Nursing homes manipulate their scores by self-reporting inflated staffing numbers and 'miscoding' residents with fake mental health diagnoses to hide the use of sedatives. While federal CMS and state inspection data is the most reliable part of the score, it only accounts for one-third of the total rating, leaving the rest open to creative accounting. To find the truth, you have to look past the overall star rating and dig into specific health inspection cycles and the Palmelle Clarity Score.

For years, nursing homes simply told the government how many people they employed, and the government took their word for it. Today, the system uses payroll data, which is harder to fake but far from foolproof. Facilities know exactly when their 'survey window' is—the period when state inspectors are likely to show up—and they staff up aggressively for those few weeks to ensure the numbers look perfect. Once the inspectors leave, the staffing levels often plummet, leaving one or two aides to care for thirty residents overnight.

There is also the phenomenon known as the 'weekend dip.' A facility might have a five-star staffing rating because they are over-staffed on Tuesdays when the administrators are in the building. However, if you look at the raw federal CMS and state inspection data, you’ll often find that staffing levels on Saturdays and Sundays drop by as much as 40%. If your parent has a fall on a Sunday afternoon, a five-star rating won't help if there isn't anyone in the hallway to hear them call for help.

To see through this, ignore the 'Staffing' stars and look for the 'Adjusted RN Staffing' hours per resident day. If that number is below 0.5 hours, it means a registered nurse is spending less than 30 minutes a day on each resident. That is the reality of the care, regardless of how many stars are on the plaque in the lobby.

One of the most cynical ways facilities game the system involves the 'Quality Measures' score. The government penalizes nursing homes that use high levels of antipsychotic drugs, which are often used as 'chemical restraints' to keep residents quiet and easy to manage. However, there is a massive loophole: if a resident has a diagnosis of schizophrenia, Huntington’s disease, or Tourette’s, their medication use is excluded from the facility’s public score.

Since 2012, the number of nursing home residents with a diagnosis of schizophrenia has skyrocketed, despite the fact that schizophrenia is a condition that almost always manifests in early adulthood, not at age 85. Facilities are essentially asking doctors to 'find' a schizophrenia diagnosis for residents with dementia so the facility can sedate them without hurting their five-star rating. A 2021 New York Times investigation found that one in nine residents had been given this diagnosis, a statistical impossibility that exists solely to protect the facility's reputation.

When you are looking at a facility, look specifically at the 'Long-Stay Quality Measures' for antipsychotic medication. If the number is low but the facility has a high rate of 'exclusionary diagnoses,' you are looking at a facility that is likely over-medicating its residents to keep labor costs down. This is why the Palmelle Clarity Score weighs these anomalies heavily—because a clean record shouldn't be bought with a bottle of Seroquel.

The 'Health Inspection' portion of the rating is the only part based on actual, third-party observations from state officials. But even this is flawed because of the time lag. State inspections generally happen every 12 to 15 months. In the world of care facilities, a year is an eternity; a facility can change ownership, fire its best director, and lose half its staff in three months. A five-star rating based on an inspection from 14 months ago is effectively a historical document, not a current report card.

Furthermore, the federal government recently added a 'Red Hand' icon to the Care Compare website to flag facilities with a history of abuse or neglect. Surprisingly, some of these facilities still maintain a three or four-star overall rating because their self-reported 'Quality Measures' are so high that they pull the average up. The system allows a facility to be literally dangerous while still appearing 'average' or 'above average' to a casual observer.

This is why we focus on the raw federal CMS and state inspection data. We look for 'Scope and Severity' codes. A letter 'G' or higher on an inspection report means 'Actual Harm' was done to a resident. If you see a 'G' in the last two years, the number of stars on the door is irrelevant. You are looking at a place that has already failed the people it was supposed to protect.