Skip to main content

Table 1 Characteristics of Included Studies Evaluating the Association Between Covariates and Warfarin Use for Stroke Prevention in Atrial Fibrillation

From: Predictors of warfarin use in atrial fibrillation in the United States: a systematic review and meta-analysis

Study, Year

Total N

Study Design

Population and Setting

Exclusion Criteria

Warfarin Definition

(% warfarin use)

Study Period

Agarwal, 2010

44,193

R, O

Patients aged ≥ 40 years who were hospitalized and had a diagnosis of AF (ICD-9: 427.xx).

Patients with hyperthyroidism or who were pregnant.

Presence of a warfarin claim during their inpatient stay

(56.2%)

2003-2004

Meschia, 2010

258

P, O

Patients aged ≥ 45 years from the Reasons for Geographic and Racial Difference in Stroke (REGARDS) study who have positive EKG evidence of AF and self-report AF during in-home visit

None

Current aspirin and warfarin treatment was defined using

an inventory of current medications that was conducted as part of the in-home visit, in which all prescription and over-the-counter medications taken in the past 2 weeks were recorded

(79.8%)

2003-2007

Lewis, 2009A

7,635

R, O (nested in the prospective GWTG database)

Consecutive patients in the Get with the Guidelines program Stroke database presenting with ischemic stroke or TIA (ICD-9: 433 to 436) and AF documented using EKG during the admission

Patients with documented contraindication to anticoagulation; patient death, leaving against medical advice, discharged to hospice, or transferred to another acute-care facility

Prescription of warfarin therapy at discharge

(78.8%)

2001-2005

Lewis, 2009B

7,826

R, O (nested in the prospective GWTG database)

Consecutive patients in the Get with the Guidelines program Stroke database presenting with ischemic stroke or TIA (ICD-9: 433 to 436) and AF documented using medical history only

Patients with documented contraindication to anticoagulation; patient death, leaving against medical advice, discharged to hospice, or transferred to another acute-care facility

Prescription of warfarin therapy at discharge

(49.4%)

2001-2005

Niska, 2009

1,771

R, O

Random, representative, and multistage sample from the National Ambulatory Medical Care Survey (NAMCS) and the National Hospital Ambulatory medical Care Survey (NHAMCS) of patient visits; patients were aged ≥ 20 years, had a diagnosis of AF (ICD-9-CM: 427.31)

Malignant or benign brain neoplasms, bleeding disorders, alcoholism, Alzheimers and other dementias, seizure disorders, chronic renal disease, cerebal hemorrhage, liver disease, peptic ulcer disease, gastritis, or duodenitis

Prescription or continuation of warfarin during office visit (52.2%)

2001-2006

Piccini, 2009

15,748

R, O (nested in the prospective GWTG database)

Patients hospitalized with HF and either AF upon admission or a prior history of AF in the Get With The Guidelines-Heart Failure (GWTG-HF) registry

Documented contraindications, intolerance, or other documented reasons for not prescribing warfarin; medical histories with < 75% completeness or conflicting data fields

Warfarin use at discharge (65.2%)

2005-2008

Glazer, 2007

572

R, O

Patient aged between 30 to 84 years with newly detected AF (first clinically recognized lifetime episode of non-surgery-related AF, ICD-9:427.31, (atrial flutter) 427.32) in a health maintenance organization (Group Health Cooperative) database

Patients who died during hospitalization, had a pacemaker implanted before AF onset, or had fewer than 4 health care visits any time before AF onset date.

Warfarin use during 6-month follow up period after AF onset (54.9%)

2001-2002

Schauer, 2007

(Patients overlap with Johnston, 2003)

6,283

R, O

White and African-American Ohio Medicaid patients with newly incident nonvalvular AF (ICD-9-CM: 427.31); patients must have at least 2 claims for AF and have a full year of continuous Ohio Medicaid enrollment without diagnosis of AF before the first diagnosis

Patients who filled any warfarin prescriptions more than 7 days prior to the diagnosis of AF; patients with a history of valvular heart disease prior to the diagnosis of AF, as ascertained by 2 or more claims for mitral valve disease, heart valve transplant, heart valve replacement, or a procedure code for mitral or aortic valve repair or replacement; patients for whom race could not be determined

Claim for a warfarin prescription at any time between 7 days prior to the initial diagnosis of AF and 30 days after the initial diagnosis

(9.1%)

