Atrial Fibrillation
Conditions
Brief summary
This cohort study plans to identify initiators of oral anticoagulants using electronic claims data from a commercial insurance database to quantify associations between anticoagulant choice (warfarin and dabigatran) and the occurrence of selected outcomes in patients with non-valvular atrial fibrillation at risk for stroke.
Interventions
None listed
Sponsors
Boehringer Ingelheim
Study design
Observational model
COHORT
Time perspective
RETROSPECTIVE
Eligibility
Sex/Gender
ALL
Age
18 Years to No maximum
Healthy volunteers
No
Inclusion criteria
* A recorded diagnosis of atrial fibrillation (AF). * Initiation of anticoagulant medication (dabigatran or warfarin). * At least 18 years of age on the date of anticoagulant initiation CHA2DS2-VASc-Score \>=1
Exclusion criteria
* Patients with missing or ambiguous age or sex information * Patients with documented evidence of valvular disease * Patients with less than 12 months enrolment in the UnitedHealth Research Database preceding the date of anticoagulant initiation * Patients with prior use of any oral anticoagulant
Design outcomes
Primary
| Measure | Time frame | Description |
|---|---|---|
| Incidence Rate of Stroke | From treatment initiation until end of follow-up; up to 20 months | Events during follow-up (October 2010- 30 June 2012) were identified using ICD 9 codes. |
| Incidence Rate of Major Bleeding | From treatment initiation until end of follow-up; up to 20 months | Events during follow-up (October 2010- 30 June 2012) were identified using ICD 9 and procedure codes. |
Secondary
| Measure | Time frame | Description |
|---|---|---|
| Incidence Rate of Deep Vein Thrombosis | From treatment initiation until end of follow-up; up to 20 months | Events during follow-up (October 2010- 30 June 2012) were identified using ICD 9 codes. |
| Incidence Rate of Pulmonary Embolism | From treatment initiation until end of follow-up; up to 20 months | Events during follow-up (October 2010- 30 June 2012) were identified using ICD 9 codes. |
| Incidence Rate of Major Lower Gastrointestinal Bleeding | From treatment initiation until end of follow-up; up to 20 months | Events during follow-up (October 2010- 30 June 2012) were identified using ICD 9 codes. |
| Incidence Rate of Major Urogenital Bleeding | From treatment initiation until end of follow-up; up to 20 months | Events during follow-up (October 2010- 30 June 2012) were identified using ICD 9 codes. |
| Incidence Rate of Major Other Bleeding | From treatment initiation until end of follow-up; up to 20 months | Events during follow-up (October 2010- 30 June 2012) were identified using ICD 9 codes. |
| Incidence Rate of Transient Ischemic Attack | From treatment initiation until end of follow-up; up to 20 months | Events during follow-up (October 2010- 30 June 2012) were identified using ICD 9 codes. |
| Incidence Rate of Major Upper Gastrointestinal Bleeding | From treatment initiation until end of follow-up; up to 20 months | Events during follow-up (October 2010- 30 June 2012) were identified using ICD 9 and procedure codes. |
| Incidence Rate of Myocardial Infarction | From treatment initiation until end of follow-up; up to 20 months | Events during follow-up (October 2010- 30 June 2012) were identified using ICD 9 codes. |
| Incidence Rate of Systemic Embolism | From treatment initiation until end of follow-up; up to 20 months | Events during follow-up (October 2010- 30 June 2012) were identified using ICD 9 codes. |
| Incidence Rate of Ischemic Stroke | From treatment initiation until end of follow-up; up to 20 months | Events during follow-up (October 2010- 30 June 2012) were identified using ICD 9 codes. |
| Incidence Rate of Hemorrhagic Stroke | From treatment initiation until end of follow-up; up to 20 months | Events during follow-up (October 2010- 30 June 2012) were identified using ICD 9 codes. |
| Incidence Rate of Stroke Uncertain Classification | From treatment initiation until end of follow-up; up to 20 months | Events during follow-up (October 2010- 30 June 2012) were identified using ICD 9. |
| Incidence Rate of Major Intracranial Bleeding | From treatment initiation until end of follow-up; up to 20 months | Events during follow-up (October 2010- 30 June 2012) were identified using ICD 9 codes. |
| Incidence Rate of Major Extracranial Bleeding | From treatment initiation until end of follow-up; up to 20 months | Events during follow-up (October 2010- 30 June 2012) were identified using ICD 9 and procedure codes. |
| Incidence Rate of Major Gastrointestinal Bleeding | From treatment initiation until end of follow-up; up to 20 months | Events during follow-up (October 2010- 30 June 2012) were identified using ICD 9 and procedure codes. |
| Incidence Rate of Stroke or Systemic Embolism | From treatment initiation until end of follow-up; up to 20 months | Events during follow-up (October 2010- 30 June 2012) were identified using ICD 9 codes. |
| Incidence Rate of Venous Thromboembolism | From treatment initiation until end of follow-up; up to 20 months | Events during follow-up (October 2010- 30 June 2012) were identified using ICD 9 codes. |
Participant flow
Recruitment details
This is an observational study. Patients were not actively recruited in this study. The data source was the UnitedHealth database.
