PP70 Identification Of Prostheses With Worse Than Expected Outcomes. (2019)

• Record Type: Journal Article
• Authors: Arias de la Torre, Jorge; Martínez, Olga; Smith, Kayla; Pons-Cabrafiga, Miquel; Prieto-Alhambra, Daniel; Valderas, Jose; Martin, Vicente; Evans, Jonathan; Marcos, Raul; Esteban, Marta; Villabrille, Susana; Espallargues, Mireia
• Is Part Of: International journal of technology assessment in health care. Volume 35(2019)Supplement 1
• Page Start: 51
• Page End: 51
• Publication Date: 2019
• Subjects: Medical technology -- Periodicals; Technology assessment -- Periodicals; 610.28
• DOI: 10.1017/S0266462319002198
• Languages: English
• ISSNs: 0266-4623

Abstract

Abstract : Introduction: Monitoring the effectiveness of knee and hip arthroplasties could be useful at the clinical, economic, and patient levels. In Catalonia, there is currently no systematic monitoring of the different prostheses available. The aims of this study were to propose an approach for the systematic identification of knee and hip prostheses with the highest revision rates, and to identify those with the poorest outcomes. Methods: Data recorded from January 2005 to December 2016 were considered from 53 out of the 61 public hospitals in Catalonia included in the Catalonian Arthroplasty Register (RACat). Specific prostheses were classified by joint, type, fixation, and, in total hip prostheses, the bearing surface. Prostheses with the worst outcomes were identified using a three-step approach, based on previous literature: (i) screening using Poisson models; (ii) comparison of prostheses using adjusted Cox models; and (iii) consensus-based review by a panel of orthopedic surgeons to detect possible sources of bias. After this process, selected prostheses were provisionally labeled as having the poorest outcomes. This process will be repeated periodically within the RACat to definitively classify the prostheses. Results: After first two steps, ten knee prostheses and eight hip prostheses were identified. After the panel discussion (third step), one knee and one hip prosthesis were excluded from the final list. The knee prosthesis was excluded because it was a unicompartmental implant, while the hip prosthesis was excluded because it was a monoblock implant. Finally, nine knee prostheses and seven hip prostheses were provisionally identified as having the worst results relative to other available prostheses. These results await confirmation in subsequent analyses. Conclusions: This study contributed to the current need to identify hip and knee prostheses whose outcomes might be worse than expected. This identification could have an impact at the patient, surgeon, industry, and stakeholder levels.