1000 PYs; ARR, 2.68; 95% CI, 1.84-3.93; ARD, 48.01; 95% CI, 30.02-65.99). The increased bleeding rate outweighed the stroke reduction, resulting in an increased rate of composite outcome (ARR, 1.62; 95% CI, 1.20-2.18; ARD, 33.91; 95% CI, 12.90-54.92).

Rivaroxaban use was associated with a substantially higher rate of major gastrointestinal bleeding (ARR, 3.62; 95% CI, 2.17-6.03; ARD, 38.13; 95% CI, 23.74-52.52) and no significant difference in intracranial bleeding (ARR, 1.21; 95% CI, 0.58-2.55; ARD, 2.14; 95% CI, −6.04 to 10.31). Rivaroxaban use was not associated with lower all-cause mortality (ARR, 0.90; 95% CI, 0.81-1.01; ARD, −51.13; 95% CI, −106.19 to 3.93) (Table 2 and Figure 1; eTable 4 in Supplement 1).

Clinical Outcomes for Warfarin Use vs Nonuse

During a median (IQR) follow-up of 110 (42-334) days, those receiving warfarin had a lower rate of ischemic stroke than those not receiving warfarin (9.8 vs 20.8 per 1000 PYs; ARR, 0.47; 95% CI, 0.25-0.88; ARD, −11.06; 95% CI, −20.09 to −2.02). They also experienced a higher rate of major bleeding (93.0 vs 30.8 per 1000 PYs; ARR, 3.02; 95% CI, 2.15-4.25; ARD, 62.22; 95% CI, 43.94-80.51). Overall, warfarin use was associated with an increased rate of composite outcome (ARR, 1.99; 95% CI, 1.50-2.63; ARD, 51.17; 95% CI, 30.71-71.64). Warfarin was also associated with increased rate of major gastrointestinal bleeding (ARR, 2.95; 95% CI, 1.92-4.53; ARD, 38.03; 95% CI, 23.63-52.43), intracranial bleeding (ARR, 3.20; 95% CI, 1.62-6.30; ARD, 16.78; 95% CI, 7.52-26.04) and all-cause mortality (ARR, 1.16; 95% CI, 1.05-1.29; ARD, 76.95; 95% CI, 24.86-129.05) (Table 2 and Figure 1; eTable 4 in Supplement 1).

ITT Analysis

Compared with the primary as-treated analyses, the 180-day ITT analyses showed attenuated associations. The reductions in ischemic stroke were no longer statistically significant, whereas the increased risks of major bleeding and the composite outcome remained significant across all 3 OAC cohorts. The lower mortality rate associated with apixaban was also attenuated (eTable 6 in Supplement 1).

Direct Active-Comparator Analyses

In the active-comparator analyses, apixaban was associated with similar ischemic stroke rates but lower rates of major bleeding and the composite outcome rates than rivaroxaban and warfarin. Compared with warfarin, apixaban was associated with lower rates of intracranial bleeding and all-cause mortality, whereas compared with rivaroxaban, it was associated with lower rates of major gastrointestinal bleeding (Table 3).

Table 3. Direct Active-Comparator Analyses of Oral Anticoagulant Use in Long-Term Nursing Home Residents With Atrial Fibrillation and Activities of Daily Living Disability.

Outcome	Apixaban vs rivaroxaban cohorts (n = 3180 pairs)				Apixaban vs warfarin cohorts (n = 3193 pairs)
	No. of events (IR per 1000 PY)				No. of events (IR per 1000 PY)
	Apixaban	Rivaroxaban	ARD (95% CI)	ARR (95% CI)	Apixaban	Warfarin	ARD (95% CI)	ARR (95% CI)
Ischemic stroke	26 (15.8)	22 (14.6)	1.2 (−7.4 to 9.9)	1.08 (0.61 to 1.92)	20 (12.1)	13 (9.2)	2.9 (−4.4 to 10.2)	1.31 (0.65 to 2.65)
Major bleeding^a	64 (39.0)	114 (76.2)	−37.2 (−54.4 to −20.0)	0.51 (0.38 to 0.70)	64 (38.7)	135 (96.3)	−57.7 (−76.7 to −38.7)	0.40 (0.30 to 0.54)
Composite of major bleeding or ischemic stroke	88 (53.8)	135 (90.4)	−36.6 (−55.9 to −17.3)	0.60 (0.45 to 0.78)	84 (50.9)	148 (105.8)	−55.0 (−75.3 to −34.6)	0.48 (0.37 to 0.63)

Open in a new tab

Abbreviations: ARD, adjusted rate difference; ARR, adjusted rate ratio; IR, incidence rate; PY, person-years.

Tipping-Point Analysis

We evaluated the weighted benefit-risk balance across increasing relative weights assigned to ischemic stroke vs major bleeding. Increasing the relative weight assigned to ischemic stroke shifted the benefit-risk balance in favor of anticoagulation. The tipping point at which the benefit from stroke reduction outweighed increased bleeding risk was 1.48 for apixaban, 3.44 for rivaroxaban, and 5.63 for warfarin (Figure 2).

Three-panel dot plots of A R D per 1000 P Y versus relative weight for three drugs. Three side-by-side panels labeled A, B, and C at the upper left of each plot. All panels share a horizontal axis labeled Relative weight with tick marks from 1 through 10, and a vertical axis labeled A R D per 1000 P Y with labeled values from 40 at the top down to minus 100 at the bottom, with a solid black horizontal line at 0. Each panel contains dark teal circular points connected by a thin black line, with vertical black error bars at each point. Along the bottom at A R D minus 100, a row of green circles appears at each relative weight value from 1 to 10. An orange dashed horizontal line overlays the 0 level, and an orange circle with an orange vertical dashed line marks the x position where the plotted A R D crosses 0. Panel A title text reads Apixaban use vs nonuse; the orange tipping-point marker sits near relative weight 2, with the teal series near 0 at weight 1 then decreasing to approximately minus 65 by weight 10. Panel B title text reads Rivaroxaban use vs nonuse; the orange tipping-point marker sits between weights 3 and 4, with the teal series around 15 at weight 1, near 0 at weight 3, and decreasing to about minus 45 by weight 10, with wider error bars at higher weights. Panel C title text reads Warfarin use vs nonuse; the orange tipping-point marker sits near weight 6, with the teal series around 20 at weight 1, near 0 at weight 6, and decreasing to roughly minus 25 to minus 30 by weight 10, with error bars extending farther at higher weights. — The relative weight (clinical impact) of ischemic stroke vs major bleeding ranges from 1 to 10. Blue circles indicate the adjusted rate difference (ARD) at each relative weight, and error bars indicate 95% CIs. Green circles indicate the relative weights evaluated. The orange horizontal dashed line indicates an ARD of 0. The orange circle and vertical dashed line indicate the tipping point at which the estimated ARD crosses 0; weighting ischemic stroke more heavily than this threshold shifts the composite outcome toward a clinical benefit of apixaban, rivaroxaban, or warfarin compared with nonuse. PY indicates person-years.