Pre-assignment details
11882 patients fulfilled the in- and ex-clusion criteria. Only 5982 patients remained after matching on the Propensity Score.
Participants by arm
| Arm | Count |
|---|---|
| Dabigatran Propensity score matched patients starting treatment with dabigatran | 2,991 |
| Warfarin Propensity score matched patients starting treatment with warfarin | 2,991 |
| Total | 5,982 |
Baseline characteristics
| Characteristic | Dabigatran | Warfarin | Total |
|---|---|---|---|
| Age, Continuous | 63.86 years STANDARD_DEVIATION 10.99 | 63.26 years STANDARD_DEVIATION 11.04 | 63.60 years STANDARD_DEVIATION 11.01 |
| Sex: Female, Male Female | 943 Participants | 877 Participants | 1820 Participants |
| Sex: Female, Male Male | 2048 Participants | 2114 Participants | 4162 Participants |
Adverse events
| Event type | EG000 affected / at risk | EG001 affected / at risk |
|---|---|---|
| deaths Total, all-cause mortality | — / — | — / — |
| other Total, other adverse events | 0 / 2,991 | 0 / 2,991 |
| serious Total, serious adverse events | 0 / 2,991 | 0 / 2,991 |
Outcome results
Primary
Incidence Rate of Major Bleeding
Events during follow-up (October 2010- 30 June 2012) were identified using ICD 9 and procedure codes.
Time frame: From treatment initiation until end of follow-up; up to 20 months
Population: Hazard ratios and confidence intervals were estimated for patients starting dabigatran or warfarin after matching on the propensity score.
| Arm | Measure | Value (NUMBER) |
|---|---|---|
| Dabigatran | Incidence Rate of Major Bleeding | 60.00 events per 1000 person years |
| Warfarin | Incidence Rate of Major Bleeding | 66.72 events per 1000 person years |
Comparison: These current feasibility data are limited by a small sample size, short follow up, and few outcome events resulting in wide 95% confidence intervals. At this early stage, no comparative conclusions are possible. Future data with an increased number of patients are planned.95% CI: [0.69, 1.36]Regression, Cox
Primary
Incidence Rate of Stroke
Events during follow-up (October 2010- 30 June 2012) were identified using ICD 9 codes.
Time frame: From treatment initiation until end of follow-up; up to 20 months
Population: Hazard ratios and confidence intervals were estimated for patients starting dabigatran or warfarin after matching on the propensity score.
| Arm | Measure | Value (NUMBER) |
|---|---|---|
| Dabigatran | Incidence Rate of Stroke | 29.09 events per 1000 person years |
| Warfarin | Incidence Rate of Stroke | 31.59 events per 1000 person years |
Comparison: These current feasibility data are limited by a small sample size, short follow up, and few outcome events resulting in wide 95% confidence intervals. At this early stage, no comparative conclusions are possible. Future data with an increased number of patients are planned.95% CI: [0.64, 1.7]Regression, Cox
Secondary
Incidence Rate of Deep Vein Thrombosis
Events during follow-up (October 2010- 30 June 2012) were identified using ICD 9 codes.
Time frame: From treatment initiation until end of follow-up; up to 20 months
Population: Hazard ratios and confidence intervals were estimated for patients starting dabigatran or warfarin after matching on the propensity score.
| Arm | Measure | Value (NUMBER) |
|---|---|---|
| Dabigatran | Incidence Rate of Deep Vein Thrombosis | 7.23 events per 1000 person years |
| Warfarin | Incidence Rate of Deep Vein Thrombosis | 27.35 events per 1000 person years |
Comparison: These current feasibility data are limited by a small sample size, short follow up, and few outcome events resulting in wide 95% confidence intervals. At this early stage, no comparative conclusions are possible.95% CI: [0.15, 0.67]Regression, Cox
Secondary
Incidence Rate of Hemorrhagic Stroke
Events during follow-up (October 2010- 30 June 2012) were identified using ICD 9 codes.
