Derivation and validation of a simple model to identify venous thromboembolism risk in medical patients.
Am J Med. 2011 Oct;124(10):947-954.e2
Authors: Woller SC, Stevens SM, Jones JP, Lloyd JF, Evans RS, Aston VT, Elliott CG
Abstract
BACKGROUND: Fewer than half of eligible hospitalized medical patients receive appropriate venous thromboembolism (VTE) prophylaxis. One reason for this low rate is the complexity of existing risk assessment models. A simple set of easily identifiable risk factors that are highly predictive of VTE among hospitalized medical patients may enhance appropriate thromboprophylaxis.
METHODS: Electronic medical record interrogation was performed to identify medical admissions from January 1, 2000-December 31, 2007 (n=143,000), and those patients with objectively confirmed VTE during hospitalization or within 90 days following discharge. Putative risk factors most predictive of VTE were identified, and a risk assessment model (RAM) was derived; 46,000 medicine admissions from January 1, 2008-December 31, 2009 served as a validation cohort to test the predictive ability of the RAM. The newly derived RAM was compared with a published VTE assessment tool (Kucher Score).
RESULTS: Four risk factors: previous VTE; an order for bed rest; peripherally inserted central venous catheterization line; and a cancer diagnosis, were the minimal set most predictive of hospital-associated VTE (area under the receiver operating characteristic curve [AUC]=0.874; 95% confidence interval [CI], 0.869-0.880). These risk factors upon validation in a separate population (validation cohort) retained an AUC=0.843; 95% CI, 0.833-0.852. The ability of the 4-element RAM to identify patients at risk of developing VTE within 90 days was superior to the Kucher Score.
CONCLUSIONS: The 4-element RAM identified in this study may be used to identify patients at risk for VTE and improve rates of thromboprophylaxis. This simple and accurate RAM is an alternative to more complicated published VTE risk assessment tools that currently exist.
PMID: 21962315 [PubMed - in process]