How the CMS 5-Star Nursing Home Rating Is Calculated: Health Inspections, Staffing, Quality Measures, Weights, and Cutpoints
The CMS 5-star Quality Rating System for nursing homes assigns a 1–5 star rating to every Medicare-certified facility. The composite rating is calculated from three domain ratings — health inspections, staffing, and quality measures — each scored independently before being combined. Understanding how each domain is scored, what data feeds it, and how the cutpoints are set is essential for interpreting what the stars mean and where they fall short.
The process, step by step
Calculate the health inspection domain score
CMS sums weighted deficiency scores across the three most recent standard surveys and any complaint surveys in the past 36 months. Each deficiency is weighted by scope-and-severity tag (A through L). Immediate jeopardy tags (J–L) receive the highest weight. The facility's summed score is compared against all other facilities in the same state to generate a state-relative percentile ranking.
Source ↗Assign the health inspection star rating
CMS assigns 1–5 stars based on the facility's state percentile rank for the health inspection score: the worst 20% receive 1 star, the next 25% receive 2 stars, the middle 25% receive 3 stars, the next 20% receive 4 stars, and the best 10% receive 5 stars. The percentiles are recalculated quarterly. A recent history of immediate jeopardy findings automatically caps the health inspection rating at 2 stars.
Source ↗Calculate staffing hours per resident day from PBJ data
CMS uses Payroll-Based Journal (PBJ) quarterly staffing reports to calculate total nurse staffing hours per resident day (HPRD) and RN-specific hours per resident day. Census data from the MDS assessment system provides the resident-day denominator. The staffing calculation uses a case-mix adjustment derived from the RUG classification of residents to account for acuity differences across facilities.
Source ↗Assign the staffing star rating
CMS converts case-mix adjusted total HPRD and RN HPRD into a combined staffing rating using a lookup table of national cutpoints. Facilities reporting no RN hours on any shift during the quarter automatically receive 1 star for staffing, regardless of total HPRD. The staffing domain was the most structurally changed in the 2023 methodology update, which replaced the old self-reported staffing data with PBJ.
Source ↗Calculate quality measure scores
CMS computes scores for 15 quality measures drawn from MDS resident assessment data and Medicare claims. Long-stay measures (residents in the facility 100+ days) include pressure ulcer prevalence, falls with injury, catheter use, and antipsychotic medication use. Short-stay measures cover high-risk pressure ulcers, pain, and functional improvement. Each measure is expressed as a percentage or rate.
Source ↗Assign the quality measures star rating
Each of the 15 quality measures is scored on a 1–5 scale using national cutpoints (not state-relative). The 15 measure scores are weighted and summed into a domain score. The domain score is then converted to a 1–5 star rating using national distribution cutpoints targeting 20%/30%/30%/15%/5% distribution from worst to best. Quality measures star ratings are updated quarterly.
Source ↗Derive the composite overall star rating
The composite overall star rating starts with the health inspection star rating, then adjusts up or down based on the staffing and quality measures ratings. If both staffing and quality measures are rated 5 stars, the overall rating can be boosted by one star. If either staffing or quality measures is 1 star, the overall rating is capped at 4 stars. The result is rounded to the nearest integer 1–5.
Source ↗
Overview: three domains, one composite
The CMS 5-Star Quality Rating System was introduced in December 2008 as a tool for consumers comparing nursing homes. It assigns a 1–5 star overall rating derived from three component domain ratings: health inspections, staffing, and quality measures. Each domain is scored independently before the composite is derived. The system was substantially updated in 2015 (adding quality measures) and again in 2022–2023 (replacing self-reported staffing with PBJ data and revising quality measure selection).
The rating is recalculated and published monthly on CMS Care Compare. It applies to all Medicare-certified skilled nursing facilities and Medicaid-certified nursing facilities. As of the current Fonteum dataset, the median overall star rating nationally is 3 stars, with 1-star facilities representing roughly 20% of the total population.
One important caveat: the rating is a relative ranking within a comparison pool, not an absolute measure. Health inspection stars are state-relative, meaning a 4-star facility in one state may have a different absolute deficiency score than a 4-star facility in another. Only quality measures are benchmarked to national cutpoints.
Health inspection domain
The health inspection domain is the most heavily weighted driver of the composite. It is calculated from three years of standard annual surveys plus complaint surveys. Each deficiency is assigned a point value based on its scope-and-severity tag: isolated potential harm (A/B/C) receives the lowest weight, widespread actual harm without immediate jeopardy (H/I) receives higher weight, and immediate jeopardy (J/K/L) receives the highest weight.
The raw score is the sum of weighted deficiency points across all surveys in the 36-month window, with more recent surveys weighted more heavily than older ones (a recency adjustment). The weighted total is then compared against all other facilities in the same state to generate a percentile ranking. This state-relative approach means a facility's star rating can change based on the average performance of its peer group — not just its own survey results.
Any nursing home with an immediate jeopardy citation in the most recent standard survey, the two preceding standard surveys, or any complaint survey in the past 12 months is automatically capped at 2 health inspection stars, regardless of its percentile score. As of Fonteum's star-ratings study, 1,753 facilities hold 4 or 5 health inspection stars while also carrying G-or-higher deficiency citations — illustrating how the scoring mechanism can produce counterintuitive results when serious citations are older than 12 months.
Staffing domain
Staffing is calculated from CMS Payroll-Based Journal (PBJ) data — the mandatory quarterly electronic payroll submission that replaced self-reported staffing in 2017. PBJ captures daily paid hours for every employee type: RNs, LPNs, CNAs, therapists, and administrative staff. Only direct nursing hours (RN, LPN, CNA) count toward the staffing rating.
