An Electronic Tool for Venous Thromboembolism Prevention in Medical and Surgical Patients

Published in: Haematologica, v. 64, no. 1, Jan. 2006, p. 64-70

Posted on on January 01, 2006

by Meyer-Michel Samama, Ola E. Dahl, Patrick Mismetti, Daniel J. Quinlan, Nadia Rosencher, Marleen Cornelis, Han de Vries, Ineke van Beusekom, James P. Kahan

Read More

Access further information on this document at

This article was published outside of RAND. The full text of the article can be found at the link above.

BACKGROUND AND OBJECTIVES: Venous thromboembolism (VTE) is a complex disorder influenced by numerous risk factors, and occurs frequently in at-risk hospitalized patients. Because appropriate prevention with thromboprophylaxis is underused, the authors wanted to create an electronic tool to provide a simple risk assessment and suggest appropriate prophylaxis. DESIGN AND METHODS: To develop the risk matrix, iterative rating of odds ratios was performed for 60 predisposing VTE risk factors, using analytical methods that account for multiple risk factors in a single patient and their non-independence. For exposing risk factors, a single score was assigned to each set of factors, both medical (25 items) and surgical conditions (144 items). A CART regression model was used to integrate the risk scales into a 4-level measure of overall risk. The validity of the level of risk and the appropriateness of 11 different prophylactic approaches was assessed using the RAND/UCLA appropriateness method and validated by expert opinion ratings (n=1998) on sample case scenarios (n=108). RESULTS: Correlation between the level of risk calculated by the risk matrix and that offered by expert opinion for individual surgical and medical clinical cases was high (65% and 70%, respectively). The matrix over-estimated the level of risk, compared with that offered by expert opinion, in 28% and 20% of surgical and medical cases, respectively, but the appropriate prophylaxis suggested was no different. Between-expert agreement on the appropriateness of the prophylaxis recommendations was high (90-94% of indications). Interpretation and CONCLUSIONS: This computer-based electronic tool for individualized assessment of venous thromboembolic risk successfully identified both the perceived risk of thrombosis and the appropriate prophylactic approach for medical and surgical patients.

Research conducted by

This report is part of the RAND Corporation external publication series. Many RAND studies are published in peer-reviewed scholarly journals, as chapters in commercial books, or as documents published by other organizations.

The RAND Corporation is a nonprofit institution that helps improve policy and decisionmaking through research and analysis. RAND's publications do not necessarily reflect the opinions of its research clients and sponsors.