BPCI stands for Bundled Payments for Care Improvement, a program created by the Centers for Medicare & Medicaid Services (CMS) to test whether paying for a whole episode of care — instead of each service separately — can lower costs without hurting quality. Under bundled payments, a group of providers receives one combined payment for a defined episode, such as a hip replacement plus the recovery period that follows. If the team delivers that care for less than the target price, it can share in the savings. If costs run higher, it may owe money back.
The model matters to ordinary people because it changes how hospitals, surgeons, and nursing homes are paid. The financial incentive shifts from doing more procedures to coordinating care and preventing complications. Whether that shift has reliably improved care is a fair question, and the evidence is genuinely mixed. Here is what the program is, how the money works, and what it has and has not achieved.
What Is BPCI And How Do Bundled Payments Work?
BPCI is a CMS payment model that links the cost of a medical episode to its quality. The original program, BPCI Model 1 through Model 4, launched in 2013. A redesigned version called BPCI Advanced began in 2018 and continues to operate. The core idea is the same across versions: define an episode, set a target price, and hold providers financially accountable for the total cost of that episode.
Traditional Medicare pays each provider separately. The hospital bills for the surgery. The surgeon bills for the operation. The skilled nursing facility bills for rehab. The home health agency bills for visits. Each gets paid for what it does, with little direct financial link between them.
Bundled payment changes that. CMS sets a target price for the entire episode, based on historical spending for similar patients. All the usual claims still get paid as they normally would. Then, after the episode closes, CMS compares actual spending to the target. If the total came in below target and quality standards were met, the participating provider earns a share of the difference. If spending exceeded the target, the provider may have to repay part of the excess.
That reconciliation step is the part people often miss. Patients do not receive a single bill or a lump-sum payment. Their Medicare coverage works the same way. The bundling happens behind the scenes, between CMS and the provider organizations that volunteered for the program.
What Counts As An Episode Of Care?
An episode is a defined window of time and a defined set of services connected to one clinical event. The window usually starts at the hospital admission or the procedure and runs for a set number of days afterward. For surgical episodes, the recovery and rehab period is included. For some medical conditions, the episode covers the initial hospitalization and follow-up care.
BPCI Advanced covers dozens of clinical episodes. Common examples include joint replacement, spinal fusion, coronary artery bypass, and certain medical admissions such as congestive heart failure or pneumonia. The exact list has changed between program versions, so the specific episodes included depend on which model a provider joined.
The length of the recovery window varies by episode type. A joint replacement episode typically extends well past discharge because rehab and nursing home care are often the largest cost drivers. This is deliberate. The designers wanted providers to think about what happens after the hospital, not just the procedure itself.
Who Participates In Bundled Payment Programs?
Participation is voluntary, and the organizations that join tend to be hospitals, physician groups, and conveners that coordinate multiple providers. A convener is an entity that takes on the financial risk on behalf of a group of smaller practices that might not be able to manage it alone.
Hospitals with strong care coordination infrastructure are more likely to participate. So are orthopedic and cardiac programs where the episode is well defined and the recovery path is relatively predictable. Smaller or rural hospitals have participated less often, partly because they may lack the data systems and care partners needed to manage an episode.
The program is not limited to one type of provider. A skilled nursing facility, a home health agency, or a physician group can be a participant if it takes on accountability for the episode. What matters is who accepts the financial risk and who can influence the total cost and quality of care.
How Does The Target Price Get Set?
CMS sets the target price using historical Medicare claims data for similar patients. The target reflects what the episode has typically cost, adjusted for the patient’s clinical characteristics and other risk factors. The goal is a fair benchmark — not so generous that savings are easy, and not so tight that providers are set up to fail.
Risk adjustment is one of the hardest parts of the design. A patient with multiple chronic conditions costs more to treat than a healthy patient having the same operation. If the target price does not account for that, providers could be penalized for taking on sicker patients, or they could avoid those patients altogether. CMS adjusts for risk, but whether the adjustment is adequate is a subject of ongoing debate among health policy researchers.
Target prices are also rebased over time. As providers become more efficient, the historical spending that sets the benchmark falls, which can make future savings harder to achieve. This is a known tension in the model.
What Are The Financial Incentives And Risks?
The incentive structure rewards efficiency and coordination. A hospital that reduces readmissions, shortens unnecessary skilled nursing stays, and prevents complications can lower episode costs and earn savings. That is the intended mechanism.
The risks are real too. Financial pressure can create incentives that are not always aligned with patient interests. If providers are rewarded for spending less, there is a concern that some may skimp on care, discharge patients too early, or steer patients away from needed services. CMS addresses this through quality requirements: providers must meet certain quality measures to receive savings, and poor performance can reduce or eliminate the payment.