1997-2002

Birman-Deych, 2006

17,272

R, O

Patients (from the National Registry of Atrial Fibrillation II) with medicare Part A and Part B claims who were hospitalized with AF

Patient who died during baseline hospitalization, had a terminal illness, had no Medicare Part B claims during follow-up, or were aged < 65 years at baseline

Patients discharged with warfarin prescription with < 91 days between successive INR tests

(49.1%)

1998-1999

Hylek, 2006

405

P, O

Consecutive patients identified by daily searches of electronic admission notes and EKGs of all admissions to Massachusetts General Hospital that had AF verified by EKG, were aged ≥ 65 years, not taking warfarin on admission, and had longitudinal care provided at the institution

Other long-term indication for warfarin therapy

Started on and discharged with warfarin according to discharge summary or electronic discharge medication list

(51%)

2001-2003

Burkiewicz, 2005

178

R, O

Patients with a AF diagnosis (ICD-9-CM:427.31) in a database shared by two ambulatory care clinics

Patients with a primary care physician at another facility

Any documented prescription for warfarin

(73.6%)

2000-2001

Abdel-Latif, 2005

117

R, O

LTC patients with chronic or paroxysmal AF either by diagnosis or EKG

NR

Warfarin use for 6 months or longer according to pharmacy or medical records

(46.1%)

NR

Lim, 2005

2,011

R, O

A random sample of Medicare fee-for-service patients discharged from Michigan's acute care hospitals (excluding Veteran's Administration) with a primary or secondary discharged diagnosis of AF (ICD9-CM:427.31); patients who met national guidelines for anticoagulant therapy

Patients with lone AF, aged < 65 years, planned surgery within 7 days of discharge or recent surgery, physician documentation of risk for falls, alcoholism or drug abuse (history or current), dual chamber pacemaker (history or current), schizophrenia/active psychosis (history or current), extensive metastatic cancer (history or current), brain or central nervous system cancer (history or current), seizures (history or current), malignant hypertension (history or current), CVA hemorrhagic (history or current), peptic ulcer (current), intracranial surgery/biopsy (current), hemorrhage (history or current), and physician documentation of rationale for not prescribing warfarin

Warfarin treatment at discharge

(53.9%)

1998-1999

Waldo, 2005

945

R, O

Randomly chosen patients from select hospitals participating in the National Anticoagulation Benchmark and Outcomes Report (NABOR) program who were discharged with a primary or secondary diagnosis of AF (ICD-9-CM: 427.31)

Patients aged < 18 years, admitted from another acute care hospital where warfarin therapy was already initiated, or discharged to another acute care hospital to continue warfarin treatment

Warfarin treatment

(53.5%)

2000-2002

Brophy, 2004a

2,217

R, O

Patients with a documented healthcare encounter in the Veterans Affairs Boston Healthcare System database, electrocardiogram-documented AF in the Marquette Universal Storage for Electrocardiograms database, and a verified diagnosis code for AF [ICD-9-CM: 427(.3,.31)] in the national Veterans Affairs database

Patients with valvular heart disease [ICD-9-CM: 391.1, 394(.0-.2), 396(.0-.3,.8), 424.0, 746(.5,.6)]

A prescription for warfarin in the Veterans Affairs Boston Healthcare System database during the study time period

(34.8%)

1998-2001

Brophy, 2004b

1,596

R, O

Patients with a documented healthcare encounter in the Veterans Affairs Boston Healthcare System database, electrocardiogram-documented AF in the Marquette Universal Storage for Electrocardiograms database, and a verified diagnosis code for AF [ICD-9-CM: 427(.3,.31)] in the national Veterans Affairs database

Patients with a contraindication to warfarin use, or valvular heart disease [ICD-9-CM: 391.1, 394(.0-.2), 396(.0-.3,.8), 424.0, 746(.5,.6)]

Any prescription for warfarin in the Veterans Affairs Boston Healthcare System database (64.2%)

1998-2001

Fang, 2004

1,335 visits

R, O

Patients with AF (ICD-9-CM: 429.31) from the National Ambulatory Medical Care Survey (NAMCS), a nationally representative assessment of office-based practice

Providers not in internal medicine, general practice, family practice, cardiology, or cardiac electrophysiology; patients with the following diagnosis: dementia, gait abnormalities, epilepsy, intracranial hemorrhages, gastritis or duodenitis, gastrointestinal ulcer disease, gastrointestinal hemorrhages, chronic liver disease, alcoholism, purpura, hematuria, and neoplasms of the central nervous system and gastrointestinal or genitourinary systems