Sensitivity Analyses

Sensitivity analyses, including the 365-day ITT analysis, high-dimensional PS adjustment, IPCW, Fine and Gray competing-risk models, and the health care-encounter restriction, yielded findings consistent with the primary analyses (eTable 5 in Supplement 1). Absolute cumulative incidence estimates at 180 and 365 days after CED from the Fine and Gray competing-risk analyses are presented in eFigure 2 in Supplement 1. Dose-stratified analyses were generally consistent with the primary findings, although estimates for rivaroxaban were less precise because of smaller sample sizes (eTable 7 in Supplement 1).

Discussion

In this large cohort study of nursing home residents with AF and ADL disability, OAC use was associated with lower ischemic stroke rates but higher major bleeding rates compared with nonuse across all 3 OAC cohorts. The reduction in ischemic stroke rates associated with apixaban appeared attenuated among frail residents, although evidence of heterogeneity by frailty was limited. Among the OACs evaluated, apixaban demonstrated the most favorable benefit-risk profile, with anticoagulation becoming favorable when ischemic stroke was weighted only 1.5 times more than major bleeding. In contrast, substantially greater weighting was required for rivaroxaban and warfarin. The favorable results for apixaban were further supported by direct comparisons with warfarin and rivaroxaban for which rates of major bleeding were substantially lower with apixaban, whereas the rates of ischemic stroke were similar across the OACs.

Our findings are consistent with prior observational studies^14,52 showing that OAC use reduces the ischemic stroke risk at the cost of increased bleeding risk in older adults with AF, including those residing in US long-term care facilities. Our study extends these findings by evaluating a contemporary national cohort of US nursing home residents with detailed functional status from MDS data and by directly comparing multiple OACs. Prior studies^13,15,16,53 have also reported higher rates of OAC discontinuation among residents with ADL disability. Together, these findings highlight the clinical complexity of anticoagulation decisions in nursing home residents, whose limited life expectancy and increased susceptibility to bleeding may attenuate the overall net clinical benefit of OAC therapy.^54,55

In our study, apixaban use was not associated with a significant increase in the composite outcome. This result contrasts with the AVERROES trial,⁵⁶ which demonstrated a clear net benefit of apixaban over aspirin use among patients with AF who were not candidates to receive warfarin. Compared with our cohort, the AVERROES trial population was substantially younger (mean age of approximately 70 years vs 82 years in our cohort); did not include measures of functional impairment, such as ADL disability; and had much lower rates of major bleeding (approximately 12-14 per 1000 PYs) than we observed (50.1-93.0 per 1000 PYs). These differences may explain the attenuated net benefit in our study and suggest treatment effects differ in frail populations. Consistent with prior observational studies,^54,57 our results are in line with the hypothesis that the net clinical benefit of OACs may diminish with advancing age and frailty.

The complementary 180-day ITT analyses showed attenuated effect estimates, although the overall pattern of findings remained similar. The associations with ischemic stroke were no longer statistically significant, whereas the increased risks of major bleeding and the composite outcome remained significant across all 3 OAC cohorts, supporting the robustness of the primary as-treated findings. The reduction in all-cause mortality associated with apixaban was also attenuated, suggesting that this finding should be interpreted cautiously because all-cause mortality is generally more susceptible to residual confounding than disease-specific outcomes in observational studies.⁵⁸ Direct active-comparator analyses demonstrated similar ischemic stroke rates but lower major bleeding and composite outcome rates with apixaban than with rivaroxaban and warfarin, providing complementary evidence from comparisons that are less susceptible to treatment-selection bias among anticoagulant users. IPCW that addressed informative censoring and the health care encounter sensitivity analysis also yielded findings consistent with the primary analyses, supporting the robustness of our results.

Previous studies^{59,60,61,62,63} have consistently shown that ischemic stroke is associated with greater mortality and long-term disability than most major extracranial bleeding events and that patients generally place greater value on stroke prevention than avoiding non-intracranial bleeding. Consistent with these observations, our tipping-point analyses showed that only a 1.5-fold greater weighting of ischemic stroke than major bleeding was required for the benefit-risk balance to favor anticoagulation with apixaban compared with 3.4-fold for rivaroxaban and 5.6-fold for warfarin. Current guidelines do not specifically address anticoagulation in nursing home residents with ADL disability. Nevertheless, our findings suggest that the overall benefit-harm balance differs across OACs.⁷

Strengths and Limitations

Our study has several strengths. First, we used a target trial emulation framework, which aligns observational analyses with key design principles of randomized clinical trials and has been validated against randomized clinical trial benchmarks.^24,25 Second, we addressed an important evidence gap by studying nursing home residents with AF and ADL disability, a population that has been largely underrepresented in randomized clinical trials. Third, linkage with the MDS 3.0 enabled adjustment for detailed geriatric characteristics, including cognitive impairment and functional status, that are unavailable in claims data alone.

However, several limitations of this study also warrant consideration. First, our analysis focused on 3 commonly used OACs (ie, warfarin, apixaban, and rivaroxaban). Edoxaban and dabigatran were not evaluated because their use was infrequent among long-term nursing home residents with ADL disability during our study period, resulting in insufficient sample sizes for reliable comparative analyses.^64,65 Second, despite using a design that focuses on new use of OACs, PS matching with 111 baseline covariates, and multiple sensitivity analyses, residual confounding by indication and informative censoring cannot be completely eliminated because anticoagulation decisions in nursing home residents are highly individualized and influenced by factors that are incompletely captured in claims data, including patient preferences, anticipated life expectancy, and goals of care. Third, although we incorporated a validated claims-based prediction of TTR, observed international normalized ratio values and actual TTR were unavailable. Therefore, the warfarin findings should be interpreted as reflecting routine clinical practice rather than optimal anticoagulation control. Fourth, frailty was defined using a claims-based index, which may have resulted in misclassification. Finally, because our study population consisted of Medicare beneficiaries only, the findings may not be generalized to uninsured or commercially insured populations.

