OBJECTIVES:

Effective stroke prevention with oral anticoagulation is the cornerstone of management in patients with atrial fibrillation (AF). However, previous literature has shown that guideline-concordant anticoagulation in patients with high stroke risk is often below 70% in the U.S. The purpose of this study is to evaluate the impact of clinical decision support (CDS) tools on anticoagulation rates.

METHODS:

This retrospective, longitudinal study used EHR data from an integrated delivery network in Texas. An electronic alert containing an anticoagulation order set was displayed to providers in outpatient settings caring for non-anticoagulated AF patients at increased stroke risk according to their CHA₂DS₂VASc score. Patients were age ≥18 years with an AF diagnosis and ≥1 outpatient encounter. An interrupted time series design was used to evaluate the effect of this decision support tool on anticoagulant prescribing rates. The pre-intervention period was April 2020-September 2020, and the post-intervention period was September 2020-November 2022. Anticoagulant prescribing rates were assessed weekly for 18 weeks before, and 117 weeks after implementation of the alert.

RESULTS:

123,764 unique AF patients had outpatient encounters with mean (SD) age 72.7 (12.6) years, 56.6% male, and mean (SD) CHA₂DS₂VASc of 3.4 (1.8). The mean (SD) number of weekly eligible patients was 10,514 (1,360) and the median (IQR) was 10,763 (10,325–11,115). Mean (SD) weekly anticoagulant prescribing rates were 60.1% (0.5%) before implementation compared to 65.6% (2.7%) after implementation of CDS.

Before the intervention, rates increased at a weekly trend of 0.06%, (SE=0.03%, P=0.0502). Immediately after BPA implementation, rates increased by 0.38%, (SE=0.30%, P=0.2167) and by 0.02%, (SE=0.03%, P=0.4266) over the 117-week period following implementation. None of these increases were statistically significant.

CONCLUSIONS:

Anticoagulant prescribing rates following implementation of clinical decision support were slightly higher than baseline anticoagulation rates, but the trends were not statistically significant.