The Centers for Medicare & Medicaid Services (CMS) finalized the Comprehensive Care for Joint Replacement Expanded Model (CJR-X) in the Inpatient Prospective Payment System (IPPS) Final Rule for the 2027 fiscal year (FY) on July 31. Starting Jan. 1, 2028, most acute-care hospitals paid under both IPPS and the Outpatient Prospective Payment System (OPPS) will be mandatorily accountable for a 90-day episode of care built around every hip, knee, and ankle replacement they perform, whether the anchor procedure happens in the inpatient or outpatient setting. More than 2,500 hospitals are expected to be included in the mandatory model.
The rollout of CJR-X is occurring while CMS is eliminating the Medicare Inpatient-Only List, providing a glimpse of what may be in store for hospitals in the years to come. Total knee and total hip arthroplasty came off the IPO List in 2018 and 2020, respectively, and inpatient volume for those procedures has fallen. The Chief Actuary’s Aug. 21, 2026 certification memo confirms that as of 2024, a total of 86 percent of knee procedures were performed in an outpatient setting or in an Ambulatory Surgery Center (ASC), and 61 percent of hip procedures were performed outside the inpatient setting.
In other words, joint replacement is no longer confined to a single site of service. A Medicare beneficiary can have a hip or knee replaced as a hospital inpatient, as a hospital outpatient, or in a freestanding ASC. Three sites, three payment mechanisms, and, as of 2028, one 90-day accountability window that follows the patient through whichever door they walked in.
This is where clinical documentation integrity (CDI) departments should start paying attention, because CJR-X does not treat those three sites the same way. Only the inpatient and hospital-outpatient anchor events trigger a CJR-X episode. For inpatient anchors specifically, CJR-X reconciles against a target price built from MS-DRG 469, 470, 521, and 522, the same DRGs that have always captured joint replacement, but the population landing in those DRGs is not the population it used to be.
Once a healthy 68-year-old with osteoarthritis and no meaningful comorbidity burden can have a knee replaced in an ASC or as same-day hospital outpatient surgery, that patient has very little clinical reason to occupy an inpatient bed. What remains for inpatient status are the patients who are not medically cleared to go home the same day. This may include patients with a Charlson Comorbidity Index of 2 or higher, uncontrolled diabetes, significant cardiopulmonary disease, morbid obesity, or a history of complications with anesthesia. The procedure got easier to perform outpatient, so those who need an inpatient bed are by definition the ones who were always going to be harder to keep complication-free.
That concentration of risk is not a side effect of CJR-X. It is the exact variable the model is trying to price correctly. The original CJR Model risk-adjusted its target prices using three factors. CJR-X uses 29 factors, including a count of specific Hierarchical Condition Category (HCC) flags at the beneficiary level. CMS built an expanded risk-adjustment methodology, in its own words, so that hospitals are not penalized for beneficiaries who are complex or for whom care is unpredictably complicated. That is a meaningful improvement over the original model’s three adjusters, and it is also functionally an admission that the remaining inpatient joint replacement population is going to look sicker – not because care has gotten worse, but because the easy cases have already migrated to the outpatient setting.
That timing detail is not lost on physician advisors, either. Ronald Hirsch, MD, pointed out on RACmonitor last week that CJR-X’s target price for its first performance years is built from a rolling baseline running from 2023 through 2027. Whether or not a given hospital is thinking about it in those terms, the discharge-planning and post-acute utilization decisions being made on today’s inpatient joint replacement patients are quietly setting the bar that CDI and utilization review will be held to three years from now, which means the documentation trail those decisions leave behind starts mattering before performance year one ever begins.
Here is the problem for hospitals, and the opportunity for CDI: those 29 risk adjusters only work if the comorbidities and HCC categories driving them are captured in the claims data. A hospital can have a genuinely excellent surgeon and a complex, high-risk inpatient joint replacement population, and still get penalized at reconciliation if the documentation does not support the acuity that justified the inpatient admission in the first place. Under CJR-X, undercoding a comorbidity is not just a lost HCC on a risk-adjustment model somewhere upstream; it directly understates the target price a hospital is measured against for that specific 90-day episode, and it weakens the medical necessity record for the admission decision itself, at a moment when payers are already scrutinizing inpatient joint replacement status more aggressively than almost any other elective procedure.
CDI’s job under CJR-X, then, is not fundamentally different from CDI’s job anywhere else, but the stakes attached to getting it right are higher and more concentrated. Query for the comorbidities that justified inpatient status, rather than letting “TKA, elective” stand alone, as if the DRG assignment settles the question. Make sure an unstable comorbidity shows up in the record in terms that map to a reportable code, not just in a preoperative risk-assessment note that never gets translated into the discharge summary. Capture complications with the same rigor CDI has always applied to present-on-admission status, since CJR-X’s 90-day window makes every readmission and every post-acute utilization decision part of the same financial episode as the surgery. None of this is new CDI methodology. It is the same clinical validation and specificity work CDI has always done, redirected at a population that is about to become smaller, sicker, and far more visible to hospital finance.
A shrinking, high-acuity inpatient joint replacement population is not a problem CDI created, and it is not one CDI can solve by working harder at the same tasks. It is solved by making sure that when a patient is sick enough to need an inpatient bed for a hip or knee replacement in 2028, the record proves it just as convincingly as the surgeon’s judgment already did. This is yet another way the relationship between utilization review, which assists with patient status, and CDI need to collaborate to support revenue integrity.
