Zero-RN days: how often US nursing homes ran a day with no registered nurse on the floor

Every figure in this study comes from a single frozen federal source: the CMS Payroll-Based Journal, the daily staffing record that every Medicare- and Medicaid-certified nursing home in the country is required to submit from its own payroll system. We read the 2025 Q2 snapshot — the quarter spanning April 1 through June 30, 2025 — and asked one narrow, answerable question. On how many days did a nursing home that had residents in its beds report zero registered-nurse direct-care hours?

The answer is 77,542 days. Out of 1,322,254 facility-days with residents present, 5.86% recorded no RN direct-care time at all. Counted a different way, 5,062 of the country's 14,537 reporting nursing homes — better than one in three — had at least one such day during the quarter. This is not a survey estimate or a model. It is a count of rows in the payroll file.

The short answer. In the CMS Payroll-Based Journal's 2025 Q2 snapshot, 5.86% of US nursing-home resident-days — 77,542 of 1,322,254 — recorded zero registered-nurse direct-care hours. More than a third of facilities (5,062 of 14,537, or 34.8%) had at least one such day. Rates ranged from 27.9% in Louisiana to 0.2% in Rhode Island.

We publish this now for a specific reason. The federal minimum staffing standard for long-term-care facilities — finalized in 2024, and including a requirement that a registered nurse be on site 24 hours a day, 7 days a week — was rescinded, with the rescission effective February 2, 2026. For the first time in the modern history of the program, there is no national floor under nurse staffing in nursing homes. The numbers below predate that change. They describe the baseline the country was already running while the rule was still on the books, and they are the baseline that now carries forward with nothing federal beneath it.

What we counted, precisely

The Payroll-Based Journal, or PBJ, is the most granular public record of who actually worked in a nursing home and when. Unlike the staffing figures that facilities once self-reported during an annual survey week, PBJ is drawn from payroll and timekeeping systems and submitted every quarter. Each row is one facility, one calendar day, with hours worked broken out by job category: registered nurses, licensed practical nurses, certified nursing assistants, nurse aides in training, and medication aides, along with the day's resident census.

PBJ separates registered-nurse time into three distinct columns. There is the RN director of nursing, the administrator-clinician who runs the nursing department. There is the RN with administrative duties, who may spend the day on care plans, compliance, and paperwork rather than at the bedside. And there is the RN direct-care field — the column for registered nurses doing hands-on resident care. Our headline metric is built on that third column alone.

A zero-RN day, throughout this report, means a facility-day on which the RN direct-care field is zero. We restrict the count to facility-days where the midnight resident census was above zero, because a day with no residents present imposes no staffing obligation and would only dilute the rate. That single filter — residents present — is what separates the 1,322,254 facility-days we analyze from the slightly larger raw row count in the file.

This definition is deliberately conservative in one direction and deliberately strict in another. It is conservative because it does not require all nursing care to be absent: a licensed practical nurse and several certified nursing assistants may well have been on duty. It is strict because it isolates the one role — a registered nurse assigned to direct care — whose absence is most consequential for catching a resident whose condition is changing. The distinction matters enough that we measured the stricter alternative too, and we report it below so readers can see exactly how much of the gap is bedside versus administrative.

Why a payroll file changed the question

For most of the program's history, the public simply could not answer how often a nursing home ran a day without a registered nurse, because the only staffing data came from a single self-reported survey week. A facility told its state surveyor how many hours it had staffed during one designated week each year, the surveyor recorded it, and that figure was extrapolated across twelve months. The method invited rounding up. Facilities knew which week they were reporting; they could not be caught on the days they were not asked about.

The Payroll-Based Journal, mandated under the Affordable Care Act and operational since 2016, replaced that annual snapshot with a daily, payroll-derived record. Because the hours are pulled from the same timekeeping systems that cut paychecks, they are far harder to inflate and can be audited against payroll tax filings. The shift is what makes a question like this one answerable at all. Before PBJ, "how many days had no bedside RN" was a guess. After PBJ, it is a query that returns 77,542. The cost of that precision is that the data is only as granular as the job codes facilities use to file it — which is why, as the weekday pattern later in this study shows, the column a nurse is booked to matters as much as whether a nurse was present.

How common a zero-RN day is

The national rate is 5.86% of resident-days. That number is easy to under-feel until it is set against the size of the system. A typical reporting facility in this snapshot carried about 85 residents on an average day. Multiply 5.86% across 1.3 million facility-days and the result is 77,542 separate occasions, last spring, on which a nursing home with residents in its beds logged no registered-nurse direct-care hours.

