Venous thromboembolism (VTE) is a known complication of hospitalization; implementation of proven prophylactic modalities should reduce VTE rates. One potential strategy is the use of computerized order sets. The objective of this study is to determine the potential impact of order sets on VTE rates in VA.
VA administrative data identified admissions to VAMCs in fiscal year (FY) 2010. ICD9 codes were used to identify all patients with a VTE either as a secondary diagnosis at an index admission or as primary or secondary diagnosis for any readmission within 30 days; baseline annual rates of VTE were calculated from FY200509. VTE order set use was determined from a survey sent to VAMCs asking if VTE prophylaxis order sets were used in FY2010. Logistic regression was utilized to determine adjusted odds ratios (OR) adjusting for comorbidity, age, gender, facility volume, and bed section.
Surveys were sent to all 128 VAMCs with inpatient services with 90 returned (70%). VTE order sets were used by 82%. The annual incidence of VTE increased steadily from 1.64% in FY2005 to 2.20% in FY2010. A total of 345,062 admissions occurred in FY10 to these 90 VAMCs, of which 2.20% experienced a VTE (VTE order set present = 2.24%; order set not present = 2.0%). For all patients, VAMCs lacking VTE order sets experienced similar rates of VTE as those with order sets (OR = 0.89, 95% CI 0.781.02; P = 0.08). Similar results were observed for medical (OR = 0.91, 95% CI 0.791.04; P = 0.16), ICU (OR = 0.82, 95% CI 0.661.01; P = 0.06), and surgical (OR = 0.90, 95% CI 0.731.11; P = 0.08) patients. For both unadjusted and adjusted rates, VAMCs with order sets had higher absolute VTE rates.
Despite studies showing computerbased VTE order sets reduce VTE rates, we were unable to find a similar association across an entire health system in which 82% of hospitals used order sets. This finding suggests that the simple presence of clinical order sets may not improve quality. As quality improvement initiatives, such as order sets, are implemented, it is important to design measurement systems to track performance. It is critical to understand what benefits are gained and whether unintended consequences occur.
To cite this abstract:Kartha A, Restuccia J, Glasgow J, Meterko M, VaughanSarrazin M, McIntosh N, Kaboli P, White R, Johnson S. Computerized Patient Record Order Sets for Venous Thromboembolism (Vte) Prophylaxis and Associated Rates of Vte. Abstract published at Hospital Medicine 2012, April 1-4, San Diego, Calif. Abstract 97610. Journal of Hospital Medicine. 2012; 7 (suppl 2). https://www.shmabstracts.com/abstract/computerized-patient-record-order-sets-for-venous-thromboembolism-vte-prophylaxis-and-associated-rates-of-vte/. Accessed May 26, 2019.