Warfarin, dicumarol, anisindione, phenprocoumon use

(NR)

1997-2000

Rahimi, 2004

290

R, O

Patients with a diagnosis of AF requiring anticoagulation therapy admitted to a community-based teaching hospital in Southeast Georgia

Patients with hypercoagulable state, hemorrhagic stroke, carotid stenosis, peripheral vascular disease, or dilated cardiomyopathy

Prescribed warfarin

(42.8%)

1997-2000

Johnston, 2003

11,699

R, O

Patients in the Ohio Medicaid Program database with a first diagnosis of AF

Patients enrolled in capitated plans in the Ohio Medicaid Program and those who did not have a full year of continuous Ohio Medicaid enrollment without diagnosis of AF before the first diagnosis; patients with other indications for warfarin including valvular heart disease and valve repair or replacement; patients with transient AF including ones with a single ICD-9-CM code for AF associated with a ICD-9-CM code for hyperthyroidism or a ICD-9-CM code for operative procedures commonly associated with perioperative or postoperative AF' patients already receiving warfarin prior to AF diagnosis

Claim in Ohio Medicaid administrative database for warfarin use (ICD-9-CM:V58.61) or warfarin prescription from 7 days preceding to 30 days after the development of AF

(9.7%)

1998-2000

McCormick, 2001

429

R, O

LTC patients in Connecticut with diagnosis of AF confirmed by EKG or written documentation by the LTC facility's physician

Patients who had resided in the LTC facility for < 30 days or had end-stage renal disease

Receipt of warfarin therapy for ≥ 2 weeks during the prior 12 months

(42%)

NR

Go, 1999

13,428

R, O

Patient in a health maintenance organization (Kaiser Permanente Medical Care Program in Northern California) database who had a diagnosis of nonvalvular AF (ICD-9-CM: 427.31) recorded in the automated outpatient database and an electrocardiogram showing AF in the electrocardiographic database (if database was available at time of diagnosis)

Patients with the following characteristics: no health membership after diagnosis of AF, age younger than 18 years, transient AF secondary to cardiac surgery, mitral stenosis or mitral or aortic valve repair or replacement, concomitant hyperthyroidism, or no outpatient, internal medicine, or cardiology care during 12 months after first diagnosis of AF

Having either a filled prescription for warfarin or dicumarol in the pharmacy database, more than one outpatient INR, or a diagnosis of "Coumadin therapy" (ICD-9: V58.61) 3 months before or after the first identified diagnosis of AF

(53.7%)

1996-1997

Smith, 1999a

144

P, O

Patients from the Cardiovascular Health Study (CHS) aged ≥ 65 years with EKG-identified prevalent AF (paroxysmal or chronic)

Patients with a mechanical pacing device; AF patients too ill to participate further or not available for follow-up

Warfarin on medication list taken at each annual clinic visit

(13%)

1989-1990 (baseline)

Smith, 1999b

135

P, O

Patients from the Cardiovascular Health Study (CHS) aged ≥ 65 years with EKG-identified prevalent AF (paroxysmal or chronic)

Patients with a mechanical pacing device; AF patients too ill to participate further or not available for follow-up

Warfarin on medication list taken at each annual clinic visit

(50%)

1995-1996 (6 year follow-up)

White, 1999

172

P, O

Subgroup of patients aged ≥ 70 years in the Cardiovascular Health Study with AF on EKG at one or more yearly examinations along with information regarding warfarin use and no pre-existing indication for its use

Patients who were in nursing homes, wheel-chair bound, had a mechanical heart valve, had a history of DVT or PE before starting warfarin therapy, being treated for cancer, or taking warfarin prior to onset of AF

Self-reported use of warfarin in 1995

(37%)

1993-1995

Brass, 1998a

278

R, O

Medicare patients aged ≥ 65 years hospitalized with a a principal diagnosis of of ischemic stroke using ICD-9 codes and discharged alive with a primary or secondary diagnosis of AF

Patients with a potential indication for anticoagulation other than AF including patients with primary diagnosis of AMI or embolic events (other than stroke); patients with retinal vascular occlusion, peripheral vascular disease, vascular insufficiency of the intestine, and vascular disorders of the kidney

Prescribed warfarin at discharge (53%)