Conclusions

Among long-term nursing home residents with AF and ADL disability, OAC use was associated with lower ischemic stroke rates but higher major bleeding rates compared with nonuse. Among the 3 OACs evaluated, apixaban demonstrated the most favorable benefit-risk profile, supported by both tipping-point and direct active-comparator analyses. These findings provide evidence to inform individualized anticoagulation decisions in this at-risk population and may also help guide OAC selection when anticoagulation is considered appropriate.

Supplement 1.

eFigure 1. Inclusion and Exclusion Criteria of the Three OAC Cohorts (i.e. Warfarin, Apixaban and Rivaroxaban Users vs Non-Users Cohorts) among Long-term Nursing Home Residents With Atrial Fibrillation and ADL Disability

eFigure 2. Cumulative Incidence Function for Major Bleeding Comparing Apixaban Users and Matched Non-users, Accounting for the Competing Risk of Death

eTable 1. Target Trial Emulation Table

eTable 2. Cohorts Selection Among Long-term Nursing Home Residents With Atrial Fibrillation and ADL Disability

eTable 3. Patient Characteristics of the OAC Users (i.e., Warfarin, Apixaban, Rivaroxaban) vs. Nonuser Among Long-term Nursing Home Residents With Atrial Fibrillation and ADL Disability

eTable 4. Follow-up and Censoring Reasons After 1:1 Propensity Score Matching among Long-term Nursing Home Residents With Atrial Fibrillation and ADL Disability

eTable 5. Primary vs Sensitivity Analyses Comparing the Primary Outcomes Among Long-term Nursing Home Residents With Atrial Fibrillation and ADL Disability

eTable 6. Adjusted Rate Ratios and Rate Differences of Major Bleeding and Ischemic Stroke in the 180-Day Intention-to-Treat (ITT) Analyses Comparing Propensity Score–Matched Oral Anticoagulant Users vs Non-users Among Long-term Nursing Home Residents With Atrial Fibrillation and ADL Disability

eTable 7. Adjusted Rate Ratios and Rate Differences of Major Bleeding, Ischemic Stroke, and Composite Outcome Comparing Standard- and Reduced-Dose Oral Anticoagulant Users with Non-users among Long-term Nursing Home Residents With Atrial Fibrillation and ADL Disability

eTable 8. Variable Definitions

jamanetwopen-e2637533-s001.pdf^{(1.1MB, pdf)}

Supplement 2.

