Atrial Fibrillation
Conditions
Brief summary
The study is an analysis using the French national health insurance database, six months after the beginning of NOAC launch in the NVAF indication. The aim is to compare the one-year, two-year and three-year benefit-risk (major bleeding, arterial thrombotic events, myocardial infarction (MI), death) between patients starting a NOAC and patients starting a VKA for NVAF in 2013
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
Patients with NVAF with a first reimbursed dispensation of Pradaxa®, Xarelto®, or VKA in 2013, with no other identified indication for anticoagulation; Without any VKA or NOAC (Pradaxa®, Xarelto®, or Eliquis®) reimbursed dispensation for the last 3 years before the first reimbursed dispensation of Pradaxa®, Xarelto®, or VKA
Exclusion criteria
None
Design outcomes
Primary
| Measure | Time frame | Description |
|---|---|---|
| Clinically Relevant Bleeding | One year | First hospitalization with primary diagnosis (Tenth Revision codes of the International Classification of Diseases (ICD-10 codes)) of: 1. Hemorrhagic stroke, 2. Other critical organ or site bleeding, 3. Other bleeding (gastro-intestinal bleeding, urogenital bleeding and other bleeding subtype). |
| Major Bleeding | 1 year | First hospitalization with primary diagnosis (ICD-10 codes) of: 1. Hemorrhagic stroke, 2. Other critical organ or site bleeding, 3. Other bleeding with transfusion, or acute post-hemorrhagic anemia or death during hospital stay. |
| Arterial Thrombotic Event | 1 year | First hospitalization with primary diagnosis (ICD-10 codes) of: 1. Ischemic or undefined stroke, 2. Systemic arterial embolism. |
| Acute Coronary Syndrome | One year | First hospitalization with primary diagnosis (ICD-10 codes) of: 1. Myocardial infarction (ST-segment elevation Myocardial infarction (STEMI) and non-ST-segment elevation Myocardial infarction(NSTEMI)), 2. Unstable angina. |
| Death (All-cause) | 1 year | All-cause death (cause of death not available in the database). |
| Composite Criterion (Clinically Relevant Bleeding, Arterial Thrombotic Events, Acute Coronary Syndrome, Death) | One year | First event among clinically relevant bleeding, arterial thrombotic event, acute coronary syndrome, or death defined above. |
Countries
France
Participant flow
Recruitment details
The study ENGEL 2 is a real-world historical cohort study in the French nationwide healthcare claims and hospitalization database (SNIIRAM) including new users of DOAC or VKA for nonvalvular atrial fibrillation (NVAF) in 2013 with a follow-up for one year (main objective).
Pre-assignment details
The main objective was to compare the risk & effectiveness for dabigatran vs VKA, & for rivaroxaban vs VKA. The main analysis was done for hdPS matched patients (pts.) with atrial fibrillation (AF) diagnosis information in the database.
Participants by arm
| Arm | Count |
|---|---|
| Dabigatran Patients with a first dispensing of dabigatran in 2013 for NVAF, matched 1:1 on the date of the first dispensing, gender, age at index date, and hdPS. | 27,060 |
| Rivaroxaban Patients with a first dispensing of rivaroxaban in 2013 for NVAF, matched 1:1 on the date of the first dispensing, gender, age at index date, and hdPS. | 31,388 |
| Vitamin K Antagonists Patients with a first dispensing of VKA in 2013 for NVAF, matched 1:1 on the date of the first dispensing, gender, age at index date, and hdPS. | 44,653 |
| Total | 103,101 |
Baseline characteristics
| Characteristic | Vitamin K Antagonists | Dabigatran | Total | Rivaroxaban |
|---|---|---|---|---|
| Age, Customized Dabigatran vs VKA (matched pop) | 75.4 Years STANDARD_DEVIATION 10.7 | 75.3 Years STANDARD_DEVIATION 10.7 | 75.4 Years STANDARD_DEVIATION 10.7 | — |
