The Manitoba Joint Replacement Registry: validation of a provincial hip and knee arthroplasty registry

BACKGROUND AND PURPOSE: Regional registries can offer increased dataset linkages and analysis opportunities compared with national registries. However, registry datasets should be properly assessed before their use. The purpose of our study was to validate the completeness and accuracy of the Manitoba Joint Replacement Registry (MJRR) by comparing it with comprehensive databases. METHODS: We identified hip and knee arthroplasty patients in the MJRR ≥ 18 years old who had surgery between 2008 and 2023. We compared our data cohort with the hospital Discharge Abstracts Database (DAD) and the Medical Services Database (MSD) through percentage agreement (proportion identical) and Cohen's kappa to analyze accuracy and completeness. RESULTS: We identified 64,472 potential patients included in at least 1 of these 3 datasets. MJRR coverage increased from 68.2% in 2008 to 97.1% in 2023. The percentage agreement between the MJRR, DAD, and MSD was high. No 2 variables matched lower than 90.6%, and all Cohen's kappa scores were at least within the substantial agreement range (0.61-0.80) with most in the near perfect agreement range (0.81-1.00). Large urban hospitals achieved almost 100% MJRR coverage for all diagnoses, while rural hospitals captured nearly all elective procedures but almost no hip arthroplasty for fractures. CONCLUSION: We showed that our regional joint replacement registry contains high completeness of data. We believe that the MJRR is a valuable resource, and its data can be used and trusted for quality initiatives and research projects.