Data Sharing Statement

jamanetwopen-e2637533-s002.pdf^{(13.4KB, pdf)}

References

1.Noubiap JJ, Tang JJ, Teraoka JT, Dewland TA, Marcus GM. Minimum national prevalence of diagnosed atrial fibrillation inferred from California acute care facilities. J Am Coll Cardiol. 2024;84(16):1501-1508. doi: 10.1016/j.jacc.2024.07.014 [DOI] [PubMed] [Google Scholar]
2.Reardon G, Nelson WW, Patel AA, Philpot T, Neidecker M. Prevalence of atrial fibrillation in US nursing homes: results from the National Nursing Home Survey, 1985-2004. J Am Med Dir Assoc. 2012;13(6):529-534. doi: 10.1016/j.jamda.2012.03.007 [DOI] [PubMed] [Google Scholar]
3.Wagner G, Smeikal M, Gisinger C, et al. Epidemiology, risk profile, management, and outcome in geriatric patients with atrial fibrillation in two long-term care hospitals. Sci Rep. 2022;12(1):18725. doi: 10.1038/s41598-022-22013-6 [DOI] [PMC free article] [PubMed] [Google Scholar]
4.Alcusky M, Lapane KL. Treatment of atrial fibrillation in nursing homes: a place for direct acting oral anticoagulants? J Nurs Home Res Sci. 2018;4:15-19. doi: 10.14283/jnhrs.2018.4 [DOI] [PMC free article] [PubMed] [Google Scholar]
5.Naccarelli GV, Varker H, Lin J, Schulman KL. Increasing prevalence of atrial fibrillation and flutter in the United States. Am J Cardiol. 2009;104(11):1534-1539. doi: 10.1016/j.amjcard.2009.07.022 [DOI] [PubMed] [Google Scholar]
6.Buck J, Fromings Hill J, Martin A, et al. Reasons for discontinuing oral anticoagulation therapy for atrial fibrillation: a systematic review. Age Ageing. 2021;50(4):1108-1117. doi: 10.1093/ageing/afab024 [DOI] [PMC free article] [PubMed] [Google Scholar]
7.Joglar JA, Chung MK, Armbruster AL, et al. ; Peer Review Committee Members . 2023 ACC/AHA/ACCP/HRS Guideline for the Diagnosis and Management of Atrial Fibrillation: a report of the American College of Cardiology/American Heart Association Joint Committee on Clinical Practice Guidelines. Circulation. 2024;149(1):e1-e156. doi: 10.1161/CIR.0000000000001193 [DOI] [PMC free article] [PubMed] [Google Scholar]
8.Parks AL, Jeon SY, Boscardin WJ, et al. Long-term individual and population functional outcomes in older adults with atrial fibrillation. J Am Geriatr Soc. 2021;69(6):1570-1578. doi: 10.1111/jgs.17087 [DOI] [PMC free article] [PubMed] [Google Scholar]
9.Stuck AE, Walthert JM, Nikolaus T, Büla CJ, Hohmann C, Beck JC. Risk factors for functional status decline in community-living elderly people: a systematic literature review. Soc Sci Med. 1999;48(4):445-469. doi: 10.1016/S0277-9536(98)00370-0 [DOI] [PubMed] [Google Scholar]
10.Inouye SK, Studenski S, Tinetti ME, Kuchel GA. Geriatric syndromes: clinical, research, and policy implications of a core geriatric concept. J Am Geriatr Soc. 2007;55(5):780-791. doi: 10.1111/j.1532-5415.2007.01156.x [DOI] [PMC free article] [PubMed] [Google Scholar]
11.Fried LP, Ferrucci L, Darer J, Williamson JD, Anderson G. Untangling the concepts of disability, frailty, and comorbidity: implications for improved targeting and care. J Gerontol A Biol Sci Med Sci. 2004;59(3):255-263. doi: 10.1093/gerona/59.3.M255 [DOI] [PubMed] [Google Scholar]
12.Mailhot T, McManus DD, Waring ME, et al. Frailty, cognitive impairment, and anticoagulation among older adults with nonvalvular atrial fibrillation. J Am Geriatr Soc. 2020;68(12):2778-2786. doi: 10.1111/jgs.16756 [DOI] [PMC free article] [PubMed] [Google Scholar]
13.Shah SJ, Fang MC, Jeon SY, Gregorich SE, Covinsky KE. Geriatric syndromes and atrial fibrillation: prevalence and association with anticoagulant use in a national cohort of older Americans. J Am Geriatr Soc. 2021;69(2):349-356. doi: 10.1111/jgs.16822 [DOI] [PMC free article] [PubMed] [Google Scholar]
14.Chen Q, Baek J, Goldberg R, Tjia J, Lapane K, Alcusky M. To treat or not to treat: a comparative effectiveness analysis of oral anticoagulant outcomes among U.S. nursing home residents with atrial fibrillation. BMC Geriatr. 2024;24(1):619. doi: 10.1186/s12877-024-05186-9 [DOI] [PMC free article] [PubMed] [Google Scholar]
15.Zhang N, Patel J, Chen Z, et al. Geriatric conditions are associated with decreased anticoagulation use in long-term care residents with atrial fibrillation. J Am Heart Assoc. 2021;10(16):e021293. doi: 10.1161/JAHA.121.021293 [DOI] [PMC free article] [PubMed] [Google Scholar]
16.Kapoor A, Foley G, Zhang N, et al. Geriatric conditions predict discontinuation of anticoagulation in long-term care residents with atrial fibrillation. J Am Geriatr Soc. 2020;68(4):717-724. doi: 10.1111/jgs.16335 [DOI] [PubMed] [Google Scholar]
17.Chen Q, Baek J, Goldberg R, Tjia J, Lapane K, Alcusky M. Discontinuation of oral anticoagulant use among nursing home residents with atrial fibrillation before hospice enrollment. J Am Geriatr Soc. 2023;71(10):3071-3085. doi: 10.1111/jgs.18512 [DOI] [PMC free article] [PubMed] [Google Scholar]
18.Heshmatollah A, Mutlu U, Koudstaal PJ, Ikram MA, Ikram MK. Cognitive and physical impairment and the risk of stroke - a prospective cohort study. Sci Rep. 2020;10(1):6274. doi: 10.1038/s41598-020-63295-y [DOI] [PMC free article] [PubMed] [Google Scholar]
19.Lee M, Saver JL, Hong KS, et al. Cognitive impairment and risk of future stroke: a systematic review and meta-analysis. CMAJ. 2014;186(14):E536-E546. doi: 10.1503/cmaj.140147 [DOI] [PMC free article] [PubMed] [Google Scholar]
20.Murakami K, Tsubota-Utsugi M, Satoh M, et al. Impaired higher-level functional capacity as a predictor of stroke in community-dwelling older adults: the Ohasama Study. Stroke. 2016;47(2):323-328. doi: 10.1161/STROKEAHA.115.011131 [DOI] [PubMed] [Google Scholar]
21.Gage BF, Birman-Deych E, Kerzner R, Radford MJ, Nilasena DS, Rich MW. Incidence of intracranial hemorrhage in patients with atrial fibrillation who are prone to fall. Am J Med. 2005;118(6):612-617. doi: 10.1016/j.amjmed.2005.02.022 [DOI] [PubMed] [Google Scholar]
22.Shih RD, Ouslander JG. Intracranial hemorrhage in older adults: implications for fall risk assessment and prevention. J Am Geriatr Soc. 2020;68(5):953-955. doi: 10.1111/jgs.16399 [DOI] [PubMed] [Google Scholar]
23.Mitchell A, Elmasry Y, van Poelgeest E, Welsh TJ. Anticoagulant use in older persons at risk for falls: therapeutic dilemmas-a clinical review. Eur Geriatr Med. 2023;14(4):683-696. doi: 10.1007/s41999-023-00811-z [DOI] [PMC free article] [PubMed] [Google Scholar]
24.Hernán MA, Wang W, Leaf DE. Target trial emulation: a framework for causal inference from observational data. JAMA. 2022;328(24):2446-2447. doi: 10.1001/jama.2022.21383 [DOI] [PubMed] [Google Scholar]