| Age, Customized Overall | 77.9 Years STANDARD_DEVIATION 11.1 | 73.2 Years STANDARD_DEVIATION 11.8 | 75.2 Years STANDARD_DEVIATION 11.7 | 73.2 Years STANDARD_DEVIATION 11.8 |
| Age, Customized Rivaroxaban vs VKA (matched pop) | 75.4 Years STANDARD_DEVIATION 10.7 | — | 75.4 Years STANDARD_DEVIATION 10.7 | 75.3 Years STANDARD_DEVIATION 10.7 |
| Sex/Gender, Customized Dabigatran vs VKA (matched pop) Female | 9325 Participants | 9325 Participants | 18650 Participants | — |
| Sex/Gender, Customized Dabigatran vs VKA (matched pop) Male | 11164 Participants | 11164 Participants | 22328 Participants | — |
| Sex/Gender, Customized Overall Female | 21785 Participants | 11807 Participants | 47327 Participants | 13735 Participants |
| Sex/Gender, Customized Overall Male | 22868 Participants | 15253 Participants | 55774 Participants | 17653 Participants |
| Sex/Gender, Customized Rivaroxaban vs VKA (matched pop) Female | 10496 Participants | — | 20992 Participants | 10496 Participants |
| Sex/Gender, Customized Rivaroxaban vs VKA (matched pop) Male | 12557 Participants | — | 25114 Participants | 12557 Participants |
Adverse events
| Event type | EG000 affected / at risk | EG001 affected / at risk | EG002 affected / at risk |
|---|---|---|---|
| deaths Total, all-cause mortality | — / — | — / — | — / — |
| other Total, other adverse events | 0 / 0 | 0 / 0 | 0 / 0 |
| serious Total, serious adverse events | 0 / 0 | 0 / 0 | 0 / 0 |
Outcome results
Primary
Acute Coronary Syndrome
First hospitalization with primary diagnosis (ICD-10 codes) of: 1. Myocardial infarction (ST-segment elevation Myocardial infarction (STEMI) and non-ST-segment elevation Myocardial infarction(NSTEMI)), 2. Unstable angina.
Time frame: One year
Population: Patients with a first dispensing of DOAC or VKA in 2013 for NVAF. For each comparison (dabigatran versus VKA and rivaroxaban versus VKA), patients were matched 1:1 on the date of the first drug (DOAC or VKA) dispensing (± 2 weeks), gender, age at index date (± 1 year) and high-dimensional propensity score (hdPS, ± 0.05).
| Arm | Measure | Value (NUMBER) |
|---|---|---|
| Dabigatran (Dabigatran vs VKA) | Acute Coronary Syndrome | 176 participants with events |
| VKA (Dabigatran vs VKA) | Acute Coronary Syndrome | 238 participants with events |
| Rivaroxaban (Rivaroxaban vs VKA) | Acute Coronary Syndrome | 230 participants with events |
| VKA (Rivaroxaban vs VKA) | Acute Coronary Syndrome | 277 participants with events |
Comparison: Outcome was analyzed during dabigatran or VKA exposure period (on treatment) using: 1) Cumulative function to estimate the incidence (with death as a competing risk), 2) Fine and Gray model to compare the 1-year risk for dabigatran versus VKA, with hazard ratio (HR) and 95%CI (and death as a competing risk).p-value: 0.014795% CI: [0.65, 0.95]Fine and Gray model
Comparison: Outcome was analyzed during rivaroxaban or VKA exposure period (on treatment) using: 1) Cumulative function to estimate the incidence (with death as a competing risk), 2) Fine and Gray model to compare the 1-year risk for rivaroxaban versus VKA, with HR and 95%CI (and death as a competing risk).p-value: 0.050195% CI: [0.71, 1]Fine and Gray model
Primary
Arterial Thrombotic Event
First hospitalization with primary diagnosis (ICD-10 codes) of: 1. Ischemic or undefined stroke, 2. Systemic arterial embolism.
Time frame: 1 year
Population: Patients with a first dispensing of DOAC or VKA in 2013 for NVAF. For each comparison (dabigatran versus VKA and rivaroxaban versus VKA), patients were matched 1:1 on the date of the first drug (DOAC or VKA) dispensing (± 2 weeks), gender, age at index date (± 1 year) and high-dimensional propensity score (hdPS, ± 0.05).