Time frame: From treatment initiation until end of follow-up; up to 20 months
Population: Hazard ratios and confidence intervals were estimated for patients starting dabigatran or warfarin after matching on the propensity score.
| Arm | Measure | Value (NUMBER) |
|---|---|---|
| Dabigatran | Incidence Rate of Hemorrhagic Stroke | 4.02 events per 1000 person years |
| Warfarin | Incidence Rate of Hemorrhagic Stroke | 3.14 events per 1000 person years |
Comparison: These current feasibility data are limited by a small sample size, short follow up, and very few outcome events resulting in wide 95% confidence intervals. At this early stage, no comparative conclusions are possible. Future data with an increased number of patients are planned.95% CI: [0.32, 5.72]Regression, Cox
Secondary
Incidence Rate of Ischemic Stroke
Events during follow-up (October 2010- 30 June 2012) were identified using ICD 9 codes.
Time frame: From treatment initiation until end of follow-up; up to 20 months
Population: Hazard ratios and confidence intervals were estimated for patients starting dabigatran or warfarin after matching on the propensity score.
| Arm | Measure | Value (NUMBER) |
|---|---|---|
| Dabigatran | Incidence Rate of Ischemic Stroke | 27.40 events per 1000 person years |
| Warfarin | Incidence Rate of Ischemic Stroke | 32.71 events per 1000 person years |
Comparison: These current feasibility data are limited by a small sample size, short follow up, and few outcome events resulting in wide 95% confidence intervals. At this early stage, no comparative conclusions are possible. Future data with an increased number of patients are planned.95% CI: [0.58, 1.55]Regression, Cox
Secondary
Incidence Rate of Major Extracranial Bleeding
Events during follow-up (October 2010- 30 June 2012) were identified using ICD 9 and procedure codes.
Time frame: From treatment initiation until end of follow-up; up to 20 months
Population: Hazard ratios and confidence intervals were estimated for patients starting dabigatran or warfarin after matching on the propensity score.
| Arm | Measure | Value (NUMBER) |
|---|---|---|
| Dabigatran | Incidence Rate of Major Extracranial Bleeding | 56.70 events per 1000 person years |
| Warfarin | Incidence Rate of Major Extracranial Bleeding | 63.53 events per 1000 person years |
Comparison: These current feasibility data are limited by a small sample size, short follow up, and few outcome events resulting in wide 95% confidence intervals. At this early stage, no comparative conclusions are possible.95% CI: [0.67, 1.35]Regression, Cox
Secondary
Incidence Rate of Major Gastrointestinal Bleeding
Events during follow-up (October 2010- 30 June 2012) were identified using ICD 9 and procedure codes.
Time frame: From treatment initiation until end of follow-up; up to 20 months
Population: Hazard ratios and confidence intervals were estimated for patients starting dabigatran or warfarin after matching on the propensity score.
| Arm | Measure | Value (NUMBER) |
|---|---|---|
| Dabigatran | Incidence Rate of Major Gastrointestinal Bleeding | 37.14 events per 1000 person years |
| Warfarin | Incidence Rate of Major Gastrointestinal Bleeding | 31.63 events per 1000 person years |
Comparison: These current feasibility data are limited by a small sample size, short follow up, and few outcome events resulting in wide 95% confidence intervals. At this early stage, no comparative conclusions are possible. Future data with an increased number of patients are planned.95% CI: [0.79, 2.01]Regression, Cox
Secondary
Incidence Rate of Major Intracranial Bleeding
Events during follow-up (October 2010- 30 June 2012) were identified using ICD 9 codes.
Time frame: From treatment initiation until end of follow-up; up to 20 months
Population: Hazard ratios and confidence intervals were estimated for patients starting dabigatran or warfarin after matching on the propensity score.
| Arm | Measure | Value (NUMBER) |
|---|---|---|
| Dabigatran | Incidence Rate of Major Intracranial Bleeding | 9.65 events per 1000 person years |
| Warfarin | Incidence Rate of Major Intracranial Bleeding | 7.34 events per 1000 person years |
Comparison: These current feasibility data are limited by a small sample size, short follow up, and few outcome events resulting in wide 95% confidence intervals. At this early stage, no comparative conclusions are possible. Future data with an increased number of patients are planned.95% CI: [0.57, 3.7]Regression, Cox
Secondary
Incidence Rate of Major Lower Gastrointestinal Bleeding
Events during follow-up (October 2010- 30 June 2012) were identified using ICD 9 codes.