25.Wang SV, Schneeweiss S, Franklin JM, et al. ; RCT-DUPLICATE Initiative . Emulation of randomized clinical trials with nonrandomized database analyses: results of 32 clinical trials. JAMA. 2023;329(16):1376-1385. doi: 10.1001/jama.2023.4221 [DOI] [PMC free article] [PubMed] [Google Scholar]
26.Jones AL, Dwyer LL, Bercovitz AR, Strahan GW. The National Nursing Home Survey: 2004 overview. Vital Health Stat 13. 2009;(167):1-155. [PubMed] [Google Scholar]
27.Fong JH, Mitchell OS, Koh BSK. Disaggregating activities of daily living limitations for predicting nursing home admission. Health Serv Res. 2015;50(2):560-578. doi: 10.1111/1475-6773.12235 [DOI] [PMC free article] [PubMed] [Google Scholar]
28.Gaugler JE, Duval S, Anderson KA, Kane RL. Predicting nursing home admission in the U.S: a meta-analysis. BMC Geriatr. 2007;7:13. doi: 10.1186/1471-2318-7-13 [DOI] [PMC free article] [PubMed] [Google Scholar]
29.Richardson DB. An incidence density sampling program for nested case-control analyses. Occup Environ Med. 2004;61(12):e59. doi: 10.1136/oem.2004.014472 [DOI] [PMC free article] [PubMed] [Google Scholar]
30.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]
31.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]
32.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]
33.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]
34.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]
35.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]
36.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]
37.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]
38.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]
39.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]
40.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]
41.Sison SDM, Shi SM, Oh G, Jeong S, McCarthy EP, Kim DH. Claims-based frailty index and its relationship with commonly used clinical frailty measures. J Gerontol A Biol Sci Med Sci. 2024;79(7):glae094. doi: 10.1093/gerona/glae094 [DOI] [PMC free article] [PubMed] [Google Scholar]
42.Lin KJ, Singer DE, Glynn RJ, et al. Prediction score for anticoagulation control quality among older adults. J Am Heart Assoc. 2017;6(10):e006814. doi: 10.1161/JAHA.117.006814 [DOI] [PMC free article] [PubMed] [Google Scholar]
43.Xu S, Ross C, Raebel MA, Shetterly S, Blanchette C, Smith D. Use of stabilized inverse propensity scores as weights to directly estimate relative risk and its confidence intervals. Value Health. 2010;13(2):273-277. doi: 10.1111/j.1524-4733.2009.00671.x [DOI] [PMC free article] [PubMed] [Google Scholar]
44.de Bakker PIW, Ferreira MAR, Jia X, Neale BM, Raychaudhuri S, Voight BF. Practical aspects of imputation-driven meta-analysis of genome-wide association studies. Hum Mol Genet. 2008;17(R2):R122-R128. doi: 10.1093/hmg/ddn288 [DOI] [PMC free article] [PubMed] [Google Scholar]
45.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]
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(1):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.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]
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.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]
52.Liu X, Baumgarten M, Smith G, et al. Warfarin usage among elderly atrial fibrillation patients with traumatic injury, an analysis of United States Medicare fee-for-service enrollees. J Clin Pharmacol. 2015;55(1):25-32. doi: 10.1002/jcph.375 [DOI] [PMC free article] [PubMed] [Google Scholar]
53.Chen Q, Lapane K, Nunes AP, Tjia J, Hugunin J, Alcusky M. Prevalence and the factors associated with oral anticoagulant use among nursing home residents. J Clin Pharm Ther. 2021;46(6):1714-1728. doi: 10.1111/jcpt.13508 [DOI] [PMC free article] [PubMed] [Google Scholar]
54.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]
55.Parks AL, Frankel DS, Kim DH, et al. Management of atrial fibrillation in older adults. BMJ. 2024;386:e076246. doi: 10.1136/bmj-2023-076246 [DOI] [PubMed] [Google Scholar]
56.Connolly SJ, Eikelboom J, Joyner C, et al. ; AVERROES Steering Committee and Investigators . Apixaban in patients with atrial fibrillation. N Engl J Med. 2011;364(9):806-817. doi: 10.1056/NEJMoa1007432 [DOI] [PubMed] [Google Scholar]
57.Søgaard M, Jensen M, Højen AA, et al. Net clinical benefit of oral anticoagulation among frail patients with atrial fibrillation: nationwide cohort study. Stroke. 2024;55(2):413-422. doi: 10.1161/STROKEAHA.123.044407 [DOI] [PubMed] [Google Scholar]
58.Krüger N, Schneeweiss S, Desai RJ, et al. Cardiovascular outcomes of semaglutide and tirzepatide for patients with type 2 diabetes in clinical practice. Nat Med. 2026;32(1):342-352. doi: 10.1038/s41591-025-04102-x [DOI] [PMC free article] [PubMed] [Google Scholar]
59.Connolly SJ, Eikelboom JW, Ng J, et al. ; ACTIVE (Atrial Fibrillation Clopidogrel Trial with Irbesartan for Prevention of Vascular Events) Steering Committee and Investigators . Net clinical benefit of adding clopidogrel to aspirin therapy in patients with atrial fibrillation for whom vitamin K antagonists are unsuitable. Ann Intern Med. 2011;155(9):579-586. doi: 10.7326/0003-4819-155-9-201111010-00004 [DOI] [PubMed] [Google Scholar]
60.Fang MC, Go AS, Chang Y, et al. Long-term survival after ischemic stroke in patients with atrial fibrillation. Neurology. 2014;82(12):1033-1037. doi: 10.1212/WNL.0000000000000248 [DOI] [PMC free article] [PubMed] [Google Scholar]
61.Raposeiras-Roubín S, Abu-Assi E, Cespón Fernández M, et al. Trade-off between the effects of embolic versus bleeding events on mortality in elderly patients with atrial fibrillation. Rev Esp Cardiol (Engl Ed). 2022;75(4):334-342. doi: 10.1016/j.rec.2021.02.003 [DOI] [PubMed] [Google Scholar]
62.Wilke T, Bauer S, Mueller S, Kohlmann T, Bauersachs R. Patient preferences for oral anticoagulation therapy in atrial fibrillation: a systematic literature review. Patient. 2017;10(1):17-37. doi: 10.1007/s40271-016-0185-9 [DOI] [PMC free article] [PubMed] [Google Scholar]
63.Devereaux PJ, Anderson DR, Gardner MJ, et al. Differences between perspectives of physicians and patients on anticoagulation in patients with atrial fibrillation: observational study. BMJ. 2001;323(7323):1218-1222. doi: 10.1136/bmj.323.7323.1218 [DOI] [PMC free article] [PubMed] [Google Scholar]
64.Ko D, Lin KJ, Bessette LG, et al. Trends in use of oral anticoagulants in older adults with newly diagnosed atrial fibrillation, 2010-2020. JAMA Netw Open. 2022;5(11):e2242964. doi: 10.1001/jamanetworkopen.2022.42964 [DOI] [PMC free article] [PubMed] [Google Scholar]
65.getPIContent.pdf. Accessed December 5, 2024. https://daiichisankyo.us/prescribing-information-portlet/getPIContent?productName=Savaysa&inline=true