| Arm | Measure | Value (NUMBER) |
|---|---|---|
| Dabigatran (Dabigatran vs VKA) | Arterial Thrombotic Event | 226 participants with events |
| VKA (Dabigatran vs VKA) | Arterial Thrombotic Event | 321 participants with events |
| Rivaroxaban (Rivaroxaban vs VKA) | Arterial Thrombotic Event | 343 participants with events |
| VKA (Rivaroxaban vs VKA) | Arterial Thrombotic Event | 351 participants with events |
Comparison: Outcome was analyzed during dabigatran or VKA exposure period (on treatment) using: 1) Cumulative function to estimate the incidence (with death as a competing risk), 2) Fine and Gray model to compare the 1-year risk for dabigatran versus VKA, with hazard ratio (HR) and 95%CI (and death as a competing risk).p-value: 0.000795% CI: [0.63, 0.88]Fine and Gray model
Comparison: Outcome was analyzed during rivaroxaban or VKA exposure period (on treatment) using: 1) Cumulative function to estimate the incidence (with death as a competing risk), 2) Fine and Gray model to compare the 1-year risk for rivaroxaban versus VKA, with HR and 95%CI (and death as a competing risk).p-value: 0.834195% CI: [0.85, 1.14]Fine and Gray model
Primary
Clinically Relevant Bleeding
First hospitalization with primary diagnosis (Tenth Revision codes of the International Classification of Diseases (ICD-10 codes)) of: 1. Hemorrhagic stroke, 2. Other critical organ or site bleeding, 3. Other bleeding (gastro-intestinal bleeding, urogenital bleeding and other bleeding subtype).
Time frame: One year
Population: Patients (pts) with a first dispensing (dispen.) of DOAC or VKA in 2013 for NVAF. For each comparison (dabigatran versus VKA and rivaroxaban versus VKA), patients were matched 1:1 on the date of the first drug (DOAC or VKA) dispensing (± 2 weeks (wks.)), gender, age at index date (± 1 year (yr)) and high-dimensional propensity score (hdPS, ± 0.05).
| Arm | Measure | Value (NUMBER) |
|---|---|---|
| Dabigatran (Dabigatran vs VKA) | Clinically Relevant Bleeding | 367 participants with event |
| VKA (Dabigatran vs VKA) | Clinically Relevant Bleeding | 668 participants with event |
| Rivaroxaban (Rivaroxaban vs VKA) | Clinically Relevant Bleeding | 635 participants with event |
| VKA (Rivaroxaban vs VKA) | Clinically Relevant Bleeding | 767 participants with event |
Comparison: Outcome was analyzed during dabigatran or VKA exposure period (on treatment) using: 1) Cumulative function to estimate the incidence (with death as a competing risk), 2) Fine and Gray model to compare the 1-year risk for dabigatran versus VKA, with hazard ratio (HR) and 95% Confidence Interval (CI ) (and death as a competing risk).p-value: <0.000195% CI: [0.51, 0.66]Fine and Gray model
Comparison: Outcome was analyzed during rivaroxaban or VKA exposure period (on treatment) using: 1) Cumulative function to estimate the incidence (with death as a competing risk), 2) Fine and Gray model to compare the 1-year risk for rivaroxaban versus VKA, with HR and 95%CI (and death as a competing risk).p-value: 0.000695% CI: [0.75, 0.92]Fine and Gray model
Primary
Composite Criterion (Clinically Relevant Bleeding, Arterial Thrombotic Events, Acute Coronary Syndrome, Death)
First event among clinically relevant bleeding, arterial thrombotic event, acute coronary syndrome, or death defined above.
Time frame: One year
Population: Patients with a first dispensing of DOAC or VKA in 2013 for NVAF. For each comparison (dabigatran versus VKA and rivaroxaban versus VKA), patients were matched 1:1 on the date of the first drug (DOAC or VKA) dispensing (± 2 weeks), gender, age at index date (± 1 year) and high-dimensional propensity score (hdPS, ± 0.05).
