Abstract TP396: Does Implementation of Enhanced Data Quality Assurance Procedures Improve Data Quality in Stroke Registries? Experience from the University of Texas Houston Stroke Registry

Abstract

Background: Limited information has been published regarding standard procedures to ensure data quality in stroke registries. We share our experience regarding the establishment of enhanced quality assurance (QA) procedures for the University of Texas Houston Stroke Registry (UTHSR) and test whether these QA procedures have improved data quality in UTHSR. Methods: All 5092 patients’ records that were abstracted and entered in UTHSR, between January 1, 2008 and December 31, 2011, were considered in this study. A random subset of 30 records was used for re-abstraction of 24 key variables by two abstractors. From these 30 records, a subset of 13 was re-abstracted by a team of experts as the “gold standard”. We assessed inter-rater reliability (IRR) between the two abstractors as well as each abstractor with the gold standard. Depending on the scale of variables, IRR was assessed with Kappa or by intra-class correlations (ICC) using a 2-way, random effects ANOVA. For assessment of data quality we re-abstracted another set of 41 patients’ records and all discrepant entries were adjudicated by a team of experts. We used Chi-square test for trend to assess whether a significant improvement in data quality has been achieved during 2008-2011. Results: The error rate dropped from 3.8 in 2008 to 1.6% in 2011, with a significant (P < 0.001) downward trend in the error rate over the 4 years. The two abstractors had an excellent IRR (Kappa or ICC > 0.75) on almost all 24 variables checked. The only variable that had moderate reliability (0.40 ≤ Kappa ≤ 0.75) was mode of arrival to hospital. Agreement between each abstractor and the gold standard for both categorical and continues variables were also excellent, except for mode of arrival to hospital. Conclusions: Establishment of a rigorous data quality assurance for our UTHSR has helped to improve the quality of data. We have observed an excellent IRR for almost all key variables in UTHSR. We recommend training of chart abstractors and systematic assessment of IRR between the abstracted data with that of the gold standard for data quality assurance processes.