TEAM started the 30-day clock for about 700 hospitals. Here is the baseline a hospital needs for each TEAM surgery.

On January 1, 2026, CMS started the clock on the Transforming Episode Accountability Model, or TEAM. It is mandatory for hospitals paid under the inpatient prospective payment system in the core-based statistical areas CMS selected, roughly 700 hospitals by the participant list on the CMS model page, and it runs five performance years through December 31, 2030. For five surgeries, the hospital now answers for what Medicare spends from the operation through 30 days after discharge. To know whether a given episode, meaning one patient's surgery and the 30 days after discharge, made or lost money, a hospital needs its own cost for that episode, because CMS's data shows only what Medicare paid. Tracking five numbers for each episode answers that.

What TEAM holds a hospital to

The five episode categories are lower extremity joint replacement, surgical hip and femur fracture treatment, spinal fusion, coronary artery bypass graft, and major bowel procedure. An episode starts with the hospital admission or outpatient procedure for the surgery, which CMS calls the anchor, and ends 30 days after discharge. It counts nearly everything Medicare Parts A and B pay for in that window: the hospital stay, physician services, readmissions, skilled nursing, inpatient rehabilitation, home health, therapy, labs, equipment, and Part B drugs, with limited exclusions such as unrelated oncology or transplant admissions. Only Original Medicare patients are included in TEAM; Medicare Advantage patients are left out.

Hospitals keep billing fee-for-service. Once a year CMS compares total episode spending in each category against a target price built from three years of regional claims (2022 through 2024 for the first year), risk adjusted, and discounted 1.5% for bypass and bowel episodes and 2% for the other three. Under the target, the hospital may receive a payment; over it, the hospital may owe one. Track 1 is upside only, meaning a hospital can earn a payment but never owes one, and it is open to every hospital in 2026 and to safety-net hospitals through 2028. From 2027, every hospital outside Track 1 takes on two-sided risk, meaning it can also owe CMS money, with repayment capped at 20% of its total target price in Track 3. A composite quality score built from hospital-wide readmissions, a patient safety composite, and a patient-reported outcome measure for hip and knee replacement moves the reconciliation amount, the yearly payment or repayment, by up to 10%, or up to 15% for a Track 2 repayment. Every patient must be referred to primary care on or before discharge.

The FY 2026 rule, published August 4, 2025, widened the waiver that lets Medicare cover skilled nursing care without a prior three-day hospital stay, and it adjusted how target prices are built. The model's basic structure did not change.

Why the CMS data feed is not a baseline

CMS says it will give participants data to track spending. That data is claims, it arrives after the fact, and it describes Medicare's spending rather than the hospital's cost. Working out a hospital's gain or loss on an episode takes two subtractions, and the CMS data covers only one of them. The first is the DRG payment minus the hospital's own cost to deliver the anchor stay. The second is the reconciliation payment or repayment, which depends on everything that happened in the 30 days after the patient left. A hospital that watches only the second will find out a year later that the first was the problem, or the reverse.

Five numbers to track for every episode

1. Anchor hospitalization cost. Length of stay, operating room minutes, implant cost, ICU hours, and the direct cost of the stay, per episode, set against the DRG payment. This is the part of the episode the hospital runs itself, so it is the number it can change most directly.
2. Post-acute mix. For each discharge: home with no services, home health, skilled nursing facility, or inpatient rehabilitation. CMS lists shorter post-acute stays among the model's goals, and the rule that created TEAM says most participants in two earlier CMS bundled-payment models saved money mainly by spending less on post-acute care.
3. 30-day readmissions. Every readmission in the window is episode spending and also feeds the quality score. For each one, record the cause, the day of the episode it happened, and how the patient came back in (emergency department, direct admission, or transfer). A rate alone tells nobody what to change.
4. Discharge-to-home rate. The share of patients who go home, by surgery, by surgeon, by day of the week. It is a simple measure of patient flow, and every patient who goes home is one who did not go to a nursing facility or inpatient rehabilitation.
5. Days from surgery to first primary-care visit. The referral is required. Whether the visit happened, and when, is the number that tells a care management director if the referral did anything.

A good quarter or a lucky one

Episode counts at a single hospital are small. CMS's own low-volume threshold is 31 episodes per category across the three-year baseline, and many hospitals will see a few dozen bowel or bypass episodes in a year. At those counts, a few unusual patients can move a quarterly average a long way. An illustrative example, not a client figure: a hospital with 30 joint replacement episodes in a quarter sends six to a skilled nursing facility instead of the usual three. Average episode spending jumps, the quarter looks bad, and nothing about the surgical program changed. The three extra nursing-facility stays happened because two patients lived alone and one had a fall. The next quarter goes back to normal, and someone gets credit for fixing it.

A baseline that separates the good quarter from the lucky one holds each episode as a row with all five numbers attached, so the question "what changed" has an answer at the patient level: which surgeon, which discharge day, which nursing facility, which readmission. Averages hide that. CMS caps outliers at the 99th percentile for pricing, but a hospital looking at its own episodes needs to see every one, including the most expensive.

What the Throughput Diagnostic produces

Kent Applied's Throughput Diagnostic is a fixed-price, three-week engagement at $14,500. For a hospital in TEAM, it uses the hospital's own data to measure a baseline for each of the five surgery categories: what a 30-day episode costs and how patients move through it. For every episode, that baseline includes the length of stay, where the patient went after discharge, any readmission, the days to the first primary-care visit and what Medicare spent, from the TEAM claims. It also shows the spread behind each average, the one step that most slows patients on the way from the operating room to discharge to their first follow-up (the binding constraint), and what each day of that delay costs. You get a written plan to relieve that step, with an estimate of how many more patients you could handle, plus the tables to re-run the analysis each quarter.

It is built from a limited data set under a Data Use Agreement, inside your environment if you prefer. If your problem looks different from this one, the first call is still free.

If your hospital is in the model, we will build a free report on where it stands in TEAM, using CMS's public files, and send it to you: request it here. It needs none of your data, and no sales emails follow.