Abstracts

RETURN TO ABSTRACT LISTING


E1503. Incidence of DVT in a County Safety Net Hospital vs. a Tertiary Care Center Using Structured Reporting: Results of >50,000 Examinations
Authors
  1. Travis Browning; UT Southwestern Medical Center
  2. Ali Tejani; UT Southwestern Medical Center
  3. Yin Xi; UT Southwestern Medical Center
  4. David Fetzer; UT Southwestern Medical Center
  5. Julia Fielding; UT Southwestern Medical Center
Objective:
This study compares the incidence of deep vein thrombosis (DVT) on extremity Venous Doppler ultrasound (VDU) at a county safety net hospital and affiliated tertiary care hospital using structured reporting templates with fixed impression picklists.

Materials and Methods:
Reports for VDU (01/20/2016 to 07/20/2019) were analyzed at a county safety net (S) and affiliated tertiary care (T) hospital. Structured reports have a fixed picklist impression: acute, calf only, or known DVT, chronic post-thrombotic change, and superficial vein thrombus. The same group of interpreters utilized the same diagnostic criteria at each site. Additionally, sonographers followed similar imaging protocol at both sites. Multivariable logistic regression was used to estimate the independent contribution of all variables and their 2-way interactions. DVT rate from the two hospitals was estimated while controlling for all possible confounders. All analyses were performed in SAS 9.4.

Results:
Of 54181 reports, 52884 (97.6%) utilized structured reports with 7254 exams (13.7%) positive for any DVT (3467, 11.3% at Hospital S compared to 3607, 17.7%, at Hospital T). 5462 exams (10.3%) were positive for acute/calf DVT (2687, 8.3% at Hospital S compared to 2775, 13.6%, at Hospital T). 58% of exams at Hospital S were ordered by the emergency department (ED) with 10.1% positive for any DVT (versus 7.4% at Hospital T, p<0.001) and 7.4% for acute/calf DVT (versus 5.1% at Hospital T, p<0.001). 64% of exams at Hospital T were ordered by inpatient services (IP) with 20.2% positive for any DVT (versus 12.8% at Hospital S, p<0.001) and 16.9% for acute/calf DVT (versus 10.6% at Hospital S, p<0.001).

Conclusion:
System-wide use of structured reporting can help monitor imaging interpretation outcomes in both safety net and tertiary care settings without manual review, which can be used to identify trends in practice patterns and test utilization based on test outcome as well as volume. We identify a higher utilization of VDU and positive test outcome in the ED at the safety net hospital and higher utilization and positive outcome in IP at the tertiary care hospital in affiliated systems. These results might reflect underlying differences between patient populations, such as increased ED demand secondary to patient volume in the safety net hospital versus interim development of DVT in patients with other primary problems admitted to the tertiary care hospital.