Institutional arthroplasty registry: what is the minimum acceptable dataset to be included in your hospital? Recommendations from a single-country national consensus using the Delphi method.

PURPOSE: Institutional arthroplasty registries are very popular nowadays; however, very few efforts have been made in order to standardize the information to be collected, thus limiting the possibility of inter-institutional data interpretation. This manuscript reports the results of a single-country consensus designed to define the minimum standardized dataset to be recorded within an institutional arthroplasty registry.
METHODS: A national consensus was carried out among all members of the Colombian Society of Hip and Knee Surgeons using the Delphi method. Eleven questions and answers comprising every potential domain of an institutional registry of hip and knee arthroplasty were defined. According to the methodology, anonymous voting and multiple discussion rounds were performed. Three levels of agreement were defined: Strong consensus: equal to or greater than 80%, weak consensus between 70 and 79.9%, and no consensus below 70%.
RESULTS: All of the questions reached consensus level. The minimum dataset was defined to include demographic and clinical information, intraoperative and implant details, follow-up and early complications, implant survival, and functional outcome scores, as well as the validation model to assess information quality within the database. Currently, this dataset is being implemented voluntarily by the members of our national society. DISCUSSION: A national consensus is a feasible method to build homogeneous arthroplasty registries. We recommend such an exercise since it establishes the basis to compare and add data between institutions and the joint analysis of said information in a national registry.