Measured by facility rather than by day, the reach is wider. Of 14,537 nursing homes that reported in the quarter, 5,062 — 34.8% — had at least one zero-RN day. The remaining 9,475 facilities reported a registered nurse in direct care on every single day they had residents. The system, in other words, is not uniformly thin. It is sharply divided between facilities that never miss and facilities that miss routinely.

That division is the real finding, and it is worth stating plainly: zero-RN days are not spread evenly across the country's nursing homes like background noise. They concentrate. A minority of facilities accounts for the overwhelming majority of the days, and within that minority a small group accounts for a wildly disproportionate share.

The geography of a missing nurse

Sort the same facility-days by state and the national average dissolves into a map of enormous disparity. The table below shows the eight states with the highest share of resident-days recording zero RN direct-care hours, and the five lowest, among states with at least twenty reporting facilities.

State	Resident-days zero-RN	Facilities with ≥1 zero-RN day
Louisiana	27.9%	79.8%
Oklahoma	25.4%	80.3%
Arkansas	20.8%	78.7%
Texas	14.2%	62.2%
Missouri	11.8%	62.2%
Georgia	8.7%	51.4%
Kansas	8.0%	52.2%
Oregon	7.5%	55.6%
…	…	…
Florida	0.9%	11.5%
Colorado	0.9%	10.7%
Maryland	0.8%	12.3%
New Hampshire	0.7%	18.9%
Rhode Island	0.2%	7.0%

The top of the table is a contiguous block. Louisiana, Oklahoma, Arkansas, Texas, and Missouri — a band running through the south-central United States — sit far above every other state. In Louisiana, more than one in four resident-days recorded no RN direct-care time, and four out of five facilities had at least one such day. In Oklahoma, 80.3% of facilities did. The bottom of the table is just as concentrated: in Rhode Island, only 0.2% of resident-days and 7.0% of facilities show the pattern at all.

The spread between Louisiana's 27.9% and Rhode Island's 0.2% is roughly 120-fold. No plausible difference in how sick residents are, how rural the geography is, or how old the buildings are can carry a gap that size on its own. A disparity this wide points at the things states actually control or shape — Medicaid reimbursement rates, state-level staffing rules layered on top of the federal floor, the depth of the local registered-nurse labor market, and the ownership structures that concentrate in some states more than others. The PBJ file cannot adjudicate among those causes. What it can do, unambiguously, is show that geography is the single strongest predictor of whether a resident's facility ran a day without a bedside registered nurse.

Reading the south-central cluster

The states at the top of the table are not a random scatter; they form a recognizable region. Louisiana, Oklahoma, Arkansas, Texas, and Missouri share several features that the staffing literature has long associated with thin nurse coverage. Medicaid pays for the majority of long-stay nursing-home days, and these states sit toward the lower end of Medicaid daily reimbursement, which compresses the budget available for the most expensive labor category — registered nurses. Several also have large rural footprints where the registered-nurse labor pool is genuinely shallow, so even a facility willing to pay cannot always fill an RN line. And several have permitted, in practice, staffing models that lean on licensed practical nurses and aides for day-to-day coverage with a registered nurse available on call rather than on site.

None of that is destiny. Within every high-rate state there are facilities with zero zero-RN days, which means the regional average is a distribution, not a uniform condition. But the consistency of the cluster is the point. When the five leanest states are geographically adjacent and share a reimbursement and labor profile, the pattern is structural rather than incidental. It is the kind of pattern a national floor is designed to lift — and the kind that, absent a floor, a low-reimbursement state has little immediate pressure to change. The lowest-rate states, by contrast — Rhode Island, New Hampshire, Maryland — tend to combine higher reimbursement, denser RN labor markets, and in some cases their own state staffing standards, which will continue to operate after the federal one is gone.

The practical implication is that the rescission of the federal floor does not land evenly. It lands hardest where the floor was doing the most work, which is exactly the south-central band at the top of this table.

For readers who want the per-facility detail behind any state, the full daily record is published, with source provenance, on the PBJ staffing dataset page, and the facility-quality context for nursing homes more broadly lives in the Care Compare nursing-home module.

The persistent few

Averages hide tails, and the tail here is the part that should hold a regulator's attention. We grouped every facility by how many zero-RN days it reported across the 91 days of the quarter.

