Understanding Differences in Administrative and Audited Patient Data in Cardiac Surgery: Comparison of the University HealthSystem Consortium and Society of Thoracic Surgeons Databases - 27/09/16

Doi : 10.1016/j.jamcollsurg.2016.06.393

Anjali Prasad, MBBS ^a, Meghana R. Helder, MD ^a, Dwight A. Brown, MIM, NRP ^b, Hartzell V. Schaff, MD ^a,^⁎
^a Division of Cardiovascular Surgery, Mayo Clinic, Rochester, MN
^b Quality Management Services, Mayo Clinic, Rochester, MN

^∗Correspondence address: Hartzell V Schaff, MD, Division of Cardiovascular Surgery, Mayo Clinic, 200 First St SW, Rochester, MN 55905.Division of Cardiovascular SurgeryMayo Clinic200 First St SWRochesterMN55905

Abstract

Background

The University HealthSystem Consortium (UHC) administrative database has been used increasingly as a quality indicator for hospitals and even individual surgeons. We aimed to determine the accuracy of cardiac surgical data in the administrative UHC database vs data in the clinical Society of Thoracic Surgeons database.

Study Design

We reviewed demographic and outcomes information of patients with aortic valve replacement (AVR), mitral valve replacement (MVR), and coronary artery bypass grafting (CABG) surgery between January 1, 2012, and December 31, 2013. Data collected in aggregate and compared across the databases included case volume, physician specialty coding, patient age and sex, comorbidities, mortality rate, and postoperative complications.

Results

In these 2 years, the UHC database recorded 1,270 AVRs, 355 MVRs, and 1,473 CABGs. The Society of Thoracic Surgeons database case volumes were less by 2% to 12% (1,219 AVRs; 316 MVRs; and 1,442 CABGs). Errors in physician specialty coding occurred in UHC data (AVR, 0.6%; MVR, 0.8%; and CABG, 0.7%). In matched patients from each database, demographic age and sex information was identical. Although definitions differed in the databases, percentages of patients with at least one comorbidity were similar. Hospital mortality rates were similar as well, but postoperative recorded complications differed greatly.

Conclusions

In comparing the 2 databases, we found similarity in patient demographic information and percentage of patients with comorbidities. The small difference in volumes of each operation type and the larger disparity in postoperative complications between the databases were related to differences in data definition, data collection, and coding errors.

Le texte complet de cet article est disponible en PDF.

Abbreviations and Acronyms : AVR, CABG, CDB, MVR, STS, UHC