Background: Symptomatic venous thromboembolism (VTE) occurs in about 1% of patients within 3 months after admission to a medicine ward.1,2 The 9th Edition American College of Chest Physicians Guidelines recommends the use of pharmacologic VTE prophylaxis in acutely ill hospitalized medical patients at increased risk of thrombosis and suggests against its use in low risk patients. Reviewing internal data in our hospital system, we found that roughly 30% of low risk patients were inappropriately receiving pharmacologic VTE prophylaxis.3
Purpose: Our objective was to create an EMR-integrated orderset requiring a documented VTE risk assessment to ultimately reduce the inappropriate use of pharmacologic VTE prophylaxis.
Description: The VTE risk assessment order set was implemented on December 15, 2020. The VTE risk assessment order set is both a required field within the general admission order set and a stand-alone order set. If “High Risk” for VTE is chosen, the user is then prompted to order VTE prophylaxis directly from the order set; however if “low risk” is chosen, the user is prompted to not order VTE prophylaxis (figure 1). To reinforce using the VTE risk assessment order set, a Best Practice Advisory (BPA) was implemented on October 15, 2021. The BPA displays if a provider attempts to order VTE prophylaxis on a patient previously deemed “low risk” (figure 2). Data regarding rate of pharmacologic prophylaxis and rate of documentation of VTE risk assessment were collected. We compared data from the pre-intervention period (June 17, 2020 – December 14, 2020), the first post-intervention period (December 15, 2020 – October 15th), and second post-intervention period (October 15th – November 7, 2021). The rate of pharmacologic prophylaxis decreased significantly after each intervention. Between the pre-intervention and first post-intervention periods, the rate of pharmacologic prophylaxis decreased by 5.5% (from 99.64 to 94.18 orders/1000 patient days, p value < 0.01). Between the first post-intervention and second post-intervention periods, the rate further decreased by 6.9% (94.18 to 87.32 orders/1000 patient days, p value < 0.01). Overall, pharmacologic VTE prophylaxis reduced by 12.4% after both interventions. The rate of VTE risk documentation increased from 20% during the pre-intervention period to 56% during the first post-intervention period, and did not significantly change from first post-intervention to second post-intervention period (56.0% to 54.6%).
Conclusions: Implementation of a VTE risk assessment order set and subsequent BPA to reinforce its use resulted in a decrease in the VTE prophylaxis orders in our hospital system. Our findings suggest that a simple VTE risk assessment tool can decrease the inappropriate use of pharmacologic VTE prophylaxis and improve documentation of VTE risk assessments.
