For Medicare ACOs
Run the acuity ladder like a P&L
Post-acute care is where ACO money quietly leaks: the wrong facility, the extra week of stay, the readmission nobody saw coming. Each is a decision somebody made, or nobody made, at a transition. Clarity puts every transition in a workflow with evidence attached, an owner named, and a clock running.
Advisory signal, team-owned callsA trail your actuary can replayStarts in shadow mode
- For
- ACO executives, CMOs, population health leaders
- Status
- In build with design partners · pipeline in production
- Starts with
- A 90-day shadow-mode baseline
Institute of Medicine
of the geographic variation in Medicare spending per beneficiary comes from post-acute care.
Institute of Medicine
share of Medicare spending that post-acute care represents.
CMS · performance year 2024
record net Medicare savings in the Shared Savings Program. Three in four ACOs earned shared savings.
Cited figures. The economics further down this page are illustrative and labeled as such.
The problem
The most variable dollars in the episode
Post-acute care is about 20% of Medicare spending and 73% of the variation in it. The leak is not the price of care, it is the choices: which setting, how long, and whether anyone was watching the handoff. Three ways it costs you.
Your visibility ends at the hospital door
The SNF, IRF, and home-health legs live in faxes and disconnected systems, so you learn what actually happened about 90 days later, from claims. Too late to act, too coarse to learn from.
Every extra rung-day erodes the benchmark
Each rung down the acuity ladder is roughly 50–70% savings on illustrative Medicare per diems. A patient sitting one rung too high, or stepped down too soon and readmitted, costs you either way.
Savings you can’t attribute, you can’t keep
Analytics flags the long-stay outlier, but nobody can show who acted, in what tool, or prove it six months later. Savings claims die in attribution disputes.
Illustrative per diems
The ladder is a price list
Every step down the ladder saves money only when it is a documented decision instead of a habit. The costs here are illustrative directional anchors, not your contracted rates.
Hospital
~$2,800 · illustrative daily cost
Is the patient ready for the next rung, and which one?
Inpatient rehab (IRF)
~$1,900 · illustrative daily cost
Is intensive daily rehab still what the record supports, or would skilled nursing do?
Long-term acute care (LTACH)
~$1,800 · illustrative daily cost
Is hospital-level care still needed, and for how many more days?
Skilled nursing
~$650 · illustrative daily cost
Which facility, and for how many days?
Home health
~$70 / day-equivalent · illustrative daily cost
Ready three days early? That is real money, if someone acts.
Home
$0 · illustrative daily cost
Did the follow-up visit happen, or does the ladder restart?
| Rung | Illustrative daily cost | The decision at each step down |
|---|---|---|
| Hospital | ~$2,800 | Is the patient ready for the next rung, and which one? |
| Inpatient rehab (IRF) | ~$1,900 | Is intensive daily rehab still what the record supports, or would skilled nursing do? |
| Long-term acute care (LTACH) | ~$1,800 | Is hospital-level care still needed, and for how many more days? |
| Skilled nursing | ~$650 | Which facility, and for how many days? |
| Home health | ~$70 / day-equivalent | Ready three days early? That is real money, if someone acts. |
| Home | $0 | Did the follow-up visit happen, or does the ladder restart? |
Illustrative directional anchors, not your contracted rates. Published hospital costs average about $3,130 per adjusted inpatient day (KFF, 2023); we model $2,800 to stay conservative.
The workflow
Every transition becomes a decision
An advisory acuity signal indicates the move. Clarity recommends where and how, with the evidence attached, and your team owns every call. The ACO offering is in build with design partners, on the cited-AI pipeline Clarity runs in production. Every engagement starts in shadow mode.
How Clarity works
See it, decide it, own it
Three moves, in the order your team already makes them, each with evidence attached and an owner named.
- 01
See the transition
Clarity puts the transition in front of your team while it is still changeable, with the full clinical picture assembled from the referral packet and nationwide record connectivity, not a 90-day-old claim.
- 02
Decide with evidence
Right setting, right facility, right length of stay, each recommendation cited to the exact page of the source record. Your clinicians see the why, click the citation, and make the call.
- 03
Own the handoff
Every transition gets a named owner and a clock. Handoffs close the loop instead of vanishing into a fax queue, and every decision lands in an audit trail your actuaries can stand behind.
Right setting
Moves indicated by evidence, decided by your team
An advisory acuity signal from your care team’s monitoring indicates when a patient is ready to move. Clarity assembles the evidence behind it, quote and page, as new records arrive. When a trajectory crosses a readiness threshold, the step-up or step-down lands in Clarity as a team-owned workflow item. It carries a ranked facility recommendation built on bed capacity, payer-contract fit, and network scorecards.
- Five gates cleared before any move: clinical, financial, regulatory, operational, choice
- The receiving SNF sees a standard referral: same queue, same gates, one accept button
- The acuity signal is advisory. Clarity recommends; your team decides
Advisory step-up signal · Clinical gate
“Weight up six pounds in three days; new orthopnea reported overnight.”
Right length of stay
The next rung is planned before admission
Clarity plans the next rung before the current placement is accepted. Length of stay at each level becomes a decision your team makes, with the evidence attached, not a number claims reports back 90 days later.
