How Physician Hip and Knee Replacement Volumes Relate to Readmission Rates

Readmission penalties for hip and knee replacements are forecasted to account for more than 50 percent of the total penalties Virginia hospitals incur as part of the Centers for Medicare & Medicaid Services Hospital Readmissions Reduction Program (RRP) in 2016. That is a key reason why identifying options for reducing readmissions following total joint replacement (TJR) has value for VHHA members. Research Corner analyses in recent weeks have examined whether factors such as discharge destination and hospital stay duration correlate to readmission rates. Now, we turn to analysis of surgical volume as a variable. Specifically, is there a relationship between 30-day readmissions and physician surgical volume? VHHA built a database matching hip and knee admissions to physicians and hospitals. Logistic regression was used to plot physician volume against the risk-adjusted readmission rate (RRSR) of a given physician’s hospital (using RRSR to control for patient acuity). Physicians with fewer than 10 admissions in a three-year period were excluded from the analysis. The final sample group consisted of 263 physicians who performed knee replacements, and 175 who performed hip replacements during the period spanning 2012-2015. The resulting analysis indicates a statistically significant association between RSRR and the number of hip and knee replacement surgeries performed by a given physician (P=.006 and P= .004, respectively). The median three-year volume is 108 surgeries for knee replacements, and 75 for hip replacements. These findings are similar to a 2006 study that examined the relationship between knee replacement volume and clinical complications in California. That study found that lower hospital volumes are associated with higher complication rates after total knee arthroplasty.


According to Virginia data, the number of TJR surgeries done by a physician has a stronger influence on readmissions, and is more predictive than simply counting the number of readmissions by given surgeons. Based on this analysis, physicians who perform 36 or more knee replacements annually, and those who perform 25 or more hip replacements annually, are likely to have a lower percentage of readmissions. Undertaking a deeper analysis of surgeons with lower procedure volumes could prove beneficial to understanding how such risk factors might contribute to readmissions. The findings of this analysis, as well as those in the aforementioned 2006 study, may help Virginia hospitals and surgeons chart a manageable action plan for preventing TJR readmissions, thereby potentially enhancing performance in the RRP. (3/11)