RESUMEN
BACKGROUND: Deep venous thrombosis and pulmonary embolus are leading preventable causes of death after surgery. Venous thromboembolism (VTE) prophylaxis management guidelines, with evidenced-based recommendations, are available in the literature. However, over 40% of "at-risk" surgical patients fail to receive appropriate VTE prophylaxis. Decision support-based interventions to reduce venous thromboembolic events were explored. METHODS: A venous thromboembolic risk stratification tool embedded in the electronic medical record, Epic, linking risk category to venous thromboembolic prophylaxis order sets was created, implemented, and analyzed for general surgery patients. Logistic regression analysis was used to compare rates of venous thromboembolic events before and after the intervention, controlling for age, gender, race, body mass index, inpatient status, transfer status, elective/emergent case status, American Society of Anesthesiologists classification, and wound classification. RESULTS: Venous thromboembolic events in the preintervention and postintervention periods were 55 (1.25%) and 12 (0.64%), respectively (P = 0.033). All-cause mortality events decreased after intervention from 49 (1.12%) to 14 (0.75%; P = 0.187). Multivariable analyses show that the risk of a venous thromboembolic event after intervention was half (odds ratio = 0.532; 95% confidence interval, 0.284-0.997; P = 0.049) as likely compared to that in the preintervention period. From 2012 to 2015, our institution moved from the ninth decile (poor) to the first decile (best) for the incidence of venous thromboembolic events among 760 National Surgical Quality Improvement Program hospitals across the nation. CONCLUSIONS: Postoperative thromboembolic events decreased after implementation of a VTE risk stratification tool, linking risk category to venous thromboembolic prophylaxis order sets, embedded in the electronic medical record, Epic.