Time frame: From treatment initiation until end of follow-up; up to 20 months
Population: Hazard ratios and confidence intervals were estimated for patients starting dabigatran or warfarin after matching on the propensity score.
| Arm | Measure | Value (NUMBER) |
|---|---|---|
| Dabigatran | Incidence Rate of Major Lower Gastrointestinal Bleeding | 35.52 events per 1000 person years |
| Warfarin | Incidence Rate of Major Lower Gastrointestinal Bleeding | 30.57 events per 1000 person years |
Comparison: These current feasibility data are limited by a small sample size, short follow up, and few outcome events resulting in wide 95% confidence intervals. At this early stage, no comparative conclusions are possible. Future data with an increased number of patients are planned.95% CI: [0.78, 2.01]Regression, Cox
Secondary
Incidence Rate of Major Other Bleeding
Events during follow-up (October 2010- 30 June 2012) were identified using ICD 9 codes.
Time frame: From treatment initiation until end of follow-up; up to 20 months
Population: Hazard ratios and confidence intervals were estimated for patients starting dabigatran or warfarin after matching on the propensity score.
| Arm | Measure | Value (NUMBER) |
|---|---|---|
| Dabigatran | Incidence Rate of Major Other Bleeding | 32.25 events per 1000 person years |
| Warfarin | Incidence Rate of Major Other Bleeding | 46.41 events per 1000 person years |
Comparison: These current feasibility data are limited by a small sample size, short follow up, and few outcome events resulting in wide 95% confidence intervals. At this early stage, no comparative conclusions are possible. Future data with an increased number of patients are planned.95% CI: [0.48, 1.14]Regression, Cox
Secondary
Incidence Rate of Major Upper Gastrointestinal Bleeding
Events during follow-up (October 2010- 30 June 2012) were identified using ICD 9 and procedure codes.
Time frame: From treatment initiation until end of follow-up; up to 20 months
Population: Hazard ratios and confidence intervals were estimated for patients starting dabigatran or warfarin after matching on the propensity score.
| Arm | Measure | Value (NUMBER) |
|---|---|---|
| Dabigatran | Incidence Rate of Major Upper Gastrointestinal Bleeding | 4.82 events per 1000 person years |
| Warfarin | Incidence Rate of Major Upper Gastrointestinal Bleeding | 7.34 events per 1000 person years |
Comparison: These current feasibility data are limited by a small sample size, short follow up, and very few outcome events resulting in wide 95% confidence intervals. At this early stage, no comparative conclusions are possible. Future data with an increased number of patients are planned.95% CI: [0.21, 1.98]Regression, Cox
Secondary
Incidence Rate of Major Urogenital Bleeding
Events during follow-up (October 2010- 30 June 2012) were identified using ICD 9 codes.
Time frame: From treatment initiation until end of follow-up; up to 20 months
Population: Hazard ratios and confidence intervals were estimated for patients starting dabigatran or warfarin after matching on the propensity score.
| Arm | Measure | Value (NUMBER) |
|---|---|---|
| Dabigatran | Incidence Rate of Major Urogenital Bleeding | 0.0 events per 1000 person years |
| Warfarin | Incidence Rate of Major Urogenital Bleeding | 0.0 events per 1000 person years |
Secondary
Incidence Rate of Myocardial Infarction
Events during follow-up (October 2010- 30 June 2012) were identified using ICD 9 codes.
Time frame: From treatment initiation until end of follow-up; up to 20 months
Population: Hazard ratios and confidence intervals were estimated for patients starting dabigatran or warfarin after matching on the propensity score.
| Arm | Measure | Value (NUMBER) |
|---|---|---|
| Dabigatran | Incidence Rate of Myocardial Infarction | 16.90 events per 1000 person years |
| Warfarin | Incidence Rate of Myocardial Infarction | 14.73 events per 1000 person years |
Comparison: These current feasibility data are limited by a small sample size, short follow up, and few outcome events resulting in wide 95% confidence intervals. At this early stage, no comparative conclusions are possible. Future data with an increased number of patients are planned.95% CI: [0.62, 2.45]Regression, Cox
Secondary
Incidence Rate of Pulmonary Embolism
Events during follow-up (October 2010- 30 June 2012) were identified using ICD 9 codes.
Time frame: From treatment initiation until end of follow-up; up to 20 months
Population: Hazard ratios and confidence intervals were estimated for patients starting dabigatran or warfarin after matching on the propensity score.
| Arm | Measure | Value (NUMBER) |
|---|---|---|
| Dabigatran | Incidence Rate of Pulmonary Embolism | 3.21 events per 1000 person years |
| Warfarin | Incidence Rate of Pulmonary Embolism | 13.65 events per 1000 person years |
Comparison: These current feasibility data are limited by a small sample size, short follow up, and few outcome events resulting in wide 95% confidence intervals. At this early stage, no comparative conclusions are possible. Future data with an increased number of patients are planned.95% CI: [0.09, 0.84]Regression, Cox
Secondary
Incidence Rate of Stroke or Systemic Embolism
Events during follow-up (October 2010- 30 June 2012) were identified using ICD 9 codes.