Associated Data

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

Supplementary Materials

Supplement 1.

eFigure 2. Cumulative Incidence Function for Major Bleeding Comparing Apixaban Users and Matched Non-users, Accounting for the Competing Risk of Death

eTable 1. Target Trial Emulation Table

eTable 2. Cohorts Selection Among Long-term Nursing Home Residents With Atrial Fibrillation and ADL Disability

eTable 3. Patient Characteristics of the OAC Users (i.e., Warfarin, Apixaban, Rivaroxaban) vs. Nonuser Among Long-term Nursing Home Residents With Atrial Fibrillation and ADL Disability

eTable 4. Follow-up and Censoring Reasons After 1:1 Propensity Score Matching among Long-term Nursing Home Residents With Atrial Fibrillation and ADL Disability

eTable 5. Primary vs Sensitivity Analyses Comparing the Primary Outcomes Among Long-term Nursing Home Residents With Atrial Fibrillation and ADL Disability

eTable 8. Variable Definitions

jamanetwopen-e2637533-s001.pdf^{(1.1MB, pdf)}

Supplement 2.

Data Sharing Statement

jamanetwopen-e2637533-s002.pdf^{(13.4KB, pdf)}

[zoi261009r1] 1.Noubiap JJ, Tang JJ, Teraoka JT, Dewland TA, Marcus GM. Minimum national prevalence of diagnosed atrial fibrillation inferred from California acute care facilities. J Am Coll Cardiol. 2024;84(16):1501-1508. doi: 10.1016/j.jacc.2024.07.014 [DOI] [PubMed] [Google Scholar]

[zoi261009r2] 2.Reardon G, Nelson WW, Patel AA, Philpot T, Neidecker M. Prevalence of atrial fibrillation in US nursing homes: results from the National Nursing Home Survey, 1985-2004. J Am Med Dir Assoc. 2012;13(6):529-534. doi: 10.1016/j.jamda.2012.03.007 [DOI] [PubMed] [Google Scholar]

[zoi261009r3] 3.Wagner G, Smeikal M, Gisinger C, et al. Epidemiology, risk profile, management, and outcome in geriatric patients with atrial fibrillation in two long-term care hospitals. Sci Rep. 2022;12(1):18725. doi: 10.1038/s41598-022-22013-6 [DOI] [PMC free article] [PubMed] [Google Scholar]

[zoi261009r4] 4.Alcusky M, Lapane KL. Treatment of atrial fibrillation in nursing homes: a place for direct acting oral anticoagulants? J Nurs Home Res Sci. 2018;4:15-19. doi: 10.14283/jnhrs.2018.4 [DOI] [PMC free article] [PubMed] [Google Scholar]

[zoi261009r5] 5.Naccarelli GV, Varker H, Lin J, Schulman KL. Increasing prevalence of atrial fibrillation and flutter in the United States. Am J Cardiol. 2009;104(11):1534-1539. doi: 10.1016/j.amjcard.2009.07.022 [DOI] [PubMed] [Google Scholar]

[zoi261009r6] 6.Buck J, Fromings Hill J, Martin A, et al. Reasons for discontinuing oral anticoagulation therapy for atrial fibrillation: a systematic review. Age Ageing. 2021;50(4):1108-1117. doi: 10.1093/ageing/afab024 [DOI] [PMC free article] [PubMed] [Google Scholar]

[zoi261009r7] 7.Joglar JA, Chung MK, Armbruster AL, et al. ; Peer Review Committee Members . 2023 ACC/AHA/ACCP/HRS Guideline for the Diagnosis and Management of Atrial Fibrillation: a report of the American College of Cardiology/American Heart Association Joint Committee on Clinical Practice Guidelines. Circulation. 2024;149(1):e1-e156. doi: 10.1161/CIR.0000000000001193 [DOI] [PMC free article] [PubMed] [Google Scholar]

[zoi261009r8] 8.Parks AL, Jeon SY, Boscardin WJ, et al. Long-term individual and population functional outcomes in older adults with atrial fibrillation. J Am Geriatr Soc. 2021;69(6):1570-1578. doi: 10.1111/jgs.17087 [DOI] [PMC free article] [PubMed] [Google Scholar]

[zoi261009r9] 9.Stuck AE, Walthert JM, Nikolaus T, Büla CJ, Hohmann C, Beck JC. Risk factors for functional status decline in community-living elderly people: a systematic literature review. Soc Sci Med. 1999;48(4):445-469. doi: 10.1016/S0277-9536(98)00370-0 [DOI] [PubMed] [Google Scholar]

[zoi261009r10] 10.Inouye SK, Studenski S, Tinetti ME, Kuchel GA. Geriatric syndromes: clinical, research, and policy implications of a core geriatric concept. J Am Geriatr Soc. 2007;55(5):780-791. doi: 10.1111/j.1532-5415.2007.01156.x [DOI] [PMC free article] [PubMed] [Google Scholar]

[zoi261009r11] 11.Fried LP, Ferrucci L, Darer J, Williamson JD, Anderson G. Untangling the concepts of disability, frailty, and comorbidity: implications for improved targeting and care. J Gerontol A Biol Sci Med Sci. 2004;59(3):255-263. doi: 10.1093/gerona/59.3.M255 [DOI] [PubMed] [Google Scholar]

[zoi261009r12] 12.Mailhot T, McManus DD, Waring ME, et al. Frailty, cognitive impairment, and anticoagulation among older adults with nonvalvular atrial fibrillation. J Am Geriatr Soc. 2020;68(12):2778-2786. doi: 10.1111/jgs.16756 [DOI] [PMC free article] [PubMed] [Google Scholar]

[zoi261009r13] 13.Shah SJ, Fang MC, Jeon SY, Gregorich SE, Covinsky KE. Geriatric syndromes and atrial fibrillation: prevalence and association with anticoagulant use in a national cohort of older Americans. J Am Geriatr Soc. 2021;69(2):349-356. doi: 10.1111/jgs.16822 [DOI] [PMC free article] [PubMed] [Google Scholar]

[zoi261009r14] 14.Chen Q, Baek J, Goldberg R, Tjia J, Lapane K, Alcusky M. To treat or not to treat: a comparative effectiveness analysis of oral anticoagulant outcomes among U.S. nursing home residents with atrial fibrillation. BMC Geriatr. 2024;24(1):619. doi: 10.1186/s12877-024-05186-9 [DOI] [PMC free article] [PubMed] [Google Scholar]

[zoi261009r15] 15.Zhang N, Patel J, Chen Z, et al. Geriatric conditions are associated with decreased anticoagulation use in long-term care residents with atrial fibrillation. J Am Heart Assoc. 2021;10(16):e021293. doi: 10.1161/JAHA.121.021293 [DOI] [PMC free article] [PubMed] [Google Scholar]

[zoi261009r16] 16.Kapoor A, Foley G, Zhang N, et al. Geriatric conditions predict discontinuation of anticoagulation in long-term care residents with atrial fibrillation. J Am Geriatr Soc. 2020;68(4):717-724. doi: 10.1111/jgs.16335 [DOI] [PubMed] [Google Scholar]

[zoi261009r17] 17.Chen Q, Baek J, Goldberg R, Tjia J, Lapane K, Alcusky M. Discontinuation of oral anticoagulant use among nursing home residents with atrial fibrillation before hospice enrollment. J Am Geriatr Soc. 2023;71(10):3071-3085. doi: 10.1111/jgs.18512 [DOI] [PMC free article] [PubMed] [Google Scholar]

[zoi261009r18] 18.Heshmatollah A, Mutlu U, Koudstaal PJ, Ikram MA, Ikram MK. Cognitive and physical impairment and the risk of stroke - a prospective cohort study. Sci Rep. 2020;10(1):6274. doi: 10.1038/s41598-020-63295-y [DOI] [PMC free article] [PubMed] [Google Scholar]

