How to compare nursing homes using CMS data

How to compare nursing homes

Most nursing-home comparisons go wrong in the same two ways: lining up facilities from different states as though their stars mean the same thing, and comparing the single overall number instead of what sits underneath it. Both are avoidable once you know why they break. This guide gives the workflow, and the reasoning behind each step.

Rule one: compare within a state

CMS builds the overall star primarily from the health-inspection rating, and it assigns that rating on a curve within each state — roughly the top tenth of a state's facilities receive five stars, the bottom fifth receive one. The comparison group is the state, never the country.

So a four-star facility in one state and a four-star facility in another were measured against different peer groups, by different state survey agencies. If you are choosing between facilities in one metro area, this does not affect you. If you are weighing options across a state line — common near borders, or when a family is split between two states — the stars are not directly comparable and the component detail matters more.

Rule two: compare components, not the overall star

The overall rating compresses three different measurements into one number, and the compression hides what you actually want to know. Two facilities showing four stars can be built completely differently: one strong on inspections and thin on staffing, the other the reverse.

Put the health-inspection, staffing, and quality-measure ratings side by side for each facility and look for disagreement. That gap is the most informative thing available, because it tells you where a facility's strength actually is — and whether it is in the area that matters for your situation.

Build a shortlist of two to four

Use provider search to filter by state and county, then add facilities to Compare from any result row or facility page. Compare holds two to four at once and lines every CMS field up in the same row, so gaps are visible rather than buried in separate tabs.

Four is a real ceiling, not a technical one. Past that the table stops being readable, which defeats the purpose of aligning the fields. If you have more candidates, narrow on location and services first, then compare.

Read the records behind the ratings

Ratings summarize; records specify. For each shortlisted facility, look at the inspection citations and their severity letters, any federal penalty records, payment denials, and whether the CMS abuse icon is shown.

Read these for pattern rather than presence. A single resolved citation from three years ago and the same deficiency recurring across successive surveys produce similar counts and mean very different things. Our guides on reading citations and on federal penalty records cover how to tell those apart.

Add local and ownership context

Every facility page benchmarks the facility against its own county and state, which is the comparison that is actually valid. County and state pages show the full local rating distribution, so you can see whether a four-star facility is typical or unusual in its own market.

Ownership pages show other facilities under the same CMS-reported ownership. That grouping is context, not a verdict — a company appearing in these records does not by itself establish that it controlled day-to-day care.

Then verify, and visit

CMS data describes a dated snapshot. Confirm what matters on Medicare's official Care Compare and with your state survey agency, then call each facility on the shortlist.

Arrive at the visit with the comparison in hand. The data's job is to turn a vague decision into three or four specific questions — about the staffing pattern on the unit you would use, about a citation you read, about what changed since the last survey. That is what it is for.

Frequently asked questions

Can I compare nursing homes in different states?

Not cleanly on stars alone. CMS assigns the health-inspection rating relative to other facilities within the same state, so a four-star facility in one state was measured against a different peer group than a four-star facility in another. Within one state or metro area, comparison is straightforward; across state lines, rely on the component ratings and the underlying records.

What should I compare first?

The three component ratings — health inspection, staffing, and quality measures — rather than the overall star. Two facilities with the same overall rating can be built very differently, and the disagreement between components tells you where each one's strength actually is.

How many facilities should I compare at once?

Two to four. Compare holds four, and beyond that the table stops being readable, which defeats the purpose of lining the fields up. Narrow on location and services first if you have more candidates.

What does a missing value mean in a comparison?

It means CMS does not publish that value for that facility — it is shown as missing and never as zero. Treat it as unknown rather than as a poor result; a facility that is new, small, or recently changed ownership commonly has gaps.

Is a higher rating always better for me?

Not necessarily. Ratings summarize CMS measures for a past period and cannot describe the specific unit, staff, or care a resident would receive. Use them to shortlist and to generate questions, then verify with a visit and the official CMS record.