1994

Brass, 1998b

203

R, O

Medicare patients aged ≥ 65 years hospitalized with a a principal diagnosis of of ischemic stroke using ICD-9 codes, discharged alive with a primary or secondary diagnosis of AF and not receiving warfarin at time of admission

Patients with a potential indication for anticoagulation other than AF including patients with primary diagnosis of AMI or embolic events (other than stroke); patients with retinal vascular occlusion, peripheral vascular disease, vascular insufficiency of the intestine, and vascular disorders of the kidney

Prescribed warfarin at discharge (41.9%)

1994

Stafford, 1998

877 visits

R, O

Nationally representative and random sample of office visits by patients with AF (ICD-9-CM: 427.31) from the National Ambulatory Medical Care Surveys

Patients with potential contraindications for anticoagulation, including peptic ulcer disease, gastritis and duodenitis, other gastrointestinal bleeding, alcoholism, gait abnormality, ataxia, Alzheimer's or other dementia, cerebral hemorrhage, seizure disorder, benign or malignant central nervous system tumors, gastrointestinal and genitourinary tract renal malignancies, thrombocytopenia, hematuria, esophageal varices, and renal insufficiency; patients < 65 years old lacking other risk factors for stroke (CHF, ischemic heart disease, diabetes mellitus, hypertension, valvular disease, or previous stroke); patient visits made to physicians other than cardiologists, general internists, family physicians, and general practitioners

A medication code during a visit for warfarin, dicumarol or anisindione

(NR)

1989-1996

Brass, 1997

488

R, O

Medicare patients aged ≥ 65 years with established AF (before hospitalization) who were hospitalized with a principal diagnosis (reason for admission) of ischemic stroke and a secondary diagnosis of AF (ICD-9:427.31); patients without stroke were selected with a primary or secondary discharge diagnosis of A and matched to one patient with stroke on age (within 1 year), sex, and secondary diagnoses of hypertension, non-insulin-dependent diabetes, insulin-dependent diabetes, congestive heart failure, angina, and myocardial infarction as a nonprimary diagnosis

Patients with a potential indication for anticoagulation other than AF including patients with primary diagnosis of AMI or embolic events (other than stroke); patients with retinal vascular occlusion, peripheral vascular disease, vascular insufficiency of the intestine, and vascular disorders of the kidney

Prescribed warfarin at time of admission to hospital

(34%)

1994

Munschauer, 1997

651

R, O

Patients discharged from hospital with AF (ICD-9: 427.31)

Patients with transient or paroxysmal AF, a recent major surgical procedure, or undergoing treatment for active malignancy

Treatment with warfarin at discharge

(36%)

1994-1995

Antani & Beyth, 1996a

189

R, O

Consecutive inpatients with nonrheumatic AF discharged alive with a discharged diagnosis of AF (ICD-9: 427.31) and confirmed by review of medical records, or outpatients with nonrheumatic AF

Patients with transient AF, history of rheumatic fever or rheumatic heart disease, or lone AF

Warfarin prescription identified by medical record review

(23%)

1990-1993

Beyth & Antani, 1996b

136

R, O

Consecutive patients with sustained or intermittent nonrheumatic AF

Patients with transient, rheumatic, or lone AF

Treated with warfarin

(24%)

1992

Stafford, 1996a

1,062 visits

R, O

Visits by patients with AF (ICD-9-CM:427.31) to randomly selected office-based physicians included in the National Ambulatory Medical Care Surveys

Visits by patients with atrial flutter (ICD-9-CM:427.32)

A medication code for warfarin (generic or proprietary names) associated with each visit (20.8%)

1980-1993

Stafford, 1996b

272 visits

R, O

Visits by patients with AF (ICD-9-CM:427.31) to randomly selected office-based physicians included in the National Ambulatory Medical Care Surveys

Visits by patients with atrial flutter (ICD-9-CM:427.32)

A medication code for warfarin (generic or proprietary names) associated with each visit (32.0%)

1992-1993

  1. Abbreviations: AF: atrial fibrillation, AMI: acute myocardial infarction, CM: clinical modification, CS: cross-sectional, EKG: electrocardiogram, CVA: cerebral vascular accident, DVT: deep-vein thrombosis, HF: heart failure, HIV: human immunodeficiency virus, ICD-9: International Classification of Diseases, Ninth Revision, LTC: long-term care, NR: not reported O: observational, P: prospective, PE: pulmonary embolism, R: retrospective, TIA: transient ischemic attack