Reference

How the CMS Five-Star staffing rating is calculated

The staffing component of a nursing home’s Five-Star rating is arithmetic performed on data the facility submits itself. This page describes that arithmetic as CMS documents it: where the numbers come from, the six measures, the point scoring, and the three exceptions that bypass scoring entirely.

What this page is, and is not. Business Management Systems builds scheduling software. We have never operated a nursing home, and nothing here is legal, compliance or clinical advice. Every factual statement below reports what a named published source says, with a link to it. Where we could not verify something against a primary source, we left it out.

Sources reviewed and links checked August 2026.

First, the question everyone asks

Is the federal 3.48 HPPD minimum still in effect?

No. The 2024 CMS Minimum Staffing Standards rule, which would have required 3.48 total nurse hours per resident day, 0.55 from registered nurses, 2.45 from nurse aides, and a registered nurse on site 24 hours a day, no longer exists.

It went through three separate unwindings, and the middle one is the reason its status is settled rather than contested:

  • Finalized May 2024. CMS publishes the Minimum Staffing Standards for Long-Term Care Facilities rule.
  • Vacated April 2025. A federal district court in the Northern District of Texas vacates the HPRD minimums and the 24-hour RN requirement, holding CMS exceeded its statutory authority.
  • Frozen July 2025. Section 71111 of P.L. 119-21 bars CMS from implementing, administering or enforcing those provisions until 2034.
  • Repealed February 2026. A CMS interim final rule formally rescinding the requirements takes effect on 2 February 2026.

Appeals arising from the litigation have continued, but the congressional moratorium operates independently of them. CMS is barred from enforcing the vacated provisions until 2034 regardless of how any appeal is decided.

Repeal published in the Federal Register (document 2025-21792), effective 2 February 2026.

What the repeal did not touch

The rating survived the rule

The repealed rule set staffing minimums. It did not create the reporting system or the public rating, and neither was affected by its removal.

Payroll-Based Journal reporting

Quarterly electronic submission of direct-care staffing data, required under Section 6106 of the Affordable Care Act since 2016. Unchanged.

The Five-Star staffing rating

Computed from PBJ submissions and published on Care Compare, where families, referral sources and hospital partners can see it. Unchanged.

The longstanding statutory baseline

The requirements predating the 2024 rule, including a registered nurse for eight consecutive hours a day and a full-time Director of Nursing, remain in the Social Security Act.

Where the numbers come from

Payroll-Based Journal, in mechanics

Every figure in the staffing rating originates in PBJ. CMS’s Technical Users’ Guide states plainly that the source for reported staffing hours is the PBJ system, that data are submitted quarterly, and that they are due 45 days after the end of each reporting period.

What a submission contains

  • Direct-care staffing hours by day, tied to a unique employee identifier that carries no personally identifiable information
  • A CMS job title code drawn from a fixed set of 40
  • A pay type code distinguishing exempt, non-exempt and contract
  • Agency and contract staff hours, which are reported alongside employee hours rather than excluded

Two mechanics that surprise people

Only accepted data counts. The Users’ Guide specifies that only data submitted and accepted by the deadline are used in staffing calculations. Practitioners quoted by McKnight’s note that CMS does not permit corrections once the submission deadline has passed.

Census is no longer submitted. Facilities do not report daily census through PBJ. As Forvis Mazars describes it, the daily resident census is derived automatically from MDS resident assessments.

Submission timing and source data per the CMS Care Compare Nursing Home Five-Star Quality Rating System Technical Users’ Guide. Correction deadline reported by McKnight’s Long-Term Care News. Census derivation described by Forvis Mazars.

The scoring

Six measures, 380 points

Before July 2022 the staffing rating averaged two measures. Since the July 2022 methodology update it uses six, each scored on predetermined cut points, with the totals compared against thresholds to produce a rating of one to five stars.

Scroll the table sideways to see every column.

Measure Maximum points What it reflects
Case-mix adjusted total nurse HPRD 100 RN, LPN and nurse aide hours per resident day, adjusted for resident acuity
Case-mix adjusted RN HPRD 100 Registered nurse hours per resident day, adjusted for acuity
Weekend total nurse HPRD 50 The same calculation averaged over Saturdays and Sundays only
Total nurse turnover 50 Percentage of nursing staff who left over a twelve-month period
RN turnover 50 Percentage of registered nurses who left over a twelve-month period
Administrator turnover 50 Number of administrators who left over a twelve-month period

Point maxima as reported by PBJ Central summarizing CMS guidance. The 380-point total and the summing method are stated in the CMS Technical Users’ Guide, which also notes that a facility missing valid data for one or more turnover measures may be rated on the remaining measures with its score rescaled against a lower maximum.

The part that bypasses the scoring

Three findings that produce a one-star staffing rating regardless of points

The six measures and the 380-point scale describe the normal path. Separately from that path, CMS applies downgrade rules that assign a one-star staffing rating outright.

A facility can score well on every measure and still publish at one star because of what its data shows or does not show.