Readmissions
A step-down that bounces back counts as a failure
Readmissions are a first-class counter-metric, not a footnote. Bounce-back risk is flagged before the door, beneficiary choice is one of the five gates every move must clear, and a step-down that readmits scores as a failure in Clarity’s own scorecard math. The platform has no incentive to push patients out.
Intervene in place
The default is to intervene in place
When acuity rises, the reflexive outcome is an ED visit and a multi-day admission. Step-downs save days. Step-ups save episodes. Clarity’s default is the lowest setting that can deliver the needed care, including a direct SNF admit from home under the MSSP 3-day rule waiver, with zero inpatient midnights.
The waiver adds that direct-to-SNF rung only when your ACO is in a two-sided track, the beneficiary is prospectively assigned, and the receiving SNF holds a written affiliate agreement with you and a rating of three stars or better. If you qualify, Clarity operationalizes it. If not, roughly $2.2M of the $2.4M illustration below stands without it.
Attributed member · 81 · CHF
Synthetic demo patientWeight up six pounds in three days; new orthopnea.
Advisory signal from the care team’s monitoring, not a Clarity score.
Why it’s different
Provable by design
Savings claims die in two rooms: the actuary’s office and the CMS audit. Clarity is built for both. Every recommendation carries a citation to the source record, every decision carries an owner and a timestamp, and every outcome is logged. When someone asks how you know the savings are real, you hand them the trail. Here is what that looks like on two ordinary days.
The early step down · synthetic patient
Leon, 79 · SNF day 14 of a planned 20
Therapy notes show he hit his mobility goals three days early, cited to the PT evaluation. Clarity flags the transition, the care manager confirms with the facility, and Leon goes home with home health on day 17. Three SNF days saved, ~$2,000, and a patient who got home sooner. Multiply that by a network and a year.
3 SNF days × ~$650 per day · illustrative
The catch · synthetic patient
Yolanda, 84 · discharged to a SNF nobody chose
The hospital broadcast her referral and an out-of-network facility answered first. Clarity is designed to surface the placement while it is hours old instead of 90 days old, your team reviews the record, and the family agrees to a transfer to a network facility with better outcomes data. The episode stays inside the network you built, where you can actually manage it.
Hours old, not 90 days old · illustrative scenario
Cited AI, and where it stops
Clarity reads packets and records and returns findings with the page attached. When the evidence is not there, it says insufficient evidence instead of guessing. The AI never moves a patient and never denies anything. Clarity recommends; your clinicians decide. That is not a disclaimer, it is the architecture.
Illustrative economics
The 20,000-life model
An illustrative reference model for a 20,000-life ACO, built on four conservative levers. A 90-day shadow-mode pilot replaces every assumption with your own risk-adjusted baseline before anyone intervenes.
Illustrative annual value
Four levers, arithmetic shown below. The shadow-mode pilot swaps in your baseline.
Per member per year
Across 20,000 attributed lives and about 1,520 post-acute episodes a year.
Stands without the waiver
The model does not depend on 3-day waiver eligibility.
Illustrative reference model, not audited customer results. Outcomes are targets, not guarantees.
Four levers, shown with their arithmetic
SNF length-of-stay right-sizing
Patients step down the day the evidence says they are ready, like Leon, instead of the day the habit says so.
−1.5 days × $650 × 1,520 episodes$1.48M
Avoided readmissions
Transitions with owners and evidence bounce back less. Each one avoided skips a whole readmission episode.
36 × $12,000$0.43M
Site-of-care optimization
Some patients skip a rung safely when the record supports it, including direct-to-SNF where you hold the waiver.
48 × $4,000$0.19M
Skilled-day leakage
Placements you can see in hours, like Yolanda’s, stay in the network you can actually manage. Wrong-rung days come out.
720 days × $450$0.32M
Total illustrative annual value
~$2.4M/yr ≈ $120 PMPY
| Lever | Arithmetic | Annual |
|---|---|---|
SNF length-of-stay right-sizing Patients step down the day the evidence says they are ready, like Leon, instead of the day the habit says so. | −1.5 days × $650 × 1,520 episodes | $1.48M |
Avoided readmissions Transitions with owners and evidence bounce back less. Each one avoided skips a whole readmission episode. | 36 × $12,000 | $0.43M |
Site-of-care optimization Some patients skip a rung safely when the record supports it, including direct-to-SNF where you hold the waiver. | 48 × $4,000 | $0.19M |
Skilled-day leakage Placements you can see in hours, like Yolanda’s, stay in the network you can actually manage. Wrong-rung days come out. | 720 days × $450 | $0.32M |
| Total illustrative annual value | ~$2.4M/yr ≈ $120 PMPY | |
Assumes 20,000 attributed lives and about 1,520 post-acute episodes a year. At 50–75% shared-savings retention, that is $1.2–1.8M a year to the ACO. Illustrative reference model: targets, not guarantees, and not audited customer results. A 90-day shadow-mode pilot replaces every line with your own risk-adjusted baseline.
Common questions
Asked first, answered first
Every step-down clears five gates, including beneficiary choice, before the door. If a step-down readmits, it scores as a failure in the same scorecard that credits the savings, so the incentive runs toward the right setting, not the cheapest one.
Start in shadow mode
A 20-minute walkthrough of the transition workflow and the 90-day baseline methodology your actuary agrees to before go-live. See your own baseline, not our slide.