CMS calculates two key metrics: total nurse hours per resident day (total HPRD = RN + LPN + CNA hours ÷ resident days) and RN-only hours per resident day. Both are case-mix adjusted using a facility-specific expected HPRD derived from the distribution of resident acuity in the MDS assessment data. The case-mix adjustment prevents facilities that specialize in high-acuity residents from appearing understaffed relative to facilities with lower-acuity populations.
Since the 2023 methodology update, the staffing domain has added a second sub-metric for registered nurse hours per resident day on weekends. Facilities that report zero RN hours on any calendar day in any quarter receive an automatic 1-star staffing rating. Fonteum's county-level staffing deserts analysis — built on 1.32 million PBJ records — found that 238 counties have no Medicare-certified nursing home at all, independent of staffing levels.
Quality measures domain
The quality measures domain covers 15 measures drawn from MDS resident assessment data and Medicare fee-for-service claims. Long-stay measures (100+ continuous days) include: percentage of residents with a urinary tract infection, with worsening pressure ulcers, with falls resulting in major injury, using an antipsychotic medication, experiencing pain, with a catheter, experiencing physical restraints, and with unexplained weight loss. Short-stay measures cover: percentage of residents with pressure ulcers, with moderate to severe pain, re-hospitalized after admission, who improved in function, and with new or worsened pressure ulcers.
Each measure is scored against national cutpoints — the distribution of all facilities nationally — not state-relative. CMS sets the cutpoints so that approximately 40% of facilities score 5 stars on quality measures nationally. The measure scores are averaged (with adjustments for data completeness) into a domain score, which is then converted to a 1–5 star rating.
Quality measures are the domain most susceptible to selection effects: facilities that specialize in short-stay rehabilitative care (post-acute recovery) will have systematically different case mixes than those serving long-stay memory care residents. CMS applies risk adjustment to most measures, but residual case-mix effects remain. The antipsychotic medication measure is one notable exception where risk adjustment is not applied, on the grounds that elevated antipsychotic use is itself a quality concern regardless of case mix.
Composite rating and cutpoints
The composite overall star rating is not a simple average of the three domain ratings. The derivation algorithm starts with the health inspection star rating, then applies adjustments based on the staffing and quality measures ratings. Specifically: if staffing is 5 stars AND quality measures is 5 stars, the overall rating is raised by 1 star (max 5). If either staffing or quality measures is 1 star, the overall rating is capped at 3 stars. If both are 1 star, the overall is capped at 2 stars.
These adjustment rules mean the health inspection rating is the floor and the most structurally important component. A facility cannot achieve a high overall rating without a decent health inspection rating, but a 5-star health inspection rating paired with poor staffing and quality measures will still produce a modest overall rating.
CMS recalculates overall ratings monthly and publishes them on Care Compare. When a health inspection survey completes, the facility's rating can change immediately upon CMS acceptance. The cutpoints for quality measures and staffing are adjusted quarterly based on the national distribution — meaning a facility's star rating can drift even if its underlying data does not change.
Limitations of the 5-star system
The 5-star rating has well-documented limitations. Because health inspection stars are state-relative, the same absolute deficiency burden produces different ratings in different states. A state with consistently high deficiency rates will have facilities that look better than their national peers on paper.
The PBJ staffing data, while more reliable than self-reported data, captures paid hours rather than hours worked on the floor. Agency and contract staff are included, which may overstate the continuity of care residents experience. PBJ data is also subject to data entry error and delayed corrections.
Fonteum's star ratings vs. actual harm study found 1,753 facilities with 4 or 5 health inspection stars that also carried G-or-higher harm-level deficiency citations in the underlying survey data. The discrepancy arises primarily because the deficiency scoring is recency-weighted and state-relative — a serious but older citation can wash out over time even if the pattern of harm is persistent.
Frequently asked questions
- How often is the CMS 5-star rating updated?
- Overall ratings are recalculated and published monthly on CMS Care Compare. Health inspection ratings update when new survey data is accepted by CMS, which occurs within days to weeks of survey completion. Quality measures and staffing ratings update quarterly when new PBJ and MDS data cycles are processed.
- Does a 5-star rating mean a nursing home is safe?
- A high star rating reflects relatively better performance on the measured dimensions — health inspection scores, staffing levels, and quality measure rates — compared to peers in the same state or nationally. It does not guarantee safety. Fonteum's study found over 1,750 facilities with 4 or 5 health inspection stars that also had documented actual-harm deficiency citations in the underlying data. Consumers should review the actual CMS-2567 deficiency record alongside the star rating.
- Why do health inspection stars use state-relative percentiles while quality measures use national cutpoints?
- CMS made this design choice at inception and has retained it through subsequent updates. The rationale is that state survey agencies conduct inspections with different frequency and intensity, making national comparisons methodologically problematic for health inspections. Quality measures, by contrast, draw from MDS and claims data submitted through nationally standardized systems, making national benchmarking more defensible.
- What is the PBJ and why does it matter for staffing stars?
- The Payroll-Based Journal (PBJ) is a CMS system that collects mandatory quarterly electronic payroll data from all certified nursing homes. Since 2017, PBJ has replaced self-reported staffing as the data source for the staffing domain of the 5-star rating. Because PBJ captures actual payroll records rather than administrator-reported estimates, it is harder to manipulate — and audit studies have found PBJ reports lower staffing levels on average than prior self-reported data.
Reviewed by Jennifer Montecillo, MD, medical reviewer. Non-practicing medical reviewer. Last reviewed: June 2026. Data last updated: 2026-06-03.
This explainer draws from federal primary sources. Every figure is traceable to a specific CMS dataset with provenance documentation. Data sources → Corrections log → Methodology library →
Federal data is in the public domain under U.S. Government Works. Fonteum analysis and synthesis are copyright Fonteum, Inc.