There is also the question of who bears the downside risk. In some models, providers take on “upside-only” risk, meaning they can earn savings but do not have to repay losses. In others, they take on “two-sided” risk, meaning they share in both savings and losses. Two-sided risk creates stronger incentives for efficiency but also stronger pressure, and it can be harder for smaller organizations to absorb.
Does Bundled Payment Actually Work?
The evidence is mixed, and honest researchers say so. Some evaluations of BPCI and similar bundled payment models have found modest reductions in episode spending, often driven by shifts to lower-cost post-acute settings such as home health instead of skilled nursing facilities. Other evaluations have found little or no savings, and some have raised concerns about whether quality was maintained.
A consistent finding across several evaluations is that savings, when they appear, tend to come from post-acute care rather than the hospital stay itself. That makes sense given the cost structure. Skilled nursing facility days are expensive, and there is often room to reduce unnecessary days without harming recovery. Whether that reduction is always appropriate for every patient is a separate question that the data cannot fully answer.
What the evidence does not clearly show is a broad, consistent improvement in patient outcomes across all conditions. Some studies suggest quality was maintained or slightly improved in certain episodes. Others are less conclusive. Anyone who claims bundled payments have definitively transformed care quality is overstating what the research supports.
What Does This Mean For Patients?
For most people, the direct experience of care under a bundled payment program looks the same as usual. You still see your doctors, you still have Medicare coverage, and you still receive bills for your care as you normally would. The bundling is a payment arrangement between CMS and providers, not a change to your benefits.
What can change is the coordination around your care. A hospital participating in a bundled program may be more proactive about discharge planning, more likely to arrange home health, or more attentive to preventing readmissions. These can be genuine improvements. They can also mean you are offered a post-acute setting that is less intensive than what you might otherwise have received, which is appropriate in some cases and not in others.
If you are scheduled for a procedure that falls under a bundled episode, it is reasonable to ask your care team how they coordinate post-discharge care and what support you will have at home. That question is useful regardless of the payment model.
How Is Bundled Payment Different From Other Medicare Models?
Bundled payment sits between traditional fee-for-service and full population-based models like accountable care organizations. Fee-for-service pays for each service with no accountability for the total. Accountable care organizations take responsibility for the total cost and quality of care for an assigned population over a year. Bundled payment focuses on a defined episode for a defined group of patients over a shorter window.
That middle position is both its strength and its limitation. It is easier to implement than full population risk because the episode is bounded. But it also means providers are accountable only for the episode, not for what happens before or long after. A patient’s overall health depends on factors outside any single episode.
The table below summarizes the broad differences.
| Model | Unit Of Accountability | Financial Risk |
|---|---|---|
| Fee-for-service | Each service | None for total cost |
| Bundled payment (BPCI) | Defined episode of care | Shared savings and possible losses |
| Accountable care organization | Assigned population over a year | Shared savings and possible losses |
What Are The Main Criticisms Of BPCI?
The most common criticism is that savings have been modest and uneven. Some participants have done well while others have exited the program, and the overall impact on Medicare spending has been smaller than early hopes suggested. That is not unusual for payment reform, which tends to produce incremental change rather than dramatic results.
Another criticism concerns patient selection. If providers can influence which patients enter an episode, there is a risk they favor healthier patients who are cheaper to treat. CMS monitors for this, but detecting subtle selection effects is difficult.
A third concern is administrative burden. Managing an episode requires data systems, care coordinators, and financial tracking that smaller practices may not have. That can concentrate participation among larger, better-resourced organizations and leave smaller ones behind.
None of these criticisms means the model is a failure. They are the normal tensions of trying to change how health care is paid for. The honest summary is that bundled payment has produced some efficiency gains, particularly in post-acute care, while the evidence on quality and broad savings remains mixed.
Frequently Asked Questions
What does BPCI stand for in Medicare?
BPCI stands for Bundled Payments for Care Improvement, a CMS program that pays a combined target price for a defined episode of care rather than paying each service separately. A newer version called BPCI Advanced began in 2018.
Do bundled payments change what patients pay?
No. Patients keep their normal Medicare coverage and receive bills as usual. The bundling is a financial arrangement between CMS and participating providers, not a change to patient benefits.
What is an episode of care in bundled payment?
An episode is a defined window of time and set of services tied to one clinical event, such as a joint replacement plus the recovery period that follows. The window length varies by procedure or condition.
Has bundled payment been proven to lower costs?
Evidence is mixed. Some evaluations found modest savings, often from shifts in post-acute care, while others found little or no reduction. Broad, consistent savings across all episodes have not been clearly demonstrated.

