In mixed-effects logistic regression models adjusting for patient sociodemographic (model 1), clinical factors (model 2), and facility and clinician factors (model 3), there were significant differences in any anticoagulant initiation with lower initiation among persons experiencing homelessness. Among patients who initiated therapy, there was no significant difference in the type of therapy used by patients with and without experiences of homelessness.
eTable. Identification of Homelessness by Diagnosis and Clinical Codes and Screening Questions
eFigure. Study Flow Diagram
Customize your JAMA Network experience by selecting one or more topics from the list below.
Wilson DA, Boadu O, Jones AL, et al. Analysis of Initiating Anticoagulant Therapy for Atrial Fibrillation Among Persons Experiencing Homelessness in the Veterans Affairs Health System. JAMA Netw Open. 2022;5(7):e2223815. doi:10.1001/jamanetworkopen.2022.23815
In the US, more than 37 000 veterans are homeless every night.1 Persons who have experienced homelessness (PEH) have a higher burden of cardiovascular diseases, such as atrial fibrillation (AF),2 documented challenges accessing health care, and suboptimal management of cardiovascular conditions.3 Stroke-preventing anticoagulant therapy improves AF outcomes, but its use among PEH is unknown.
This cohort study used data from the Race, Ethnicity, and Anticoagulation Choice in Atrial Fibrillation (REACH-AF) cohort to compare rates and types of anticoagulant therapy among PEH vs non-PEH.4 Using administrative and clinical data from the Veterans Health Administration (VA), we defined the REACH-AF cohort as patients with a new AF diagnosis between January 1, 2010, and December 31, 2020, continuous VA enrollment for 2 years before diagnosis, and an outpatient confirmatory AF diagnosis within 180 days of the index diagnosis (eFigure in the Supplement). We excluded patients with valvular heart disease, cardiac ablation, hyperthyroidism, an anticoagulant prescription in the 2 years before the index diagnosis, or death or hospice care in the 90 days after diagnosis. The institutional review board at the VA Pittsburgh Healthcare System approved the study and granted a waiver of informed consent owing to use of deidentified data. We followed the STROBE reporting guideline.
Our primary outcome was anticoagulant therapy initiation, defined as the first outpatient prescription for any anticoagulant within 90 days of the index diagnosis. Among individuals initiating anticoagulant therapy, we determined whether patients were prescribed warfarin or a direct oral anticoagulant (DOAC). Our independent variable was homeless experience documented with International Classification of Diseases, Ninth Revision, or International Statistical Classification of Diseases and Related Health Problems, Tenth Revision codes, receipt of VA homeless services, or a positive screen on the VA annual homeless screener (eTable in the Supplement) in the year before the index diagnosis.5 We estimated differences in anticoagulant initiation, then type of therapy, using mixed-effects logistic regression models adjusted for sociodemographic characteristics, including self-reported race and ethnicity,4 clinical, and clinician and facility covariates. We included a random effect for VA site in all models and used a threshold of P < .05 (2-tailed) for statistical significance. Data analysis was performed using SAS Enterprise Guide, version 8.2 (SAS Institute Inc).
Among 168 003 patients with incident AF from 2014 to 2020, 164 396 were men (97.9%), 3607 were women (2.1%), with a mean (SD) age of 66.0 (10.6) years; 6362 (3.8%) had experiences of homelessness (Table). Anticoagulant initiation was lower among PEH (3576 [56.2%]) than non-PEH (106 265 [65.7%]) (P < .001). In our final adjusted model, the odds of initiating any anticoagulant were significantly lower for PEH (adjusted odds ratio [aOR], 0.79; 95% CI, 0.74-0.84) (Figure). Among anticoagulant initiators, DOAC use appeared to be lower for PEH (2514 [70.3%]) than for non-PEH (79 691 [75.0%]) (P < .001). However, these differences were not statistically significant in the final adjusted model (aOR, 0.96; 95% CI, 0.87-1.06).
In a national cohort of veterans with AF, we found substantial disparities in anticoagulant therapy prescribing between PEH and non-PEH. Among those who initiated therapy, however, there was no significant difference in DOAC use. The limitations include the use of a VA cohort that may affect generalizability to a broader population and the inability to assess prescription copayment and other social barriers to pharmacotherapy access. Furthermore, we were unable to assess the duration of homelessness or exclude residual confounding. To our knowledge, this is the first report observing national differences in anticoagulation among PEH with AF in the DOAC era.6 Our findings suggest that efforts to prevent stroke and improve AF management among PEH may best focus on addressing treatment initiation barriers.
That these findings persisted after adjusting for sociodemographic and clinical factors within an integrated health system with a low-cost, uniform drug formulary and integrated medical and social services has implications for equitable AF management. Research into the determinants of these observed inequities, including clinician bias, determination of VA priority groups, and differential shared decision-making among high-risk populations will be critical for improving the quality of AF care.
Accepted for Publication: June 9, 2022.
Published: July 26, 2022. doi:10.1001/jamanetworkopen.2022.23815
Open Access: This is an open access article distributed under the terms of the CC-BY License. © 2022 Wilson DA et al. JAMA Network Open.
Corresponding Author: Utibe R. Essien, MD, MPH, Center for Health Equity Research and Promotion, VA Pittsburgh Healthcare System & Department of Medicine, University of Pittsburgh School of Medicine, 3609 Forbes Ave, Ste 2, Pittsburgh, PA 15213 (email@example.com).
Author Contributions: Ms Kim and Dr Essien 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: Wilson, Mor, Hausmann, Essien.
Acquisition, analysis, or interpretation of data: Boadu, Jones, Kim, Mor, Hausmann, Essien.
Drafting of the manuscript: Wilson, Boadu, Kim, Essien.
Critical revision of the manuscript for important intellectual content: Wilson, Boadu, Jones, Mor, Hausmann, Essien.
Statistical analysis: Kim, Mor.
Obtained funding: Essien.
Supervision: Hausmann, Essien.
Conflict of Interest Disclosures: Mr Wilson received a stipend from the University of Pittsburgh School of Medicine during this study. Dr Jones reported receiving grants from the Department of Veterans Affairs (VA) during the conduct of the study. Dr Hausmann reported receiving grants from the VA during the conduct of the study and outside the submitted work. Dr Essien reported receiving grants from the VA outside the submitted work. No other disclosures were reported.
Funding/Support: Dr Essien received funding from the Department of Veterans Affairs Health Services Research and Development Division (CDA-20-049) for this work.
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.
Meeting Presentation: This work was presented in virtual format on May 20, 2021, at the 2021 American Heart Association Epidemiology and Prevention–Lifestyle and Cardiometabolic Health 2021 Scientific Sessions.