American Journal of Cardiovascular Drugs

, Volume 10, Issue 1, pp 37–48

Predictors of Warfarin Use in Atrial Fibrillation Patients in the Inpatient Setting

  • Shuchita Agarwal
  • Dimitri Bennett
  • David J. Smith
Original Research Article

DOI: 10.2165/11318870-000000000-00000

Cite this article as:
Agarwal, S., Bennett, D. & Smith, D.J. Am J Cardiovasc Drugs (2010) 10: 37. doi:10.2165/11318870-000000000-00000



There is substantial published evidence that warfarin reduces the risk of stroke in patients with atrial fibrillation (AF). However, the current literature suggests that not all patients who could benefit from warfarin receive the drug.


To evaluate patient-related demographic and clinical factors that could influence warfarin use or other anticoagulant use in hospitalized patients with AF.

Study Design

Retrospective observational study using claims data from the Wolters Kluwer Pharma Solutions Hospital Patient Level Database, evaluating characteristics of patients hospitalized in the US between 1 November 2003 and 31 October 2004.


Hospital care.


The study included 44 193 patients aged ≥40 years who were hospitalized between 1 November 2003 and 31 October 2004 and had a diagnosis of AF during hospitalization (AF did not need to be the cause of hospitalization).


Use of warfarin or other anticoagulants (unfractionated heparin [UFH] or low-molecular-weight heparin [LMWH]) was evaluated.

Main Outcome Measures

A logistic regression model was used to identify factors associated with warfarin use, international normalized ratio (INR) monitoring, or the use of anticoagulants (UFH or LMWH).


In this analysis of hospitalized patients with AF in the real-world setting, about 56% of patients received anticoagulation therapy with warfarin. Elderly patients aged ≥75 years were less likely to be treated with warfarin than younger patients, but patients between the ages of 60 and 74 years were more likely to use warfarin than their younger counterparts. Except for patients with congestive heart failure or vascular malformation, patients with other bleeding risk factors (hepatic disease, renal disease, aspirin use, and fractures) were significantly less likely to receive warfarin than those without these risk factors. CHADS2 scores for stroke risk of 2 and 3 were associated with a significantly higher likelihood of warfarin treatment than scores of 0 or 1. Patients admitted through a routine admission (an outpatient department) were significantly more likely to be prescribed warfarin than patients admitted through an emergency room. Patients aged ≥75 years and aspirin users were more likely to have their INR monitored during hospitalization. With respect to other anticoagulant use, females and older patients (≥65 years) were less likely to use UFH or LMWH, and patients with renal disease or vascular malformation and those receiving aspirin were more likely to use UFH or LMWH than patients without these conditions/not receiving aspirin. Patients admitted through the emergency room were more likely to receive an anticoagulant than patients admitted through an outpatient department, an inpatient transfer, or any other source.


Older age, female sex, and certain risk factors for bleeding, including hepatic disease, renal disease, aspirin use, and fractures, were associated with a lower likelihood of warfarin treatment, while a higher stroke risk (as indicated by CHADS2 scores) was associated with a higher likelihood of warfarin treatment, in hospitalized patients with a diagnosis of AF. The likelihood of INR being monitored increased for patients aged ≥75 years and for aspirin users. Older patients and female patients were less likely to be prescribed other anticoagulants (UFH or LMWH) also.

Copyright information

© Adis Data Information BV 2010

Authors and Affiliations

  • Shuchita Agarwal
    • 1
  • Dimitri Bennett
    • 2
  • David J. Smith
    • 1
  1. 1.Wolters Kluwer Pharma SolutionsYardleyUSA
  2. 2.GlaxoSmithKlinePhiladelphiaUSA