Monitoring Quality of Care for Joint Replacements: Assessing Alternative Statistical Methods to Accurately Estimate Time to Revision

Date
2015-02-04
Journal Title
Journal ISSN
Volume Title
Publisher
Abstract
With increased longevity and as younger, more physically active patients are receiving joint replacements, it has become increasingly important to measure the failure rate for joint replacements using the cumulative incidence of revision. The most commonly applied survival analysis technique, the Kaplan-Meier (KM) method, does not account for competing risks (i.e., patient deaths) and consequently overestimates the cumulative incidence. This thesis examines alternative methods for estimating the cumulative incidence of revision using population-based cohorts of hip and knee replacements performed in Alberta and Sweden. In comparing the KM method to the cumulative incidence function, which accounts for competing risks, the KM method overestimated the cumulative incidence at each time point. The magnitude of overestimation increased with follow-up time and higher mortality rates. Application of three regression models demonstrated competing risks models require careful interpretation. Competing risks methods are recommended to accurately estimate revision rates for healthcare planning purposes.
Description
Keywords
Health Care Management
Citation
Lacny, S. L. (2015). Monitoring Quality of Care for Joint Replacements: Assessing Alternative Statistical Methods to Accurately Estimate Time to Revision (Master's thesis, University of Calgary, Calgary, Canada). Retrieved from https://prism.ucalgary.ca. doi:10.11575/PRISM/27000