Skilled nursing is a unique provider class. While SNFs are commonly associated with hospitals and home care agencies, financial realities differ widely across the healthcare continuum. My first cross-industry study examined the striking similarities between skilled nursing facility and cruise ship economics. That realization changed my approach to analytics and marked the start of SNFonomic theory. This one isn’t quite so profound. Take it for what it’s worth, which isn’t nothing.

A recent drive from one suburban outpost to another less than an hour away took me past three sets of Golden Arches, the fourth most recognized logo on Earth. A lively debate began about whether the country had more McDonald’s restaurants or SNFs. The debate became frustratingly heated, which would not have been so odd if I hadn’t been alone in my car. As it turns out, myself and I drew to a statistical tie: the two are just about even.

Before anyone starts using this for underwriting the next great SNF acquisition strategy, relax. The purpose of the McSNF Index is not to induce a Big Mac Attack. It is simply a reminder that sometimes the market sees things before the industry does (or in the case of skilled nursing, long before they can do anything about it).

The McSNF Index started as analytical satire; then the numbers showed up and ruined the joke. National comparisons involving nursing facilities require some housekeeping in the source data to mitigate distortion. This one required very little. As always, I removed Alaska, Hawaii, and U.S. territories. The only other SNFs removed from the data were those with fewer than 30 certified beds, most of which are located on the far side of rural and are often government sponsored. Sensitivity testing using nearby bed thresholds produced nearly identical results, suggesting the filter was not driving the outcome.

Ray Kroc opened his first franchised McDonald’s roughly a decade before Medicare and Medicaid transformed nursing homes into the modern SNF industry. Today, both are mature industries with remarkably stable geographic footprints.

Across 48 states, there are 14,070 active SNFs and 13,653 McDonald’s locations. Using SNFs as the denominator, the national McSNF Index equals 0.97, almost one for one.

McDonald’s does not study Medicaid rate construction, Medicare Advantage penetration, hospital discharge patterns, staffing levels, or cost reports. But the fast-food giant knows market analytics. It knows roads, traffic, labor pools, commercial nodes, land use, and demographic shifts. SNFs depend on many of the same conditions. The difference is that McDonald’s moves at the speed of business—they know what is happening and can pull normal economic levers in response to market changes. Providers cannot, and that is what causes distortion and drives imbalanced reimbursement rates, occupancy, and staffing costs.

Remarkable Parallel

The correlation between McDonald’s locations and active SNFs with 30 or more beds is 0.938. Remove the four highest and four lowest ratio outliers, and it rises to 0.951. That is not a punchline, and it is no parlor trick. It is a remarkable parallel where none should exist.

Occupancy was the one variable that could have turned the comparison into a statistical illusion, so we tested it. After applying state occupancy rates in multiple configurations, the correlation remained around 0.938.

Occupancy as a standalone measure was almost uncorrelated with the McSNF ratio. In other words, the McSNF Index held. McDonald’s tells us whether the SNF map resembles the commercial map. Occupancy tells us whether the mismatch matters.

No, McDonald’s does not predict nursing home demand. Nursing home residents are hardly McDonald’s target demographic. The McSNF Index is not a bed-need formula. It is modest but telling: it shows where the SNF map no longer looks like the commercial map. In an industry with distorted national data, this oddity is worth noting.

What the Question Is

The McSNF Index may have no practical utility for operators, but it’s real. Dig deeper into the statistics and we begin to see how the measure may identify state-level changes that could have informed policy as the need evolved, rather than years after the fact.

If a state has far more SNFs than McDonald’s, there is a reason. Perhaps legacy institutional density is the explanation, or rural county-seat facilities, old CON geography, Medicaid long-stay dependence, or a fragmented provider map that never consolidated because healthcare specializes in preserving yesterday’s problem with tomorrow’s regulation.

If a state has far more McDonald’s than SNFs, maybe commercial growth moved faster than post-acute infrastructure. Maybe the market shifted. Maybe assisted living, home care, MA behavior, labor, zoning, or facility economics explain it.

The index does not answer the question. It tells you where the question is, and that’s why it works. It may be ridiculous and nutritionally questionable, but the McSNF Index is a surprisingly effective way to ask whether CMS is still reading the same map as everyone else.

Either way, those fries are still the best.

The Big Stats

There are 13,653 McDonald’s locations and 14,070 active SNFs with at least 30 beds in the 48-state dataset. That means there are 0.97 McDonald’s locations for every SNF. In practical terms, the two industries exist in almost identical numbers nationally. The ratio does not tell us whether a state is over- or under-bedded. It simply tells us how closely the SNF footprint resembles the commercial footprint represented by McDonald’s.

R² measures how much of the variation in one variable is explained by the other. Our comparison yielded an R² of 0.881, meaning approximately 88 percent of the differences in McDonald’s counts among states can be explained simply by knowing the number of SNFs in those states. The remaining 12 percent of variation is likely influenced by factors such as tourism, population growth, rurality, certificate-of-need policy, assisted living growth, and legacy institutional density.

The Regression Slope was 0.994, indicating that for every additional SNF in a state, there is almost exactly one additional McDonald’s location. If the slope were 0.6 or 1.8, the exercise would still be interesting. A slope of 0.994, in context with the residual variation, is measurable and meaningful.

Pearson correlation measures how closely two variables move together. A value of 0.00 means no relationship; a value of 1.00 means a perfect relationship. The McSNF Index produced an exceptionally strong correlation of 0.938. States with more SNFs almost always have more McDonald’s locations, and states with fewer SNFs almost always have fewer McDonald’s locations. More telling, when the top and bottom four outliers are removed, the correlation increases to 0.951.

McDonald’s does not predict nursing home demand. The McSNF Index is useful because two completely unrelated and mature industries grew to draw almost the same map of America. The ratio shows how close they are, the correlation demonstrates the relationship is real, R² shows it is not random, and the remaining variance highlights where the most interesting market questions exist.

I’m lovin’ it.

Marc Zimmet is the CEO of Zimmet Healthcare Services Group.

Comments or questions? Contact Patrick Connole at pconnole@parkplacelive.com.