Post-Acute Care: The Unclaimed Prize

Medicare’s fee-for-service program paid skilled nursing facilities about $25 billion in 2023, and per-capita spending on skilled nursing and home health has been roughly flat since 2011, according to MedPAC. More than a decade of payment reform has bent that curve very little.

This is not for lack of a working method. Accountable care organizations have been shown to cut post-acute spending by about 9% without harming patients. Post-acute care remains the prize that value-based care keeps identifying and keeps failing to claim at scale. The reason is that the method depends on data that mostly does not move.

The savings are real

An analysis of the Medicare Shared Savings Program, covering more than 25 million patient-years, found that ACO participation was associated with a reduction in post-acute spending of $106 per beneficiary, about 9.0% of the pre-contract average. The savings came from two levers: fewer discharges to facilities rather than home, and shorter stays for the patients who did go.

Just as important is what did not happen. The reductions were not associated with significant changes in 30-day readmissions, mortality, or the use of highly rated facilities. That is a rare result. It means a meaningful share of post-acute utilization was discretionary rather than necessary, and removing it did not harm patients.

The savings come from managing individual patients

How the savings happened matters more than that they happened. The reductions were not produced by hospitals steering patients to a preferred network of facilities, nor by hospital-wide discharge redesign. They were most consistent with clinicians working inside hospitals and skilled nursing facilities to influence the care of ACO patients specifically, one patient at a time.

A separate national study of nearly 11.7 million beneficiaries confirms this from the opposite side. When a hospital joined an ACO, post-acute use did not change for the hospital’s patients in general. The ACO was not changing how the hospital worked. It was changing what happened to its own attributed patients.

That is a demanding mechanism. It requires knowing which patients are yours, following them into a facility you do not own, and influencing decisions made by clinicians you do not employ. It is care management conducted at arm’s length.

The data stops at the facility door

That mechanism depends on seeing into the skilled nursing facility, and that is the weakest point in the entire information chain. As prior AIVBC research on hospital data exchange documented, hospitals send records to post-acute providers at twice the rate they receive records back: 17% versus 8%.

Put those together. The savings depend on knowing what is happening to your attributed patient inside the facility, in time to act. Only about one hospital in twelve routinely gets information back from that facility. An organization accountable for the total cost of care can see the admission it paid for and is nearly blind to the two-week stay that follows.

What this implies

Three things follow, and none is a call for a new payment model.

• The data connection to post-acute providers is a value-based care priority, not an IT backlog item. Taking on total cost of care without visibility into skilled nursing and home health means accepting risk on a population you cannot observe.
• Investment should follow the mechanism that worked. The evidence points to embedded clinical relationships, people who follow attributed patients into facilities, not to preferred-network strategies or discharge-pathway redesign, which the data does not credit for the savings.
• Expect per-patient returns, not institutional ones. Because the effects do not spill over to other patients, an organization cannot buy post-acute savings once and apply them across a hospital. The work scales with the attributed panel and must be staffed that way.

What the evidence does not settle

The ACO effect estimates come from observational difference-in-differences studies of the Shared Savings Program’s early years (roughly 2010 through 2014), which predate both the current generation of risk models and the growth of Medicare Advantage. The spending figures here are current; the effect sizes are not, and a 9% reduction then is not a guarantee of 9% now.

What has not changed is the shape of the problem. Post-acute spending remains large and flat, and the one mechanism shown to reduce it safely depends on information crossing a boundary that, on the most recent national data, it still mostly does not cross.

Related

• Hospital at Home Is Linked to Fewer Emergency Department Visits
• Interoperability Reached 70% of Hospitals by 2023, but the Data Often Goes Unused

Sources

• McWilliams JM, Gilstrap LG, Stevenson DG, et al. Changes in Postacute Care in the Medicare Shared Savings Program. JAMA Internal Medicine. 2017;177(4):518-526. doi:10.1001/jamainternmed.2016.9115 (opens in a new tab)
• Navathe AS, Bain AM, Werner RM. Do Changes in Post-acute Care Use at Hospitals Participating in an Accountable Care Organization Spillover to All Medicare Beneficiaries? Journal of General Internal Medicine. 2018;33(6):831-838. doi:10.1007/s11606-018-4368-z (opens in a new tab)
• Health Care Spending and the Medicare Program: A Data Book. Section 8, Post-Acute Care. Medicare Payment Advisory Commission; July 2025. medpac.gov (opens in a new tab)
• Gabriel MH, Richwine C, Strawley C, et al. Interoperable Exchange of Patient Health Information Among U.S. Hospitals: 2023. ONC Data Brief No. 71. Office of the National Coordinator for Health Information Technology; May 2024. healthit.gov (opens in a new tab)