[zoi261009r19] 19.Lee M, Saver JL, Hong KS, et al. Cognitive impairment and risk of future stroke: a systematic review and meta-analysis. CMAJ. 2014;186(14):E536-E546. doi: 10.1503/cmaj.140147 [DOI] [PMC free article] [PubMed] [Google Scholar]

[zoi261009r20] 20.Murakami K, Tsubota-Utsugi M, Satoh M, et al. Impaired higher-level functional capacity as a predictor of stroke in community-dwelling older adults: the Ohasama Study. Stroke. 2016;47(2):323-328. doi: 10.1161/STROKEAHA.115.011131 [DOI] [PubMed] [Google Scholar]

[zoi261009r21] 21.Gage BF, Birman-Deych E, Kerzner R, Radford MJ, Nilasena DS, Rich MW. Incidence of intracranial hemorrhage in patients with atrial fibrillation who are prone to fall. Am J Med. 2005;118(6):612-617. doi: 10.1016/j.amjmed.2005.02.022 [DOI] [PubMed] [Google Scholar]

[zoi261009r22] 22.Shih RD, Ouslander JG. Intracranial hemorrhage in older adults: implications for fall risk assessment and prevention. J Am Geriatr Soc. 2020;68(5):953-955. doi: 10.1111/jgs.16399 [DOI] [PubMed] [Google Scholar]

[zoi261009r23] 23.Mitchell A, Elmasry Y, van Poelgeest E, Welsh TJ. Anticoagulant use in older persons at risk for falls: therapeutic dilemmas-a clinical review. Eur Geriatr Med. 2023;14(4):683-696. doi: 10.1007/s41999-023-00811-z [DOI] [PMC free article] [PubMed] [Google Scholar]

[zoi261009r24] 24.Hernán MA, Wang W, Leaf DE. Target trial emulation: a framework for causal inference from observational data. JAMA. 2022;328(24):2446-2447. doi: 10.1001/jama.2022.21383 [DOI] [PubMed] [Google Scholar]

[zoi261009r25] 25.Wang SV, Schneeweiss S, Franklin JM, et al. ; RCT-DUPLICATE Initiative . Emulation of randomized clinical trials with nonrandomized database analyses: results of 32 clinical trials. JAMA. 2023;329(16):1376-1385. doi: 10.1001/jama.2023.4221 [DOI] [PMC free article] [PubMed] [Google Scholar]

[zoi261009r26] 26.Jones AL, Dwyer LL, Bercovitz AR, Strahan GW. The National Nursing Home Survey: 2004 overview. Vital Health Stat 13. 2009;(167):1-155. [PubMed] [Google Scholar]

[zoi261009r27] 27.Fong JH, Mitchell OS, Koh BSK. Disaggregating activities of daily living limitations for predicting nursing home admission. Health Serv Res. 2015;50(2):560-578. doi: 10.1111/1475-6773.12235 [DOI] [PMC free article] [PubMed] [Google Scholar]

[zoi261009r28] 28.Gaugler JE, Duval S, Anderson KA, Kane RL. Predicting nursing home admission in the U.S: a meta-analysis. BMC Geriatr. 2007;7:13. doi: 10.1186/1471-2318-7-13 [DOI] [PMC free article] [PubMed] [Google Scholar]

[zoi261009r29] 29.Richardson DB. An incidence density sampling program for nested case-control analyses. Occup Environ Med. 2004;61(12):e59. doi: 10.1136/oem.2004.014472 [DOI] [PMC free article] [PubMed] [Google Scholar]

[zoi261009r30] 30.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]

[zoi261009r31] 31.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]

[zoi261009r32] 32.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]

[zoi261009r33] 33.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]

[zoi261009r34] 34.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]

[zoi261009r35] 35.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]

[zoi261009r36] 36.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]

[zoi261009r37] 37.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]

[zoi261009r38] 38.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]

[zoi261009r39] 39.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]

[zoi261009r40] 40.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]

[zoi261009r41] 41.Sison SDM, Shi SM, Oh G, Jeong S, McCarthy EP, Kim DH. Claims-based frailty index and its relationship with commonly used clinical frailty measures. J Gerontol A Biol Sci Med Sci. 2024;79(7):glae094. doi: 10.1093/gerona/glae094 [DOI] [PMC free article] [PubMed] [Google Scholar]

[zoi261009r42] 42.Lin KJ, Singer DE, Glynn RJ, et al. Prediction score for anticoagulation control quality among older adults. J Am Heart Assoc. 2017;6(10):e006814. doi: 10.1161/JAHA.117.006814 [DOI] [PMC free article] [PubMed] [Google Scholar]

[zoi261009r43] 43.Xu S, Ross C, Raebel MA, Shetterly S, Blanchette C, Smith D. Use of stabilized inverse propensity scores as weights to directly estimate relative risk and its confidence intervals. Value Health. 2010;13(2):273-277. doi: 10.1111/j.1524-4733.2009.00671.x [DOI] [PMC free article] [PubMed] [Google Scholar]

[zoi261009r44] 44.de Bakker PIW, Ferreira MAR, Jia X, Neale BM, Raychaudhuri S, Voight BF. Practical aspects of imputation-driven meta-analysis of genome-wide association studies. Hum Mol Genet. 2008;17(R2):R122-R128. doi: 10.1093/hmg/ddn288 [DOI] [PMC free article] [PubMed] [Google Scholar]

[zoi261009r45] 45.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]

[zoi261009r46] 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]

[zoi261009r47] 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(1):142. doi: 10.1186/1471-2288-13-142 [DOI] [PMC free article] [PubMed] [Google Scholar]

[zoi261009r48] 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]

[zoi261009r49] 49.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]

[zoi261009r50] 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]

[zoi261009r51] 51.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]

[zoi261009r52] 52.Liu X, Baumgarten M, Smith G, et al. Warfarin usage among elderly atrial fibrillation patients with traumatic injury, an analysis of United States Medicare fee-for-service enrollees. J Clin Pharmacol. 2015;55(1):25-32. doi: 10.1002/jcph.375 [DOI] [PMC free article] [PubMed] [Google Scholar]

[zoi261009r53] 53.Chen Q, Lapane K, Nunes AP, Tjia J, Hugunin J, Alcusky M. Prevalence and the factors associated with oral anticoagulant use among nursing home residents. J Clin Pharm Ther. 2021;46(6):1714-1728. doi: 10.1111/jcpt.13508 [DOI] [PMC free article] [PubMed] [Google Scholar]

