Working smarter, not harder: evaluating a population health approach to anticoagulation therapy management.

Abstract

Inappropriate direct acting oral anti-coagulants (DOAC) prescribing increases the risk of adverse events. Population health management tools (PMTs) could help reduce adverse events through the early, efficient identification of questionable prescribing practices, but the impact of such a tool remains unknown. We evaluated the effect of PMT use on questionable DOAC dosing rates within 40 VHA medical centers and whether this effect differed by DOAC indication or agent. Medical centers were divided into PMT user or standard of care (SOC) groups based upon high or low tool access in the prior year. Questionable DOAC dosing rate was defined as the proportion of patients prescribed DOACs who were also flagged by the tool. Chi-square tests were used to determine if PMT user versus SOC groups differed with high (above 15.3%) versus low (below 15.3%) questionable dosing rates. T-tests were used to determine if mean questionable dosing rates significantly differed between the PMT user and SOC groups. DOAC PMT users were classified less frequently as being 'High" questionable dosage rate compared to SOCs (25% PMT vs. 75% SOC, respectively, p = 0.002). DOAC PMT utilization within the overall cohort was associated with a 4.3% absolute reduction in questionable DOAC dosing rates (13.2% PMT vs 17.5% SOC; p = 0.01). Tool use within the atrial fibrillation (AF) subgroup was associated with a 5.1% absolute reduction in questionable dosing rates (10.4% SOC vs. 5.3% PMT, p < 0.001). Tool use was also associated with lower questionable dosing rates in the apixaban (p < 0.001), dabigatran (p = 0.03) and AF plus venous thromboembolism (p < 0.001) subgroups. In our study, PMT use was associated with reduced questionable DOAC dosing, a difference most pronounced within AF patients. A population health approach has the potential to reduce adverse events among patients prescribed DOACs.