After the first year in the Comprehensive Care for Joint Replacement (CJR) model, hospitals must repay Medicare for spending above a target price. Hospitals are incentivized to reduce spending in a 90-day episode and generate internal cost savings through, for example, the use of lower-cost implants.
We used a Markov model to compare quality-adjusted life-years and lifetime costs of total hip arthroplasty, under Medicare fee-for-service (baseline) and under alternative revision rate assumptions (prospective CJR scenarios). Results were generated for 65-year-old and 75-year-old male and female Medicare beneficiaries using baseline spending and revision rates from Medicare claims. We estimated the impact of CJR on 90-day spending. We ran sensitivity analyses for revision rates.
Under willingness-to-pay thresholds of $50,000, $100,000, and $150,000, the baseline scenario was more cost-effective than the CJR scenario for a 65-year-old male patient if the revision risk increases by at least 7% (95% confidence interval for CJR savings: 4%-22%), 5% (range, 3%-7%), or 3% (range, 1%-5%), respectively. For males aged 75 years and females, revision risk needs to increase by a greater percentage under CJR relative to baseline for Medicare fee-for-service to be more cost-effective.
The CJR model holds great promise. However, it incentivizes hospitals to choose lower-cost implants and adopt newer technology more slowly, which could potentially increase revision rates and offset benefits of the program. Policy makers should monitor revision rates and consider changes to the CJR model to ensure beneficiary access to valuable technology.