Failing to submit

A facility that does not submit staffing data for the quarter receives a one-star staffing rating. Reporting on the practical effect, McKnight’s describes the penalty as lasting for the quarter.

Four or more days with no RN hours

The Technical Users’ Guide states that providers submitting data showing four or more days in the quarter with no RN staffing hours, under job codes 5 through 7, on days when one or more residents were in the nursing home, receive a one-star staffing rating for the quarter.

Audit failure or non-response

PBJ submissions are subject to CMS audit. Facilities that do not respond to an audit, or whose audit identifies significant discrepancies, receive the same downgrade.

A distinction worth noting. The four-day threshold governs the rating. It is not the threshold surveyors use. PBJ Central, summarizing CMS guidance, reports that while the staffing data report uses four or more days without RN coverage, surveyors are instructed to consider a citation where there is a single day without coverage.

Downgrade rules per the CMS Technical Users’ Guide and the Oklahoma State Department of Health’s PBJ 101 briefing. Quarterly duration reported by McKnight’s, which counted 535 nursing homes affected by the RN coverage rule in a single quarter.

The measure that behaves unlike the others

How turnover is actually computed

Three of the six measures are turnover measures, so together they carry 150 of the 380 points. Unlike the staffing level measures, they are not calculated from a single quarter, and they depend on something no other measure depends on: the stability of your employee identifiers.

The definitions

  • The calculation requires six consecutive quarters of PBJ data
  • Only employees working at least 120 hours in a 90-day period are eligible to be counted
  • An employee who does not work for 60 days is counted as separated
  • Both employed and contracted workers are included
  • Employee ID numbers are what CMS uses to identify gaps in employment

Why that last point matters

Because separations are detected by watching identifiers disappear, anything that changes identifiers looks identical to staff leaving.

Practitioners describe turnover figures spiking toward 100% in a quarter following the implementation of new systems that change the employee IDs used in the calculation. Changing payroll vendors or ownership can produce the same effect. The staff never left. The numbers say they did.

The same reporting notes a second version of this: a facility that submits its corporate staff one quarter and omits them the next will have CMS read those people as turned over.

There is a documented remedy. Reporting on PBJ pitfalls, McKnight’s quotes practitioner Napier describing that facilities have the ability to link an old employee ID to a new one by providing CMS with a “linking table” to convert the information. If an identifier change is coming, this is the mechanism that exists to address it.

Turnover definitions per the CMS Technical Users’ Guide and Bonadio. Identifier effects and the linking table reported by McKnight’s and StarPRO. A separate July 2023 exclusion rule applies where a facility reports administrator hours for five or more individuals on four or more days in a quarter, per LeadingAge New York.

The calculation itself

What HPPD is, arithmetically

Hours per patient day, also written HPRD or PPD, divides direct-care hours by resident days over the same period. For a single day it is total direct-care hours worked divided by the resident census.

Over a quarter, the numerator is the sum of direct-care hours across all days and the denominator is the sum of daily census across those days.

Three properties follow from the arithmetic, and they explain most of the confusion people have with the number:

  • It moves when census moves. The same staffing hours produce a different HPPD at a census of 78 than at 94.
  • A facility average hides unit variation. Averaging across units can show an acceptable figure while an individual unit sits well below it.
  • Case-mix adjustment is separate. The rating uses case-mix adjusted figures derived from resident acuity, so a facility’s own unadjusted HPPD will not equal the adjusted figure CMS scores.

Daily

HPPD = total direct-care hours ÷ resident census

Over a period

HPPD = Σ direct-care hours ÷ Σ daily census

This page does not publish a target HPPD. There is no single correct figure: requirements differ by state, the rating is scored on curves derived from national distributions, and the case-mix adjustment means an unadjusted number is not directly comparable to a scored one.

State requirements

Separate from the federal question entirely

State staffing requirements were never created by the federal rule and were not removed with it. They are set independently, they differ substantially in structure as well as in number, and they change in both directions.

Rather than reproduce figures that would go out of date, here is how the requirements differ structurally, so you know what to look for in your own state:

How the floor is expressed

Some states set a numeric direct-care hours per resident day minimum. Others set ratios of staff to residents by shift. Many set neither, requiring only “sufficient” staffing without a defined figure.

Whether composition is specified

Several states divide the total, requiring a minimum share from licensed nurses or from certified aides. A facility can meet a total figure and still miss a composition requirement.

How compliance is measured

Some states measure against the same PBJ data submitted to CMS, on a quarterly basis. Others use their own inspection or documentation processes, on different periods.

What non-compliance carries

Enforcement ranges from citation to financial penalty, and in some states penalties are calculated per day. Several states have adjusted their requirements upward and others downward in recent years.

For current requirements in a specific state, the authoritative source is that state’s own health department or the statute itself. The Long-Term Care Community Coalition and the National Consumer Voice for Quality Long-Term Care both publish periodically updated multi-state overviews. We do not reproduce state figures here because they change, and a stale number is worse than no number.

Where scheduling software fits

What software can and cannot do about any of this

We make scheduling software, so treat this section as what it is. It is also the part where being straight is more useful than being enthusiastic.

