Comparing Direct Oral Anticoagulants to Vitamin K Antagonists on bleeding and cardiovascular outcomes in patients with atrial fibrillation: effect of using a common protocol only versus a syntactically harmonized Common Data Model

Study type
Date of Approval
Study reference ID
Lay Summary

The use of electronic health records from several countries in the assessment of drug safety and effectiveness can provide us with more data across different populations allowing us to make more robust assessments. This allows researchers to compare the effects of drugs between countries, regions and healthcare systems; investigate specific groups of patients or identify rarer side-effects. However, such studies have several additional complexities, because recording and storage of patient information might be different in different data sources, making comparisons and overall assessments more challenging. There are several methods to address these differences which include a) the use of a single, but common protocol that can be followed and b) restructuring all databases into a single standard format, known as a common data model.
We will compare these differences using the electronic healthcare records of two European countries, the United Kingdom (using CPRD GOLD) and the Netherlands (PHARMO Database Network). We will measure the risk of several clinical events (stroke, major bleeding, cardiovascular disease and glaucoma) after the use of drugs used to prevent blood clots in patients with atrial fibrillation, a condition where parts of the heart are out of sync. We will carry out the case study in the two countries twice with two methods method to combat database differences to see if there are differences in the results: one which uses only a common protocol and a one which uses a well implemented European common data model.

Technical Summary

In the last years many studies assessed whether the risk of major bleeding events among users of Direct Oral Anticoagulants (DOACs) is different compared to use of Vitamin K Antagonists (VKAs). In this project among patients with non-valvular atrial fibrillation we will also measure and compare the risk of stroke, major bleeding, cardiovascular disease, glaucoma and hip fractures following use of DOACs/VKAs. Incidence rates of outcome events and Cox proportional hazards analysis will be used to calculate hazard ratios (HRs) in the United Kingdom (as well as in the Netherlands). The other objective is methodological and addresses the impact of using different data harmonisation methods. Nowadays, use of multiple health databases is a preferred method for generating evidence on the safety and effectiveness of licensed medicines. This allows researchers to compare drug safety and effectiveness between countries, regions and healthcare systems; investigate specific groups of patients or identify rarer outcomes. However, these larger studies have an additional number of complexities. In part, these are due heterogeneity amongst data sources, a challenge only amplified when using multiple internationally distributed databases. Harmonisation methods such as the use of a common protocol (CP) and/or common data model (CDM) can mitigate bias and improve precision. A CDM can broadly be characterised as either developed in accordance with a protocol to fit study-specific data (i.e. a study specific CDM) or prior to a specific research protocol (i.e. a general CDM). In this study, we will focus on comparing the effect estimates and baseline characteristics when using the ConcePTION CDM, a syntactically harmonising CDM filled with study-specific data and a common protocol only approach that does not change the original data. Better understanding of such harmonisation methods will guide the selection/interpretation of epidemiological methods, allowing for more transparent and informed studies, thereby indirectly improving patient care.

Health Outcomes to be Measured

Risk of stroke (haemorrhagic, ischaemic, or unclassified but not transient ischaemic attack), major bleeding (defined in line with previous studies within our group and based on the International Society on Thrombosis and Haemostasis definition as a symptomatic bleeding in an organ or other critical area), cardiovascular disease (acute myocardial infarction, stroke, acute coronary syndrome, other ischemic heart disease), glaucoma and hip fracture. See code lists in the appendix.


Olaf Klungel - Chief Investigator - Utrecht University
Patrick Souverein - Corresponding Applicant - Utrecht University
Helga Gardarsdottir - Collaborator - Utrecht University
Marloes Bazelier - Collaborator - Utrecht University
Nicholas Hunt - Collaborator - Utrecht University