Impact of a Simple Computerized Decision Support on Venous Thromboembolism Prophylaxis in the Inpatient Setting

1University of Pennsylvania, Philadelphia, PA
2University of Pennsylvania, Philadelphia, PA
3University of Pennsylvania, Philadelphia, PA
4University of Pennsylvania, Philadelphia, PA
5University of Pennsylvania, Philadelphia, PA
6University of Pennsylvania, Philadelphia, PA
7University of Pennsylvania Health System, Philadelphia, PA
8University of Pennsylvania, Philadelphia, PA
9University of Pennsylvania, Philadelphia, PA

Meeting: Hospital Medicine 2010, April 8-11, Washington, D.C.

Abstract number: 146


Venous thromboembolism (VTE) is an important cause of morbidity and mortality in hospitalized patients, so payers and regulators are encouraging hospitals to invest in systematic approaches to prevent them. Here, we describe an initiative to implement VTE prophylaxis best practices through computerized decision support (CDS), and measure its impact.


We integrated content from locally adapted national guidelines into a CDS intervention and implemented it across the 3 hospitals in our health system. The CDS was linked to the inpatient admission order set and prompted physicians to either accept or decline prophylaxis based on assessed patient risk. Those accepting would proceed to an order set grid providing guidance on prophylaxis by indication. Those declining would be prompted for a reason. Study subjects included all adults admitted to the 3 hospitals in our health system in the 12 months before and after the CDS was launched (April 2008). Interrupted time series analyses were used to measure the impact of the CDS on process and outcome measures. Process measures included percentage of patients on any prophylaxis (i.e., pharmacologic or nonpharmacologic), percentage of patients on pharmacologic prophylaxis (i.e., unfractionated heparin, enoxapann, or coumadin), and percentage of patients on recommended prophylaxis (per our local guideline). Outcome measures included hospital‐associated VTE and bleeds. VTE was defined as inpatient encounters with secondary discharge diagnostic codes consistent with PE and DVT. Bleeding was defined similarly using diagnostic codes consistent with bleeding as well as E codes specific for drug‐induced adverse events. Data were extracted from administrative and clinical databases. The study was approved by the institutional review board.


There were minimal differences in the characteristics of the study population before and after the CDS. Overall, the administration of “any prophylaxis' and “recommended prophylaxis” significantly increased after CDS implementation from 57% to 74% and 27 to 42%, respectively (P < 0.01). Using interrupted time series models, the estimated increase in “any prophylaxis” across the 3 hospitals ranged from 5.5 to 12.5%, and the estimated increase in “recommended prophylaxis” ranged from 4.8% to 11.7% (P < 0.011). There was no significant change in the use of “pharmacologic” prophylaxis. The percentage of patients with VTE events did not change in the general population, but decreased from 2.11% to 1.95% in a subset of patients with procedure codes (representing 113 fewer patients with VTE). The number of bleeds did not significantly change.


The CDS improved prophylaxis rates while decreasing VTE. Given its simplicity, this intervention could be used by any hospital with computerized physician order entry to improve prophylaxis and reduce VTE.

Author Disclosure:

C. Umscheid, none; A. Hanish, none; J. Chittams, none; M. Massary, none; C. Kean, none; S. Sood, none; T. Hecht, none; M. Weiner, none.

To cite this abstract:

Umscheid C, Hanish A, Chittams J, Hecht T, Sood S, Kean C, Massary M, Smith C, Werner M. Impact of a Simple Computerized Decision Support on Venous Thromboembolism Prophylaxis in the Inpatient Setting. Abstract published at Hospital Medicine 2010, April 8-11, Washington, D.C. Abstract 146. Journal of Hospital Medicine. 2010; 5 (suppl 1). Accessed March 31, 2020.

« Back to Hospital Medicine 2010, April 8-11, Washington, D.C.