| Arm | Measure | Value (NUMBER) |
|---|---|---|
| Dabigatran (Dabigatran vs VKA) | Composite Criterion (Clinically Relevant Bleeding, Arterial Thrombotic Events, Acute Coronary Syndrome, Death) | 1340 participants with events |
| VKA (Dabigatran vs VKA) | Composite Criterion (Clinically Relevant Bleeding, Arterial Thrombotic Events, Acute Coronary Syndrome, Death) | 1970 participants with events |
| Rivaroxaban (Rivaroxaban vs VKA) | Composite Criterion (Clinically Relevant Bleeding, Arterial Thrombotic Events, Acute Coronary Syndrome, Death) | 1967 participants with events |
| VKA (Rivaroxaban vs VKA) | Composite Criterion (Clinically Relevant Bleeding, Arterial Thrombotic Events, Acute Coronary Syndrome, Death) | 2328 participants with events |
Comparison: Outcome was analyzed during dabigatran or VKA exposure period (on treatment) using: 1) Kaplan-Meier to estimate the cumulative incidence, 2) Cox proportional hazard risk model to compare the 1-year risk for dabigatran versus VKA, with HR and 95%CI.p-value: <0.000195% CI: [0.66, 0.76]Cox proportional hazard risk model
Comparison: Outcome was analyzed during rivaroxaban or VKA exposure period (on treatment) using: 1) Kaplan-Meier to estimate the cumulative incidence, 2) Cox proportional hazard risk model to compare the 1-year risk for rivaroxaban versus VKA, with HR and 95%CI.p-value: <0.000195% CI: [0.79, 0.89]Cox proportional hazard risk model
Primary
Death (All-cause)
All-cause death (cause of death not available in the database).
Time frame: 1 year
Population: Patients with a first dispensing of DOAC or VKA in 2013 for NVAF. For each comparison (dabigatran versus VKA and rivaroxaban versus VKA), patients were matched 1:1 on the date of the first drug (DOAC or VKA) dispensing (± 2 weeks), gender, age at index date (± 1 year) and high-dimensional propensity score (hdPS, ± 0.05).
| Arm | Measure | Value (NUMBER) |
|---|---|---|
| Dabigatran (Dabigatran vs VKA) | Death (All-cause) | 686 participants with events |
| VKA (Dabigatran vs VKA) | Death (All-cause) | 983 participants with events |
| Rivaroxaban (Rivaroxaban vs VKA) | Death (All-cause) | 908 participants with events |
| VKA (Rivaroxaban vs VKA) | Death (All-cause) | 1186 participants with events |
Comparison: Outcome was analyzed during dabigatran or VKA exposure period (on treatment) using: 1) Kaplan-Meier to estimate the cumulative incidence, 2) Cox proportional hazard risk model to compare the 1-year risk for dabigatran versus VKA, with HR and 95%CI.p-value: <0.000195% CI: [0.67, 0.82]Cox proportional hazard risk model
Comparison: Outcome was analyzed during rivaroxaban or VKA exposure period (on treatment) using: 1) Kaplan-Meier to estimate the cumulative incidence, 2) Cox proportional hazard risk model to compare the 1-year risk for rivaroxaban versus VKA, with HR and 95%CI.p-value: <0.000195% CI: [0.71, 0.84]Cox proportional hazard risk model
Primary
Major Bleeding
First hospitalization with primary diagnosis (ICD-10 codes) of: 1. Hemorrhagic stroke, 2. Other critical organ or site bleeding, 3. Other bleeding with transfusion, or acute post-hemorrhagic anemia or death during hospital stay.
Time frame: 1 year
Population: Patients with a first dispensing of DOAC or VKA in 2013 for NVAF. For each comparison (dabigatran versus VKA and rivaroxaban versus VKA), patients were matched 1:1 on the date of the first drug (DOAC or VKA) dispensing (± 2 weeks), gender, age at index date (± 1 year) and high-dimensional propensity score (hdPS, ± 0.05).
| Arm | Measure | Value (NUMBER) |
|---|---|---|
| Dabigatran (Dabigatran vs VKA) | Major Bleeding | 178 participants with events |
| VKA (Dabigatran vs VKA) | Major Bleeding | 341 participants with events |
| Rivaroxaban (Rivaroxaban vs VKA) | Major Bleeding | 280 participants with events |
| VKA (Rivaroxaban vs VKA) | Major Bleeding | 417 participants with events |
Comparison: Outcome was analyzed during dabigatran or VKA exposure period (on treatment) using: 1) Cumulative function to estimate the incidence (with death as a competing risk), 2) Fine and Gray model to compare the 1-year risk for dabigatran versus VKA, with hazard ratio (HR) and 95%CI (and death as a competing risk).p-value: <0.000195% CI: [0.46, 0.66]Fine and Gray model
Comparison: Outcome was analyzed during rivaroxaban or VKA exposure period (on treatment) using: 1) Cumulative function to estimate the incidence (with death as a competing risk), 2) Fine and Gray model to compare the 1-year risk for rivaroxaban versus VKA, with HR and 95%CI (and death as a competing risk).p-value: <0.000195% CI: [0.58, 0.79]Fine and Gray model