Big Data and Total Hip Arthroplasty: How Do Large Databases Compare?

Abstract

Background Use of large databases for orthopedic research has become extremely popular in recent years. Each database varies in the methods used to capture data and the population it represents. The purpose of this study was to evaluate how these databases differed in reported demographics, comorbidities, and postoperative complications for primary total hip arthroplasty (THA) patients. Methods Primary THA patients were identified within National Surgical Quality Improvement Programs (NSQIP), Nationwide Inpatient Sample (NIS), Medicare Standard Analytic Files (MED), and Humana administrative claims database (HAC). NSQIP definitions for comorbidities and complications were matched to corresponding International Classification of Diseases, 9th Revision/Current Procedural Terminology codes to query the other databases. Demographics, comorbidities, and postoperative complications were compared. Results The number of patients from each database was 22,644 in HAC, 371,715 in MED, 188,779 in NIS, and 27,818 in NSQIP. Age and gender distribution were clinically similar. Overall, there was variation in prevalence of comorbidities and rates of postoperative complications between databases. As an example, NSQIP had more than twice the obesity than NIS. HAC and MED had more than 2 times the diabetics than NSQIP. Rates of deep infection and stroke 30 days after THA had more than 2-fold difference between all databases. Conclusion Among databases commonly used in orthopedic research, there is considerable variation in complication rates following THA depending upon the database used for analysis. It is important to consider these differences when critically evaluating database research. Additionally, with the advent of bundled payments, these differences must be considered in risk adjustment models.