Why Nursing Home Star Ratings Are Not The Whole Story

Nursing Home Quality · June 7, 2026

You found a five-star facility. You toured it on a Tuesday afternoon. The lobby smelled like something warm was baking. The administrator shook your hand and handed you a brochure with the CMS badge printed right on the cover. You felt relief.

That relief is exactly what the rating system is designed to produce. And that is the problem.

The CMS Five-Star Quality Rating System was introduced in 2008 because Congress noted it was easier for an American consumer to shop for a washing machine than to evaluate a nursing home. That framing tells you everything about what the system was built to do: reduce a complex clinical environment to something digestible enough that families stop asking harder questions. Federal audits, investigative reporting, and peer-reviewed research have spent the years since 2008 documenting exactly how well that worked, and how badly it failed the people it was supposed to protect.

This is not an argument against using the star rating. It is an argument for knowing what it is actually measuring, and for understanding the machinery behind it before you hand someone you love over to it. We cover the full breakdown of how to read CMS ratings on our CMS ratings guide, and you will find a fuller set of evaluation tools on our Silent Voices resources page.

Three numbers inside one star

The overall star rating is a composite of three separate domains. Health Inspections, which is the only domain based on actual third-party evaluations by state surveyors. Staffing, which historically relied on data the facility reported about itself. And Quality Measures, which is calculated almost entirely from clinical assessments the facility fills out about its own residents.

The Staffing and Quality Measures ratings can each add or subtract a star from the baseline Health Inspection score. Which means a facility can fall short on its independent inspections and still achieve a high overall rating by posting strong numbers in the two domains it controls. CMS built a mathematical cap into the formula to limit that, capping facilities with a one-star inspection at two stars overall regardless of their other scores. That cap is an admission. If the self-reported data were reliable, the cap would be unnecessary.

A grading curve no one tells families about

Here is something the brochure does not mention. Health Inspection ratings are not graded against a national standard of safe care. They are graded on a curve within each state. CMS locks the distribution so that exactly 10% of facilities in every state receive five stars in the Health Inspection domain, and exactly 20% receive one star, regardless of the actual quality of care in that state.

A facility that would receive two stars in a state with aggressive enforcement could receive four or five stars in a state with lax oversight, simply by outperforming its local competitors. The rating tells you how a facility compares to its neighbors. It says nothing about whether those neighbors are safe. In a state where most facilities are inadequate, the five-star facilities are just the least inadequate ones in the room.

The staffing numbers were invented

For years, nursing homes self-reported their staffing levels during a two-week window immediately before their scheduled annual inspections. Facilities staffed up for the window, posted the numbers, and reverted to skeleton crews once the surveyors left. This was not speculation. When the government finally transitioned to auditable payroll data, the gap became visible.

An analysis by the Center for Public Integrity of over 10,000 nursing homes found that more than 80% had reported higher registered nurse coverage on their public profiles than what their audited Medicare financial records showed. In more than a quarter of those facilities, the advertised RN staffing was at least double the daily reality. When CMS implemented the Payroll-Based Journal system, the data showed that in the last quarter of 2017, 25% of facilities reported having zero registered nurses on duty for at least one day, in direct violation of federal licensing requirements.

Analysis of nearly 500 million nurse shifts found median annual nursing staff turnover of 94%, with RN turnover reaching 140.7% and CNA turnover at 129.1%, according to research published in Health Affairs. High turnover disrupts continuity of care and makes subtle clinical changes easier to miss. When a facility replaces its entire registered nurse staff more than once a year, the person caring for your loved one today may have no institutional knowledge of who they were six months ago.

The diagnosis that changed the metric

The Quality Measures domain tracks 15 clinical outcomes, including antipsychotic medication use. CMS flags high antipsychotic use as a negative quality indicator because these drugs are frequently used as chemical restraints on dementia patients who are difficult to manage in understaffed environments. There is a carveout in the methodology: residents with a documented schizophrenia diagnosis are excluded from that metric entirely.

A March 2026 report by the HHS Office of Inspector General found that schizophrenia diagnoses were being added to residents’ records at rates inconsistent with clinical prevalence, removing those residents from the antipsychotic quality measure and allowing facilities to maintain high scores on that metric regardless of actual medication practices. A September 2025 OIG report found separately that nursing homes failed to report 43% of falls resulting in major injuries and hospitalizations among Medicare residents. Unreported falls do not count against the score.

The Quality Measures domain is built on data the facility generates about itself, with no routine independent verification before CMS uses it to calculate the rating. An empirical study published in INQUIRY estimated that between 6% and 8.5% of all nursing homes actively inflate their self-reported measures. The facilities most likely to do it are large, for-profit, chain-affiliated facilities with the most financial incentive to boost their public score. Because a five-star Quality Measures rating automatically adds a full star to the overall composite, the inflation lands directly in what families see on Care Compare.

The citations that disappear during appeals

When state surveyors cite a facility for a severe violation, that facility can appeal the citation. Certain appealed deficiencies may not be reflected in public ratings until the appeal process concludes. While an appeal is active, those citations are excluded from the star rating calculation. The appeals process can run for over a year. During that time, residents and their families have no access to those proceedings. Negotiations take place between the facility’s legal team and government attorneys, and the government frequently agrees to downgrade or withdraw citations to avoid prolonged litigation.

A New York Times investigation identified approximately 2,700 dangerous incidents, documented by state inspectors, that were completely hidden from the public due to active or negotiated appeals. According to CMS data, only 50 Administrative Law Judge decisions regarding nursing home penalties were issued in 2020, a fraction of the thousands of serious deficiency citations filed during that same period.

The rating you see on Care Compare is not the rating the facility earned. It is the rating that survived the paperwork.

What to look at instead

Use the star as a door. Then walk through it and do this.

The star rating is a starting point. It was never designed to be the last word. For a full walkthrough of how to read each component of the CMS rating, visit our CMS ratings guide or go directly to CMS Care Compare to pull a facility’s inspection history. The families who walk in knowing what the star does and does not measure are the ones who ask better questions, catch what the brochure hides, and are ready to act when something changes.

---