What it does not do. No scheduling system files your PBJ submission, changes how CMS scores you, or interprets your state’s requirements. Snap Schedule holds the staffing data a submission is built from. The filing stays with you.

What it does do. Every figure above originates in hours that were scheduled and worked. A system that records those hours accurately by employee, day, job classification and unit, keeps employees and contract staff distinguishable, and shows HPPD against census before the quarter closes, means the reconciliation is a review rather than a reconstruction.

Snap Schedule 365 scheduled HPPD report for a nursing home showing facility census and hours with per unit breakdowns for licensed nurses, aides and supervisors

Frequently asked questions

Common questions about PBJ and the staffing rating

Is the federal 3.48 HPPD minimum staffing rule still in effect?

No. The 2024 CMS Minimum Staffing Standards rule, which would have required 3.48 total nurse hours per resident day, 0.55 from registered nurses, 2.45 from nurse aides and a registered nurse on site 24 hours a day, was vacated by a federal district court in April 2025, frozen by Congress in July 2025 under Section 71111 of P.L. 119-21, and formally repealed by an interim final rule effective 2 February 2026. The congressional moratorium bars enforcement until 2034 independently of any appeal.

How is the CMS Five-Star staffing rating calculated?

It is computed from Payroll-Based Journal submissions using six measures: case-mix adjusted total nurse hours per resident day, case-mix adjusted RN hours per resident day, weekend total nurse hours per resident day, total nurse turnover, RN turnover and administrator turnover. Each measure earns points against predetermined cut points, the points are summed, and the total is compared against thresholds to assign one to five stars. The maximum possible score is 380 points.

What causes an automatic one-star staffing rating?

CMS applies downgrade rules that bypass the point scoring. Failing to submit staffing data for the quarter produces a one-star staffing rating. So does submitting data showing four or more days in the quarter with no RN staffing hours on days when one or more residents were present. So does failing to respond to a CMS audit, or an audit identifying significant discrepancies.

When is PBJ data due?

The CMS Technical Users’ Guide states that PBJ data are submitted quarterly and are due 45 days after the end of each reporting period. It also specifies that only data submitted and accepted by the deadline are used in staffing calculations.

Can PBJ data be corrected after the deadline?

Practitioners quoted by McKnight’s Long-Term Care News report that CMS does not allow corrections once the submission deadline has passed. The Technical Users’ Guide’s statement that only data submitted and accepted by the deadline are used is consistent with this.

How does staff turnover affect the Five-Star staffing rating?

Turnover accounts for three of the six measures, carrying up to 150 of the 380 available points across total nurse turnover, RN turnover and administrator turnover. The calculation uses six consecutive quarters of PBJ data. Employees are eligible to be counted if they worked at least 120 hours in a 90-day period, and an employee who does not work for 60 days is counted as separated. Both employed and contracted workers are included.

Why did our reported turnover jump after we changed payroll systems?

CMS identifies separations by tracking employee ID numbers across quarters, so a change in identifiers is indistinguishable from staff leaving. Practitioners describe turnover figures spiking toward 100% in the quarter following a system change that altered employee IDs. McKnight’s reports that facilities can link an old employee ID to a new one by providing CMS with a linking table to convert the information.

How is HPPD calculated?

Hours per patient day divides direct-care hours by resident days. For a single day, it is total direct-care hours divided by the resident census. Over a longer period, it is the sum of direct-care hours divided by the sum of daily census across those days. The Five-Star rating uses case-mix adjusted figures derived from resident acuity, so an unadjusted internal calculation will not match the adjusted figure CMS scores.

Do agency and contract staff count in PBJ reporting?

Yes. Contract and agency hours are reported alongside employee hours, distinguished by pay type code rather than excluded. Contracted workers are also included in the turnover measures.

Does CMS still require facilities to report daily census in PBJ?

No. Forvis Mazars describes the daily resident census as being derived automatically from MDS resident assessments rather than reported by the facility through PBJ.

Do state staffing requirements still apply after the federal repeal?

State requirements are set independently of the federal rule and were not affected by its repeal. They differ in structure as well as in figures: some states set numeric hours per resident day minimums, some set staff-to-resident ratios by shift, and many require only sufficient staffing without a defined number. Several also specify composition, such as a minimum share of hours from licensed nurses. For current requirements, the authoritative source is the state health department or statute.

Does the four-day RN rule mean three days without RN coverage is acceptable?

The four-day threshold governs the staffing rating, not survey citations. PBJ Central, summarizing CMS guidance, reports that while the staffing data report uses four or more days without RN coverage, surveyors are instructed to consider a citation where there is a single day without coverage. The separate longstanding statutory requirement for a registered nurse for eight consecutive hours a day was not affected by the 2026 repeal.

Related

Scheduling software for skilled nursing and long-term care

Snap Schedule 365 keeps the staffing record these figures are built from: hours by employee, day, job classification and unit, employees and contract staff kept distinct, and HPPD against census while the quarter is still open.

Prefer to talk? Call 1-800-450-4230, Monday to Friday, 9:00 a.m. to 5:00 p.m. Pacific Time.