Data integrity of the Pediatric Cardiac Critical Care Consortium (PC4) clinical registry. (11th September 2015)

• Record Type: Journal Article
• Authors: Gaies, Michael; Donohue, Janet E.; Willis, Gina M.; Kennedy, Andrea T.; Butcher, John; Scheurer, Mark A.; Alten, Jeffrey A.; William Gaynor, J.; Schuette, Jennifer J.; Cooper, David S.; Jacobs, Jeffrey P.; Pasquali, Sara K.; Tabbutt, Sarah
• Is Part Of: Cardiology in the young. Volume 26:Number 6(2016)
• Page Start: 1090
• Page End: 1096
• Publication Date: 2015-09-11
• Subjects: Cardiac, -- critical care, -- database, -- audit, -- paediatric; Pediatric cardiology -- Periodicals; 618.9212
• DOI: 10.1017/S1047951115001833
• Languages: English
• ISSNs: 1047-9511

Abstract

Abstract: Background: Clinical databases in congenital and paediatric cardiac care provide a foundation for quality improvement, research, policy evaluations and public reporting. Structured audits verifying data integrity allow database users to be confident in these endeavours. We report on the initial audit of the Pediatric Cardiac Critical Care Consortium (PC 4 ) clinical registry. Materials and methods: Participants reviewed the entire registry to determine key fields for audit, and defined major and minor discrepancies for the audited variables. In-person audits at the eight initial participating centres were conducted during a 12-month period. The data coordinating centre randomly selected intensive care encounters for review at each site. The audit consisted of source data verification and blinded chart abstraction, comparing findings by the auditors with those entered in the database. We also assessed completeness and timeliness of case submission. Quantitative evaluation of completeness, accuracy, and timeliness of case submission is reported. Results: We audited 434 encounters and 29, 476 data fields. The aggregate overall accuracy was 99.1%, and the major discrepancy rate was 0.62%. Across hospitals, the overall accuracy ranged from 96.3 to 99.5%, and the major discrepancy rate ranged from 0.3 to 0.9%; seven of the eight hospitals submitted >90% of cases within 1 month of hospital discharge. There was no evidence for selective case omission. Conclusions: Based on a rigorous audit process, data submitted to the PC 4 clinical registry appear complete, accurate, and timely. The collaborative will maintain ongoing efforts to verify the integrity of the data to promote science that advances quality improvement efforts.