External validation of the society of thoracic surgeons general thoracic surgery database

Mitchell J. Magee, Cameron D. Wright, Donna McDonald, Felix G. Fernandez, Benjamin D. Kozower

Research output: Contribution to journalArticle

30 Scopus citations

Abstract

Background The Society of Thoracic Surgeons (STS) General Thoracic Surgery Database (GTSD) reports outstanding results for lung and esophageal cancer resection. However, a major weakness of the GTSD has been the lack of validation of this voluntary registry. The purpose of this study was to perform an external, independent audit to assess the accuracy of the data collection process and the quality of the database. Methods An independent firm was contracted to audit 5% of sites randomly selected from the GTDB in 2011. Audits were performed remotely to maximize the number of audits performed and reduce cost. Auditors compared lobectomy cases submitted to the GTSD with the hospital operative logs to evaluate completeness of the data. In addition, 20 lobectomy records from each site were audited in detail. Agreement rates were calculated for 32 individual data elements, 7 data categories pertaining to patient status or care delivery, and an overall agreement rate for each site. Six process variables were also evaluated to assess best practice for data collection and submission. Results Ten sites were audited from the 222 participants. Comparison of the 559 submitted lobectomy cases with operative logs from each site identified 28 omissions, a 94.6% agreement rate (discrepancies/site range, 2 to 27). Importantly, cases not submitted had no mortality or major morbidity, indicating a lack of purposeful omission. The aggregate agreement rates for all categories were greater than 90%. The overall data accuracy was 94.9%. Conclusions External audits of the GTSD validate the accuracy and completeness of the data. Careful examination of unreported cases demonstrated no purposeful omission or gaming. Although these preliminary results are quite good, it is imperative that the audit process is refined and continues to expand along with the GTSD to insure reliability of the database. The audit results are currently being incorporated into educational and quality improvement processes to add further value.

Original languageEnglish
Pages (from-to)1734-1739
Number of pages6
JournalAnnals of Thoracic Surgery
Volume96
Issue number5
DOIs
StatePublished - Nov 1 2013
Externally publishedYes

Fingerprint Dive into the research topics of 'External validation of the society of thoracic surgeons general thoracic surgery database'. Together they form a unique fingerprint.

  • Cite this