Understanding CMS Star Ratings: How Nursing Home Quality Is Measured

The overall star rating is not a simple average of the three domains. CMS uses a specific methodology:

1. Start with the health inspection rating. This is the foundation. A facility's overall rating begins equal to its health inspection star rating.
2. Add staffing. If the staffing rating is 4 or 5 stars (and above a minimum threshold), the overall rating gets +1 star. If the staffing rating is 1 star, the overall rating gets -1 star.
3. Add quality measures. If the quality measures rating is 5 stars, the overall rating gets +1 star. If the quality measures rating is 1 star, the overall rating gets -1 star.
4. Cap at 5, floor at 1. The overall rating cannot exceed 5 or drop below 1.

This means the health inspection rating has the strongest influence on the overall rating. A facility with 1-star inspections cannot achieve more than 3 stars overall, even with perfect staffing and quality measures. Conversely, a facility with 5-star inspections can drop to 3 stars overall with poor staffing or quality measures.

This weighting is intentional: CMS considers health inspections the most direct measure of actual care quality, since they involve on-site observation by trained surveyors rather than self-reported data.

Health inspection ratings are based on the results of the three most recent standard (comprehensive) surveys, plus any complaint investigations during that period. More recent surveys are weighted more heavily than older ones.

For each survey, every cited deficiency receives a score based on its scope (how many residents were affected) and severity (how much harm occurred or could have occurred). The scoring grid ranges from:

• Scope: Isolated (1-2 residents) → Pattern (several residents) → Widespread (systemic)
• Severity: No actual harm with potential for minimal harm → No actual harm with potential for more than minimal harm → Actual harm → Immediate jeopardy

Each deficiency's scope-severity combination maps to a point value. The facility's total points across all deficiencies from the weighted three-year history determine its health inspection rating. CMS then distributes ratings on a bell curve within each state: roughly 10% of facilities get 1 star, 20% get 2 stars, 35% get 3 stars, 25% get 4 stars, and 10% get 5 stars.

Critical limitation: Because the distribution is done within each state, a 3-star facility in a state with generally good care (like Minnesota) may actually be providing better care than a 4-star facility in a state with lower overall standards. Cross-state comparisons using star ratings alone can be misleading.

Since 2018, CMS has used Payroll-Based Journal (PBJ) data instead of self-reported staffing numbers. Facilities must submit actual payroll records documenting the hours worked by every nurse and aide. This was a significant improvement, as previous self-reported data was found to be inflated by up to 12% on average.

The staffing rating considers two primary metrics:

• Total nursing staff hours per resident per day (HPRD) — includes RNs, LPNs/LVNs, and CNAs
• RN hours per resident per day — registered nurses specifically

Both metrics are case-mix adjusted, meaning they account for the acuity (medical complexity) of the facility's resident population. A facility caring for predominantly high-acuity residents is expected to have more staffing hours than one with lower-acuity residents.

The approximate thresholds for each star level (case-mix adjusted):

• 1 star: Total HPRD below about 3.3 or RN HPRD below about 0.4
• 3 stars: Total HPRD around 3.6-3.9, RN HPRD around 0.55-0.65
• 5 stars: Total HPRD above about 4.1, RN HPRD above about 0.75

Weekend staffing is now also reported separately, allowing families to see whether the facility maintains adequate staffing seven days a week. Many facilities staff significantly lower on weekends — a gap that directly affects care quality for those days.

Limitation: PBJ data captures hours worked but not skill mix within categories. An hour of care from an experienced 20-year CNA is counted the same as an hour from a newly certified aide.

The quality measures domain is based on 15 clinical quality indicators derived from the Minimum Data Set (MDS), a standardized assessment that nursing homes must complete for every resident at admission, quarterly, and whenever there is a significant change in condition. The data is self-reported by the facility.

The 15 measures include:

• Long-stay measures (11): percentage of residents with pressure ulcers, UTIs, falls with major injury, use of physical restraints, excessive weight loss, depressive symptoms, need for increased help with activities of daily living, catheter use, antipsychotic medication use, and successful discharge to community
• Short-stay measures (4): percentage of short-stay residents with new or worsening pressure ulcers, rehospitalization within 30 days, emergency department visit, and functional improvement

Each measure is risk-adjusted to account for the facility's case mix. The facility's performance across all 15 measures is compared to the national distribution to determine its quality measures star rating.

Important caveat: Because MDS data is self-reported by the facility, there is an inherent conflict of interest. Studies have found that some facilities underreport negative outcomes like falls and pressure ulcers. CMS audits MDS accuracy but cannot verify every data point. This is why health inspections (where independent surveyors verify conditions on-site) are weighted more heavily in the overall rating.

The five-star system has been criticized on several fronts, and understanding these limitations helps you use the ratings more intelligently:

1. Gaming the system. Some facilities have learned to optimize their star ratings without genuinely improving care. This can include coaching staff before surveys, temporarily increasing staffing during expected survey windows, and strategically coding MDS assessments. A 2019 GAO report found evidence of these practices.
2. State-by-state variation. Health inspection ratings are distributed within each state, meaning the same level of care might receive different ratings depending on the state. States also vary in how rigorously their survey agencies conduct inspections. Some states cite 50% more deficiencies per survey than others, which may reflect different standards rather than different care quality.
3. Lag time. Ratings are updated monthly, but they rely on data that may be 3-12 months old. A facility that has recently changed ownership, hired new leadership, or experienced a significant event may have current quality that differs substantially from its rating.
4. What is not measured. Star ratings do not capture many things that matter to residents and families: food quality, social engagement, cultural sensitivity, staff friendliness, noise levels, responsiveness to call lights, or overall "feel" of the facility. These subjective quality factors can only be assessed through visits and conversations with current residents and families.
5. Special focus facilities. CMS designates some consistently poor-performing facilities as "Special Focus Facilities" (SFF) that receive more frequent inspections. If a facility is on the SFF list, that is a significant warning — but not all low-rated facilities are on the list due to limited SFF slots.

Given these complexities, here is our recommended approach to using CMS star ratings:

• Use overall ratings to create a shortlist, not a final decision. Filter to 3+ stars to eliminate the bottom tier, then look deeper at individual domains.
• Prioritize health inspection ratings. This is the most reliable domain because it is based on independent, on-site observation. A facility with a 5-star overall rating but a 2-star health inspection rating deserves scrutiny.
• Look at staffing data directly, not just the rating. Check the actual RN hours per resident per day and total nursing HPRD on CareFindPeek's facility pages. Compare to the national average (approximately 3.6 total HPRD and 0.55 RN HPRD).
• Read the deficiency details, not just the count. Five deficiencies with minimal harm potential are very different from five deficiencies with actual harm or immediate jeopardy findings.
• Check the trend. Is the facility improving or declining? Compare the last 2-3 survey cycles. A facility trending from 3 stars to 4 stars is often a better bet than one trending from 5 to 4.
• Compare within your geographic area. Since health inspection ratings are state-normalized, comparing facilities within the same state or metro area gives you the most meaningful comparison.

Star ratings are a powerful starting tool, but they are not a substitute for visiting facilities, asking hard questions, and trusting your own observations. Use our facility comparison directory to see all the underlying data behind the ratings.