No abstract

Aims There is currently no consensus on whether atrial fibrillation (AF) patients at low risk for stroke (one non-sex-related CHA2DS2-VASc point) should be treated with an oral anticoagulant. Methods and results We conducted a multi-country cohort study in Sweden, Denmark, Norway, and Scotland. In total, 59 076 patients diagnosed with AF at low stroke risk were included. We assessed the rates of stroke or major bleeding during treatment with a non-vitamin K antagonist oral anticoagulant (NOAC), a vitamin K antagonist (VKA), or no treatment, using inverse probability of treatment weighted (IPTW) Cox regression. In untreated patients, the rate for ischaemic stroke was 0.70 per 100 person-years and the rate for a bleed was also 0.70 per 100 person-years. Comparing NOAC with no treatment, the stroke rate was lower [hazard ratio (HR) 0.72; 95% confidence interval (CI) 0.56–0.94], and the rate for intracranial haemorrhage (ICH) was not increased (HR 0.84; 95% CI 0.54–1.30). Comparing VKA with no treatment, the rate for stroke tended to be lower (HR 0.81; 95% CI 0.59–1.09), and the rate for ICH tended to be higher during VKA treatment (HR 1.37; 95% CI 0.88–2.14). Comparing NOAC with VKA treatment, the rate for stroke was similar (HR 0.92; 95% CI 0.70–1.22), but the rate for ICH was lower during NOAC treatment (HR 0.63; 95% CI 0.42–0.94). Conclusion These observational data suggest that NOAC treatment may be associated with a positive net clinical benefit compared with no treatment or VKA treatment in patients at low stroke risk, a question that can be tested through a randomized controlled trial. Key question What is the association between anticoagulant treatment and stroke and bleeding rate, in patients with one non-sex-related risk factor for stroke? Key findings Take-home message These observational data suggest that NOAC treatment may be associated with a positive net clinical benefit compared with no treatment or VKA treatment in patients at low stroke risk, a hypothesis that can be tested through a randomized controlled trial.

Clinical trials are the standard approach for evaluating new treatments, but may lack the power to assess rare outcomes. Trial results are also necessarily restricted to the population considered in the study. The availability of routinely collected healthcare data provides a source of information on the performance of treatments beyond that offered by clinical trials, but the analysis of this type of data presents a number of challenges. Hierarchical methods, which take advantage of known relationships between clinical outcomes, while accounting for bias, may be a suitable statistical approach for the analysis of this data. A study of direct oral anticoagulants in Scotland is discussed and used to motivate a modeling approach. A Bayesian hierarchical model, which allows a stratification of the population into clusters with similar characteristics, is proposed and applied to the direct oral anticoagulant study data. A simulation study is used to assess its performance in terms of outcome detection and error rates.

Recent approaches to the statistical analysis of adverse event (AE) data in clinical trials have proposed the use of groupings of related AEs, such as by system organ class (SOC). These methods have opened up the possibility of scanning large numbers of AEs while controlling for multiple comparisons, making the comparative performance of the different methods in terms of AE detection and error rates of interest to investigators. We apply two Bayesian models and two procedures for controlling the false discovery rate (FDR), which use groupings of AEs, to real clinical trial safety data. We find that while the Bayesian models are appropriate for the full data set, the error controlling methods only give similar results to the Bayesian methods when low incidence AEs are removed. A simulation study is used to compare the relative performances of the methods. We investigate the differences between the methods over full trial data sets, and over data sets with low incidence AEs and SOCs removed. We find that while the removal of low incidence AEs increases the power of the error controlling procedures, the estimated power of the Bayesian methods remains relatively constant over all data sizes. Automatic removal of low‐incidence AEs however does have an effect on the error rates of all the methods, and a clinically guided approach to their removal is needed. Overall we found that the Bayesian approaches are particularly useful for scanning the large amounts of AE data gathered.