[zoi261009r54] 54.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]

[zoi261009r55] 55.Parks AL, Frankel DS, Kim DH, et al. Management of atrial fibrillation in older adults. BMJ. 2024;386:e076246. doi: 10.1136/bmj-2023-076246 [DOI] [PubMed] [Google Scholar]

[zoi261009r56] 56.Connolly SJ, Eikelboom J, Joyner C, et al. ; AVERROES Steering Committee and Investigators . Apixaban in patients with atrial fibrillation. N Engl J Med. 2011;364(9):806-817. doi: 10.1056/NEJMoa1007432 [DOI] [PubMed] [Google Scholar]

[zoi261009r57] 57.Søgaard M, Jensen M, Højen AA, et al. Net clinical benefit of oral anticoagulation among frail patients with atrial fibrillation: nationwide cohort study. Stroke. 2024;55(2):413-422. doi: 10.1161/STROKEAHA.123.044407 [DOI] [PubMed] [Google Scholar]

[zoi261009r58] 58.Krüger N, Schneeweiss S, Desai RJ, et al. Cardiovascular outcomes of semaglutide and tirzepatide for patients with type 2 diabetes in clinical practice. Nat Med. 2026;32(1):342-352. doi: 10.1038/s41591-025-04102-x [DOI] [PMC free article] [PubMed] [Google Scholar]

[zoi261009r59] 59.Connolly SJ, Eikelboom JW, Ng J, et al. ; ACTIVE (Atrial Fibrillation Clopidogrel Trial with Irbesartan for Prevention of Vascular Events) Steering Committee and Investigators . Net clinical benefit of adding clopidogrel to aspirin therapy in patients with atrial fibrillation for whom vitamin K antagonists are unsuitable. Ann Intern Med. 2011;155(9):579-586. doi: 10.7326/0003-4819-155-9-201111010-00004 [DOI] [PubMed] [Google Scholar]

[zoi261009r60] 60.Fang MC, Go AS, Chang Y, et al. Long-term survival after ischemic stroke in patients with atrial fibrillation. Neurology. 2014;82(12):1033-1037. doi: 10.1212/WNL.0000000000000248 [DOI] [PMC free article] [PubMed] [Google Scholar]

[zoi261009r61] 61.Raposeiras-Roubín S, Abu-Assi E, Cespón Fernández M, et al. Trade-off between the effects of embolic versus bleeding events on mortality in elderly patients with atrial fibrillation. Rev Esp Cardiol (Engl Ed). 2022;75(4):334-342. doi: 10.1016/j.rec.2021.02.003 [DOI] [PubMed] [Google Scholar]

[zoi261009r62] 62.Wilke T, Bauer S, Mueller S, Kohlmann T, Bauersachs R. Patient preferences for oral anticoagulation therapy in atrial fibrillation: a systematic literature review. Patient. 2017;10(1):17-37. doi: 10.1007/s40271-016-0185-9 [DOI] [PMC free article] [PubMed] [Google Scholar]

[zoi261009r63] 63.Devereaux PJ, Anderson DR, Gardner MJ, et al. Differences between perspectives of physicians and patients on anticoagulation in patients with atrial fibrillation: observational study. BMJ. 2001;323(7323):1218-1222. doi: 10.1136/bmj.323.7323.1218 [DOI] [PMC free article] [PubMed] [Google Scholar]

[zoi261009r64] 64.Ko D, Lin KJ, Bessette LG, et al. Trends in use of oral anticoagulants in older adults with newly diagnosed atrial fibrillation, 2010-2020. JAMA Netw Open. 2022;5(11):e2242964. doi: 10.1001/jamanetworkopen.2022.42964 [DOI] [PMC free article] [PubMed] [Google Scholar]

[zoi261009r65] 65.getPIContent.pdf. Accessed December 5, 2024. https://daiichisankyo.us/prescribing-information-portlet/getPIContent?productName=Savaysa&inline=true

PERMALINK

Anticoagulation Outcomes in Nursing Home Residents With Atrial Fibrillation and ADL Disability

Qiaoxi Chen, PhD

Daniel Singer, MD

Chun-Ting Yang, PhD

Kevin T Pritchard, PhD

Dae Hyun Kim, MD

Kueiyu Joshua Lin, MD

Abstract

Importance

Objective

Design, Setting, and Participants

Exposures

Main Outcomes and Measures

Results

Conclusions and Relevance

Key Points

Question

Findings

Meaning

Introduction

Methods

Study Design and Data Sources

Study Population

Use and Nonuse of OACs

Outcomes and Follow-Up

Baseline Covariates

Statistical Analysis

Main Analysis

Subgroup and Sensitivity Analyses

Results

Overall Baseline Characteristics

Table 1. Characteristics of Long-Term Nursing Home Residents With Atrial Fibrillation and ADL Disability Receiving Oral Anticoagulants.

Clinical Outcomes for Apixaban Use vs Nonuse

Table 2. Clinical Outcomes for Oral Anticoagulant Use vs Nonuse in Long-Term Nursing Home Residents With Atrial Fibrillation and Activities of Daily Living Disability.

Figure 1. Dot Plots of Adjusted Rate Ratios (ARRs) of Secondary Outcomes in Long-Term Nursing Home Residents With Atrial Fibrillation and Activities of Daily Living Disability Between Propensity Score–Matched Patients Taking vs Not Taking Oral Anticoagulants.

Clinical Outcomes for Rivaroxaban Use vs Nonuse

Clinical Outcomes for Warfarin Use vs Nonuse

ITT Analysis

Direct Active-Comparator Analyses

Table 3. Direct Active-Comparator Analyses of Oral Anticoagulant Use in Long-Term Nursing Home Residents With Atrial Fibrillation and Activities of Daily Living Disability.

Tipping-Point Analysis

Figure 2. Dot Plots of Tipping-Point Analysis of the Benefit-Risk Balance Between Ischemic Stroke and Major Bleeding in Long-Term Nursing Home Residents With Atrial Fibrillation and Activities of Daily Living Disability in Patients Receiving vs Not Receiving Oral Anticoagulants.

Sensitivity Analyses

Discussion

Strengths and Limitations

Conclusions

References

Associated Data

Supplementary Materials

ACTIONS

PERMALINK

RESOURCES

Similar articles

Cited by other articles

Links to NCBI Databases