CSP #716B - The 80+ Hemorrhagic Cohort Study
CSP #716B - The 80+ Hemorrhagic Cohort Study
This is a prospective observational cohort study that is in the process of enrolling and following patients on warfarin therapy by utilizing anticoagulation clinics in VA hospitals across the US. There are currently 14 sites actively enrolling patients in the study. For each octogenarian enrolled a randomly selected patient with atrial fibrillation (AF) younger than 80 years of age from the same clinic and receiving warfarin for approximately the same duration is enrolled.
Objectives: The primary objectives are: (1) to quantify the absolute risk of major hemorrhagic complications associated with anticoagulation for AF among octogenarians; (2) to determine if risk and severity of major hemorrhagic complications is increased in octogenarians compared with younger patients maintained on warfarin for AF after adjustment for potential confounders; (3) to explore predictors of major hemorrhagic complications among octogenarians maintained on warfarin for AF; and (4) to determine the effectiveness of utilizing the VA Intranet to collect and manage patient data in the conduct of longitudinal studies.
Research Plan: This is a prospective observational cohort study that is in the process of enrolling and following patients on warfarin therapy by utilizing anticoagulation clinics in VA hospitals across the US. There are currently 14 sites actively enrolling patients in the study. For each octogenarian enrolled a randomly selected patient with AF younger than 80 years of age from the same clinic and receiving warfarin for approximately the same duration is enrolled.
Methods: Baseline data is collected and prospective follow-up data is captured every 3 months. Based on their duration of previous warfarin therapy, octogenarians with AF are assembled into either a survivor or an inception cohort. For analysis, patients with two weeks or less of prior warfarin exposure constitute the inception cohort. The primary endpoint is major hemorrhage as defined by the Consensus Recommendations published by the American College of Chest Physicians. Data is being collected to allow for verification of fatal bleeds and verification of the resultant morbidity from non-fatal major bleeds. Endpoints will be adjudicated by hemostasis experts blinded to the age of the patient. Since this is an observational study, all reported events and deaths are not related to or the result of any study intervention. Data on development of new comorbidities, medications, average daily warfarin dose or discontinuation and degree of anticoagulation as measured by the International Normalized Ratio (INR) is also being collected.
All aspects of the study are being conducted through the VA Intranet. Participating sites utilize the Intranet to transmit baseline and follow-up data forms on study patients, removing the need for detailed paper correspondences. Sites also have the opportunity to view and edit previously submitted forms, view the study manual, download paper copies of data collection forms, and utilize the bulletin board to post questions or comments about the study. Site Investigators and Coordinators access the Intranet from terminals in their own facility and are then presented with a unique study menu and bulletin board. To prevent unauthorized access to the data, Site Investigators and Coordinators have each been assigned an Electronic Signature (login name and password) in order to access the data forms on the Intranet. All data entered is submitted directly into the study database, obviating the need for subsequent transcription of data and reducing the opportunity for errors. Additionally, electronic data collection is designed to reject forms when predefined critical fields have not been completed or when values fall outside of field parameters. Data verification is done in real-time such that a site coordinator entering an invalid value is required to either skip the field or re-enter the corrected data to proceed. This should greatly reduce the time consumed searching through patient records to retrieve data that was entered incorrectly originally.
Findings: Our goal is to enroll and follow a total of 700 patients over 5 years of recruitment. To date, 16 sites have enrolled a total of 502 patients into the study, 251 in the "survivor" cohort, 24 in the "inception" cohort, and 227 in the "control" cohort. There have been 66 reported deaths and 54 hemorrhagic events, 23 of which have been confirmed as major bleeds by the adjudication committee. The cohort to date has accumulated a total of 824 person-years of observation.