Will Unreasonable CMS Expectations Worsen Knee Osteoarthritis Care?

In the study by Zalikha et al., only two-thirds of a contemporary sample of >64,000 patients from the American Joint Replacement Registry (AJRR) who underwent primary total knee arthroplasty (TKA) for osteoarthritis met the U.S. Centers for Medicare & Medicaid Services (CMS) threshold for substantial clinical benefit (SCB), defined as a 20-point improvement in their Knee Injury and Osteoarthritis Outcome Score for Joint Replacement (KOOS-JR) from preoperatively to 1 year postoperatively. Most surgeons should initially meet this result with resistance. Although patient dissatisfaction following TKA remains an issue, well over two-thirds of our patients would consider their improvement in quality of life after TKA to be “substantial.” Among patients with severe bilateral knee osteoarthritis, well over two-thirds undergo contralateral TKA1,2. The rate of TKA also continues to rise, in part, because of patient outcomes3. There are several theories that can be used to interpret the results of this study. First, it is possible that the results of this study are false. Maybe far greater than two-thirds of patients who undergo TKA achieve at least a 20-point improvement in their KOOS-JR at 1 year postoperatively. Although there was likely some selection bias introduced by using AJRR data and analyzing only patients with follow-up data, as the authors acknowledge, it is not clear that these factors would bias the results to underestimate the proportion of patients achieving the SCB. If anything, patients managed at centers reporting data to the AJRR and providing follow-up data may do better than average. We know that patients who are socioeconomically disadvantaged are generally less likely to report PROMs and have worse scores4,5. On the basis of this study and other work, it seems unlikely that all satisfied patients have a 20-point KOOS-JR improvement6. Second, it is possible that these results are true but that changes in the KOOS-JR do not measure patient satisfaction or reflect anything substantial. The KOOS-JR consists of 7 questions regarding the extent to which a patient’s knee has stiffness; pain with activities involving twisting, straightening, ascending or descending stairs, and standing; and function involving rising and bending. Although no 7 questions could capture all relevant aspects, the tool measures much of what surgeons consider important when discussing TKA with patients: decreased pain and improved function. Prior research has shown that the KOOS-JR is correlated with satisfaction, even if it is not strongly predictive7. Although the KOOS-JR may have ceiling effects and other limitations, it seems as reasonable a tool as any8. Third, and most likely, it is possible that these results are true but that the CMS threshold for defining the SCB using the KOOS-JR is false. For instance, Zalikha et al. found that over three-fourths of patients achieved a 14-point improvement in the KOOS-JR at 1 year and that five-sixths of patients had at least a 7.5-point improvement. The threshold of a 14-point improvement was selected as representing anchor-based estimates of the minimal clinically important difference, and the threshold of a 7.5-point improvement was selected as distribution-based estimates of the minimal clinically important difference. These thresholds or a value between them may better capture, for the CMS and the public, whether TKA was successful. Prior literature has found smaller changes of around 7 points to be meaningful9,10. The results of this study highlight the need for surgeons to work with the CMS to reconsider the threshold values used to define success based on the change in KOOS-JR, especially given the desire to publicly report these rates. Some have already highlighted the potential harms that could come to patients, especially vulnerable populations, from such reporting4,5. Surgeons and hospitals may be disincentivized from offering TKA to patients less likely to complete surveys or to have larger score improvement. For instance, this study found that Black patients were about one-half as likely to have an SCB following TKA. Although much work is needed to better understand this complex association, it is vital that such statistical realities should not be used to limit access to care. Unfortunately, the pathway from using this CMS definition to exacerbating inequalities may be indirect. Even if surgeons and hospitals caring for vulnerable populations are not directly influenced by concerns over public reporting, if these rates are later tied to reimbursement or other changes in volume, then these groups may have fewer resources to expand their services despite their best intentions. Additionally, a misunderstanding regarding the value of TKA may lead patients to pursue less-established treatments, exposing them to risks and further driving health-care costs11,12. The CMS needs to work with surgeons and patients to select a more appropriate KOOS-JR threshold value for reporting TKA success to the public.