The Standard Set

What's in a name? That which we call a rose
By any other name would smell as sweet...

I doubt skilled nursing was on Shakespeare's mind when he wrote those words in the late sixteenth century, but I'll take the liberty of assuming it was, if only so I can disagree. Juliet was off base. Names matter, particularly when they’re attached to a regulatory framework built on inconsistent federal and state provider designations. Names shape how performance is defined and analyzed. "Romeo and SNF" is a very different story than "Romeo and CCRC."

I've written extensively about variation across the SNF provider class and the public policy problems caused by treating all facilities as equivalent. There are two issues to unpack here.

First, identical facilities in adjacent markets can perform differently because the markets themselves differ, and that variation grows as we cross state lines. There is an insidious flip side to that coin, one I've spent considerably less time surfacing: the SNF provider class includes facilities that differ fundamentally in profile and operating model, making them noncomparable even within the same market.

Those differences reach beyond semantics. They affect how we measure access, cost, quality, and nearly every other metric drawn from a federal provider class that combines fundamentally discrete care settings.

The federal SNF designation covers facilities with different purposes, populations, physical plants, payers, and operating profiles. A hospital-based transitional care unit, a small SNF wing on a senior living campus, a county-run facility, a state-licensed specialty care provider, and a conventional freestanding nursing facility may all appear in the same dataset. Treating them as interchangeable can materially distort measures CMS has positioned as a pillar of (reported) excellence.

When we hear "nursing home," society does not picture a transitional care unit, a sprawling senior living campus where SNF care accounts for less than 10 percent of utilization, or a mixed-use facility where SNF beds share a physical plant with patients receiving other levels of care. We think of, well, a nursing home - the type of place people rushed to in the spring of 2020.

“Nursing Home” is the only term that trended in public searches of the SNF provider class.

It’s more than confusing - it’s inscrutable. The differences are too big to ignore, yet impossible to reconcile. The question is a simple one: What is a Skilled Nursing Facility?

Enter the Standard Set

Meaningful comparisons require comparable providers. A single federal designation encompasses facilities with fundamentally different profiles and operating models; in some cases, the differences are extreme. Treating every SNF as equivalent distorts policy and payment to the point of arbitrariness. Roughly one-third of certified SNFs drive most of that distortion. Performance must be analyzed against a standardized cohort that represents the conventional freestanding SNF market.

The Standard Set is Zimmet Healthcare's classification for conventional freestanding SNFs – a comparable cohort across states and markets that anchors SNFonomics and our State-of-States ratings. All else equal, included providers should respond to market and policy changes in reasonably similar and predictable ways. Excluded providers may not; their fixed costs, payer mix, resident acuity, and business models are fundamentally different.

Standard Set facilities must meet the following criteria:

·   Operate as freestanding providers with SNF-certified beds only, with limited allowances for Nursing Facility beds in certain states.

·   Participate in both the Medicare and Medicaid programs.

·   Remain active for the full reporting year.

·   Operate at least 40 certified beds.

·   Report occupancy of at least 40 percent for the applicable period.

The Standard Set excludes:

·   Hospital-based SNFs.

·   Continuing Care Retirement Communities.

·   Government-operated SNFs, except qualifying upper payment limit program participants supported by allowable intergovernmental transfers.

·   Medicare-only facilities that do not participate in Medicaid.

·   Facilities that combine SNF beds with other inpatient or residential care under a single financial reporting structure.

·   Specified outlier profiles that are economically noncomparable.

To put economic differences in perspective, 467 of the 500 SNFs reporting the highest 2024 net patient service revenue per day were excluded from the Standard Set. Most were licensed to provide more than skilled nursing. CCRCs, hospital-based units, and other mixed-use providers often consolidate financial reporting under a single business entity. The differences are obvious and substantial, yet CMS and many states continue to regulate these providers as one big, happy family.

The Standard Set does not eliminate complexity. It removes avoidable distortion so that differences among states and markets more accurately reflect structural economics rather than provider composition. Each state profile discloses the number and percentage of active SNFs represented by its Standard Set. A facility with a nonrecurring outlier in one measure remains in the cohort, but that specific value may be excluded or capped at a defined limit.

How Averages Change the Answer

The Standard Set is the starting point for meaningful comparison, but findings are too often expressed in ways counterproductive to advancing consistent regulation and equitable reimbursement. Specifically, aggregate provider performance is often reduced to an average. Attaching the word "national" or "state" to a measure is the most effective way of turning data into a useless number. Occupancy, for example, is a critical indicator of provider health, yet the answer changes with the cohort, the averaging method, and the question being asked.

There are two ways to average a statistic: simple and weighted. A simple average describes the experience of the typical provider. A weighted average describes the experience of the typical bed or patient day, giving larger facilities more influence and often shifting the result toward urban markets with substantially greater geographic bed density. Neither is inherently better. Each method can answer a valid question, but not the same question. Weighting does not solve industry problems; it surfaces geographic problems. A precisely weighted average of unlike providers is still an average of unlike providers. Similarly, failing to capture representative performance leads to uneven policy because policymaking occurs at the state level, leaving markets running with or against the regulatory wind. In other words, big-city performance too often overshadows statewide realities.

An Oklahoma Example

Putting cohort definition and averaging method together shows how much the answer can move. For example, most sources show Oklahoma's 2024 occupancy at roughly 61 percent. That number treats either the entire provider class as one giant facility or the state as a homogeneous market. Neither is true, and policy based on that figure can misfire. CMS HCRIS cost-report data show what happens when the results are separated by cohort and averaging method:

To say this only scratches the surface only scratches the surface of what I’m saying. Beyond the stark variation in average occupancy, the table leaves 41 SNFs unaccounted for. Where are they? Some are hospital-based and file within an IPPS cost report; some are simply not there. They may be released tomorrow, which would require any analysis based on today’s report availability to be updated. Either way, the missing reports are outside the Standard Set. The effect of adding them is unpredictable, which is exactly why they were excluded in the first place. In other words, adding providers with fundamentally different financial profiles would introduce additional distortion the Standard Set is meant to remove. In other-other words, it doesn’t matter.

I could go on and on, but I’ll leave you with one final nugget before calling it a day. Nationally, the simple average overall rating of excluded providers is 3.3 stars, while the Standard Set average is 2.8 stars. What's in a name? Apparently, the sun, the moon, and half a star.

Reconciling the Standard Set

In 2024, the methodology excluded 5,300 of 16,067 certified providers, leaving 10,767 in the Standard Set. The excluded facilities remain an important component of the healthcare delivery system, but they should be studied for what they are rather than used to define the economics of a conventional nursing facility.

The exclusions overlap, so the gross count for a category is often much larger than the number of providers newly removed at that step. The unique subtraction column prevents the same provider from being counted twice.

Equitable regulation and sustainable reimbursement begin with comparable providers. CMS and state regulators must distinguish ordinary statistical noise from structural differences when modeling how changes will affect conventional nursing facilities.

This is especially important when Medicaid serves as the safety-net payer for both residents and providers. A payment change can leave Standard Set providers worried about payroll while excluded providers that maintain SNF beds as a matter of convenience barely notice because many don’t participate in Medicaid.

That is why the name matters. Welcome to the Standard Set, where Skilled Nursing data makes a little more sense than usual. More to come.

Marc Zimmet is CEO of Zimmet Healthcare Services Group.