The Journal of Arthroplasty, Volume 34, Issue 8, 1831 - 1836

The Validity of All-Cause 30-Day Readmission Rate as a Hospital Performance Metric After Primary Total Hip and Knee Arthroplasty: A Systematic Review

Ali, Adam M. et al.
Hip Knee


Risk-adjusted all-cause 30-day readmission rate (ACRR) following total hip arthroplasty (THA) and total knee arthroplasty (TKA) is currently used as a metric of hospital performance as part of the Hospital Readmissions Reduction Program. However, the extent to which it is determined by hospital-related factors and is therefore a fair method of determining reimbursement remains unclear.


Our aim was to systematically review the available literature pertaining to whether ACRR is a valid metric of hospital performance after elective primary THA or TKA as determined by (1) its association with other performance metrics, (2) the extent to which variation in ACRR can be explained by between-hospital variation, and (3) the relative importance of hospital-related versus surgeon- or patient-related factors in determining ACRR. The MEDLINE, EMBASE, and Health Management Information Consortium databases were searched from inception to November 2018 and reference lists of selected articles scanned. The final list of articles was determined by consensus.


Eight articles were included. Correlation of ACRR with established composite metrics of both outcome and process measures was poor. There was a weak positive correlation between ACRR and mortality. Only 1.5% of the variation in readmission rates for THA and TKA was found to be attributable to hospital-level factors, with patient-related factors such as age and comorbidities having much greater influence. Use of composite outcome metrics, for example, combining readmission and mortality, or considering the “surgical” readmission rate, improved the sensitivity to detect important between-hospital variation.


There is insufficient evidence in the current literature to justify the use of ACRR following elective THA and TKA for financially penalizing hospitals. Further work is needed to define what is acceptable variation. The use of a composite metric or surgical readmission rate may improve the ability to detect between-hospital variation.

Download article