Atrial Fibrillation (AF) is a heart condition in which patients have an irregular heartbeat. Patients with AF are at a higher risk of stroke and may be prescribed a type of medication called anticoagulants to reduce this risk. Traditionally this treatment has been with an anticoagulant called warfarin, but recently newer options called direct oral anticoagulants (DOACs) have been developed. Warfarin works by stopping the actions of vitamin K, and some studies suggest this may put patients at a greater risk of developing diabetes. Two recent studies investigating this found that patients taking DOACs compared to warfarin had lower risks of diabetes, but results in different age groups and among men and women varied. One of the reasons for these conflicting results might be a lack of power due to a small sample size. The aim of this study is therefore to repeat these analyses using data from the larger databases CPRD GOLD and Aurum. We will calculate the absolute and relative risks of developing diabetes among AF patients taking DOACs vs. warfarin, taking into account clinical differences between patients using statistical modelling. In sensitivity analyses, we will use alternative statistical methods to control for differences between patients. We will also investigate whether any association between DOACs and warfarin varies according age and sex, and according to the level of Vitamin K inhibition, as such a dose-response relationship would strengthen the evidence for a causal association.
Atrial Fibrillation (AF) is a common cardiovascular condition, typically managed with long-term oral anticoagulation to lower the risk of stroke. In the past decade several direct oral anticoagulants (DOACs) have been approved, which provides an alternative to standard treatment with the Vitamin K inhibitor Warfarin. Laboratory and epidemiological evidence suggest levels of Vitamin K might influence insulin sensitivity and therefore the development of type II diabetes. Two recent observational studies from Taiwan and Hong Kong found that patients taking DOACs compared to warfarin had a reduced incidence of type II diabetes, but there were conflicting results for subgroups defined by age and sex, as well as for individual DOACs. The aim of this study is therefore to compare the risk of developing type II diabetes among patients taking DOACs vs. warfarin in a large cohort comprising patients with AF from CPRD Aurum. Secondary objectives include investigating whether this association varies according to age and sex. We will also investigate any potential dose-response relationship according to the level of Vitamin K inhibition (measured using values of the international normalized ratio [INR]). We will use time-to-event analyses and describe the absolute risk of type II diabetes in both cohorts. Multivariable cox regression models, adjusting for a range of pre-specified confounders, will be used to estimate hazard ratios and 95%CI. In sensitivity analyses, we will use inverse probability of treatment weighting by propensity scores to produce absolute risk estimates taking confounding into account.
Incident Type II Diabetes (time-to event)
Anna Schultze - Chief Investigator - London School of Hygiene & Tropical Medicine ( LSHTM )
Anna Schultze - Corresponding Applicant - London School of Hygiene & Tropical Medicine ( LSHTM )
Emma Powell - Collaborator - London School of Hygiene & Tropical Medicine ( LSHTM )
Ian Douglas - Collaborator - London School of Hygiene & Tropical Medicine ( LSHTM )
Kevin Wing - Collaborator - London School of Hygiene & Tropical Medicine ( LSHTM )
Usha Gungabissoon - Collaborator - GSK
Yan-Ling Lu - Collaborator - London School of Hygiene & Tropical Medicine ( LSHTM )
HES Admitted Patient Care;ONS Death Registration Data;Patient Level Index of Multiple Deprivation;Practice Level Index of Multiple Deprivation