BPCI stands for Bundled Payments for Care Improvement, a Medicare program that pays hospitals and doctors a single target price for an entire episode of care rather than billing separately for every test, procedure, and follow-up visit. The idea is straightforward: if a hospital knows it will receive one fixed payment for, say, a hip replacement and all the recovery care that follows, it has a strong incentive to coordinate that care efficiently and avoid unnecessary costs. The program is run by the Centers for Medicare & Medicaid Services (CMS) and currently operates as BPCI Advanced, which is scheduled to run through December 31, 2025.
How Bundled Payments Work
Under traditional Medicare, every provider involved in a patient’s care sends a separate bill. The surgeon bills for the operation, the hospital bills for the room and supplies, the physical therapist bills for rehab, and so on. No single provider has a financial reason to think about the total cost or whether the patient truly needs each service. Bundled payments flip that structure by setting a target price for the full episode, from hospital admission through a defined recovery window.
Medicare still pays each provider individually as services are delivered. But after the episode ends, CMS adds up all those payments and compares the total to the target price. If the total spending came in below the target, the participating hospital or physician group keeps the difference. If spending exceeded the target, the participant owes money back to Medicare. This retrospective reconciliation is the core financial mechanism that drives the program.
The Original Four BPCI Models
CMS launched the original BPCI initiative with four distinct models, each testing a different scope of bundled payment:
- Model 1 covered only the inpatient hospital stay itself. Medicare paid the hospital a discounted amount, while physicians were still paid separately under the standard fee schedule.
- Model 2 cast a wider net. The episode included the hospital stay plus all post-acute care and related services for 30, 60, or 90 days after discharge. Actual spending was reconciled against the target price after the fact.
- Model 3 focused specifically on post-acute care. The episode was triggered by a hospital stay but began when the patient started recovery services at a skilled nursing facility, rehab facility, long-term care hospital, or with a home health agency. It also ran for 30, 60, or 90 days.
- Model 4 was a prospective model where the hospital received a single, predetermined payment for the entire episode upfront.
These original models have since ended. Model 2 attracted the most participants and generated the most data, which helped CMS design the successor program.
BPCI Advanced: The Current Program
BPCI Advanced launched on October 1, 2018, replacing the original models with a single, streamlined design. It uses the same retrospective reconciliation approach as the original Model 2, where Medicare pays providers as usual and then settles up against the target price afterward. But it added mandatory quality measurement and required participants to accept real financial risk, not just the possibility of earning bonuses.
As of 2024, 247 acute care hospitals and physician group practices participate in the program, split almost evenly between 123 hospitals and 124 physician practices. That’s down from 280 participants in 2023, according to MedPAC, the commission that advises Congress on Medicare policy.
Clinical Episodes Covered
BPCI Advanced covers a broad range of medical conditions and surgical procedures, organized into categories. Orthopedic procedures like hip and knee replacements are among the most common, but the program also includes cardiac care (heart failure, heart attacks, arrhythmias), cardiac procedures (bypass surgery, valve replacements, stents), gastrointestinal conditions (bowel obstruction, GI bleeding, inflammatory bowel disease), and medical conditions like sepsis, pneumonia, COPD, kidney failure, stroke, and urinary tract infections. Some episodes apply only to inpatient stays, while others cover both inpatient and outpatient settings.
How Target Prices Are Calculated
The target price is not a simple average. CMS builds it from several layers of adjustment to make it fair across different hospitals and patient populations. The starting point is historical spending data for similar episodes at the hospital. That baseline is then adjusted for patient severity, so a hospital treating sicker patients gets a higher target. CMS also accounts for persistent spending differences across peer groups of similar hospitals and applies a trend factor to project costs into the current period.
After the performance period ends, CMS applies one more retrospective adjustment to account for unexpected, system-wide cost changes that no individual hospital could have predicted. This re-centers the target around what peer hospitals actually spent nationally, preventing participants from being penalized or rewarded for broad economic shifts rather than their own efficiency.
Quality Measures and Payment Adjustments
Cost savings alone don’t determine whether a participant earns or loses money. CMS aggregates several quality measures into a Composite Quality Score that adjusts the final payment. Two measures apply to every clinical episode: all-cause hospital readmission rates and whether the patient had an advance care plan in place. Depending on the episode type, up to three additional measures may apply, drawn from a mix of claims data, hospital-reported data, and clinical registries.
This quality link matters for another reason. BPCI Advanced qualifies as an Advanced Alternative Payment Model under Medicare’s Quality Payment Program. That designation gives participating clinicians an exemption from the standard Merit-Based Incentive Payment System (MIPS) reporting requirements. To qualify, participants must accept financial risk of up to 20% of the target price for each episode, well above the 3% minimum threshold. They must also use certified electronic health record technology and have their payments tied to quality measures comparable to MIPS.
Does BPCI Actually Save Money?
The most recent CMS evaluation, covering Model Year 5, found that BPCI Advanced saved Medicare $344 million, roughly 4% of what payments would have been without the program. On a per-episode basis, that translated to $1,014 less in spending per episode. Readmission rates and mortality rates were unaffected, meaning the savings did not come at the cost of worse clinical outcomes.
The savings largely came from reducing discharges to post-acute care facilities like skilled nursing homes and inpatient rehab centers. Instead, more patients recovered at home with home health services or outpatient therapy. Functional recovery was similar between BPCI Advanced patients and a comparison group. However, some BPCI Advanced patients rated their care experience and satisfaction slightly lower than patients outside the program. CMS flagged that this could reflect how providers communicate about discharge plans, particularly whether they set clear expectations or consider patient preferences when deciding where recovery will happen.
What BPCI Means for Patients
If you’re a Medicare beneficiary, you won’t receive a bill that says “bundled payment” or be asked to opt in. The program operates behind the scenes. You still see the same doctors, go to the same hospitals, and receive the same covered services. The difference is that your care team has a financial incentive to coordinate your treatment more tightly, avoid duplicative tests, and plan your discharge and recovery carefully.
In practice, this often means more attention to discharge planning. A care coordinator may follow up with you after you leave the hospital, check that you have your medications, and ensure your follow-up appointments are scheduled. You may be more likely to recover at home with visiting nurses or outpatient rehab rather than being sent to a skilled nursing facility. For many patients this is preferable, though it works best when the care team explains the plan and involves you in the decision.