Time frame: From treatment initiation until end of follow-up; up to 20 months
Population: Hazard ratios and confidence intervals were estimated for patients starting dabigatran or warfarin after matching on the propensity score.
| Arm | Measure | Value (NUMBER) |
|---|---|---|
| Dabigatran | Incidence Rate of Stroke or Systemic Embolism | 39.73 events per 1000 person years |
| Warfarin | Incidence Rate of Stroke or Systemic Embolism | 60.38 events per 1000 person years |
Comparison: These current feasibility data are limited by a small sample size, short follow up, and few outcome events resulting in wide 95% confidence intervals. At this early stage, no comparative conclusions are possible. Future data with an increased number of patients are planned.95% CI: [0.5, 1.08]Regression, Cox
Secondary
Incidence Rate of Stroke Uncertain Classification
Events during follow-up (October 2010- 30 June 2012) were identified using ICD 9.
Time frame: From treatment initiation until end of follow-up; up to 20 months
Population: Hazard ratios and confidence intervals were estimated for patients starting dabigatran or warfarin after matching on the propensity score.
| Arm | Measure | Value (NUMBER) |
|---|---|---|
| Dabigatran | Incidence Rate of Stroke Uncertain Classification | 20.98 events per 1000 person years |
| Warfarin | Incidence Rate of Stroke Uncertain Classification | 14.70 events per 1000 person years |
Comparison: These current feasibility data are limited by a small sample size, short follow up, and few outcome events resulting in wide 95% confidence intervals. At this early stage, no comparative conclusions are possible. Future data with an increased number of patients are planned.95% CI: [0.83, 3.08]Regression, Cox
Secondary
Incidence Rate of Systemic Embolism
Events during follow-up (October 2010- 30 June 2012) were identified using ICD 9 codes.
Time frame: From treatment initiation until end of follow-up; up to 20 months
Population: Hazard ratios and confidence intervals were estimated for patients starting dabigatran or warfarin after matching on the propensity score.
| Arm | Measure | Value (NUMBER) |
|---|---|---|
| Dabigatran | Incidence Rate of Systemic Embolism | 11.27 events per 1000 person years |
| Warfarin | Incidence Rate of Systemic Embolism | 30.55 events per 1000 person years |
Comparison: These current feasibility data are limited by a small sample size, short follow up, and few outcome events resulting in wide 95% confidence intervals. At this early stage, no comparative conclusions are possible. Future data with an increased number of patients are planned.95% CI: [0.21, 0.76]Regression, Cox
Secondary
Incidence Rate of Transient Ischemic Attack
Events during follow-up (October 2010- 30 June 2012) were identified using ICD 9 codes.
Time frame: From treatment initiation until end of follow-up; up to 20 months
Population: Hazard ratios and confidence intervals were estimated for patients starting dabigatran or warfarin after matching on the propensity score.
| Arm | Measure | Value (NUMBER) |
|---|---|---|
| Dabigatran | Incidence Rate of Transient Ischemic Attack | 11.27 events per 1000 person years |
| Warfarin | Incidence Rate of Transient Ischemic Attack | 12.61 events per 1000 person years |
Comparison: These current feasibility data are limited by a small sample size, short follow up, and few outcome events resulting in wide 95% confidence intervals. At this early stage, no comparative conclusions are possible. Future data with an increased number of patients are planned.95% CI: [0.47, 2.2]Regression, Cox
Secondary
Incidence Rate of Venous Thromboembolism
Events during follow-up (October 2010- 30 June 2012) were identified using ICD 9 codes.
Time frame: From treatment initiation until end of follow-up; up to 20 months
Population: Hazard ratios and confidence intervals were estimated for patients starting dabigatran or warfarin after matching on the propensity score.
| Arm | Measure | Value (NUMBER) |
|---|---|---|
| Dabigatran | Incidence Rate of Venous Thromboembolism | 8.83 events per 1000 person years |
| Warfarin | Incidence Rate of Venous Thromboembolism | 34.79 events per 1000 person years |
Comparison: These current feasibility data are limited by a small sample size, short follow up, and few outcome events resulting in wide 95% confidence intervals. At this early stage, no comparative conclusions are possible. Future data with an increased number of patients are planned.95% CI: [0.15, 0.59]Regression, Cox