Zero-RN days in the quarter	Facilities	Share of all facilities
0	9,475	65.2%
1–6	2,650	18.2%
7–29	1,497	10.3%
30–90	864	5.9%
All 91	51	0.4%

Most facilities that have any zero-RN days have only a handful. But 864 facilities reported zero RN direct-care hours on between 30 and 90 days of the quarter — that is, on a third to nearly all of their reported days. And 51 facilities reported zero RN direct-care hours on every single day they had residents. Across the entire 91-day quarter, by the payroll record they submitted themselves, a registered nurse was never logged in direct care. Among the 5,062 facilities with any zero-RN days, the average count was 15.3 days; 536 facilities recorded a zero-RN day on at least half of their resident-days.

These are not facilities that had a bad weekend when a nurse called out. They are facilities whose ordinary operating model, for an entire quarter, did not include a registered nurse at the bedside. That a facility can run that way and remain certified is precisely what the now-rescinded 24-hour, 7-day RN requirement was written to change — and precisely what, with the requirement gone, nothing federal now prevents.

Per our research doctrine, we do not name individual facilities or attach these counts to any provider profile. CMS publishes the underlying file at the facility level for anyone who wishes to look; our role is to report the aggregate pattern, not to rate, warn against, or endorse any specific home.

Direct care versus the whole department

Because the headline metric isolates the RN direct-care column, an obvious question follows: on those 77,542 days, was there truly no registered nurse in the building at all, or simply none assigned to direct care?

The honest answer is usually the latter. When we widen the definition to require zero RN hours across all three job codes — direct care, administrative, and director of nursing combined — the count falls sharply, to 7,399 facility-days, or 0.56% of resident-days, spread across 1,373 facilities. So on the large majority of zero-direct-care-RN days, the payroll shows an RN somewhere in the department, most often the director of nursing or an administrative RN.

That nuance cuts in two directions, and both belong in the record. It means the 5.86% figure should not be read as "no nurse of any kind was present" — that would overstate it. But it also means that on those days, the registered nurse who was present was booked to administration or department management, not to assessing residents. A director of nursing buried in compliance paperwork is not the same safeguard as a registered nurse walking the floor. The 0.56% all-RN-zero figure is the more extreme failure; the 5.86% direct-care figure is the more common and, for day-to-day resident safety, arguably the more revealing one. We report both rather than choosing the one that tells a cleaner story.

Nationally, the same file puts registered-nurse direct-care staffing at about 0.43 hours per resident-day, RN hours across all three codes at about 0.62 per resident-day, and total nurse staffing — RNs, LPNs, and aides combined — at about 3.70 hours per resident-day. The rescinded federal rule had set a total floor of 3.48 hours and an RN-specific expectation well above what the direct-care column alone delivers. The zero-RN-day count is the discrete, countable edge of that broader thinness.

What a registered nurse is there to do

The reason a missing registered nurse matters more than a missing aide is a matter of scope of practice, not headcount. Certified nursing assistants deliver the bulk of hands-on personal care — bathing, feeding, transferring, toileting — and licensed practical nurses administer many medications and perform routine clinical tasks. Both are essential, and on a zero-RN-direct-care day they are very often present. But neither role is authorized to perform the full nursing assessment that turns a vague change — a resident who is suddenly confused, breathing harder, or running a low-grade fever — into a clinical decision: escalate, call the physician, send to the emergency department, or watch and reassess.

That assessment-and-escalation function is the registered nurse's distinct contribution, and it is time-sensitive. A resident whose sepsis is caught in its early hours has a very different trajectory from one whose decline is noticed at the next shift change. The early-warning literature on nursing homes consistently finds that registered-nurse presence is associated with fewer avoidable hospital transfers and faster response to acute change. A day with aides and an LPN but no bedside RN is not a day without care; it is a day without the specific clinician trained to recognize, and act on, the moment a stable resident stops being stable.

This is why the count is built on the direct-care column rather than the broader department total. A director of nursing finishing a survey-readiness binder in the office is a real registered nurse doing real work, but that work is not standing between a deteriorating resident and a missed window. The 77,542 days are days when, by the facility's own record, that bedside safeguard was not staffed.

A counterintuitive weekday pattern

Staffing analysts expect weekends to be the lean shifts, and for total nursing hours that holds. For zero-RN-direct-care days specifically, the 2025 Q2 file shows the opposite: 6.51% of weekday resident-days recorded zero RN direct-care hours, against 4.26% on weekends.

The most likely explanation is a bookkeeping one rather than a clinical one, and it reinforces the direct-care-versus-administrative distinction. On weekdays, the registered nurse who is present is more often coded to administrative or director-of-nursing duties — the care-plan meetings, survey preparation, and family conferences that cluster Monday through Friday — leaving the direct-care column at zero even though an RN is in the building. On weekends, with the administrative load lighter, the one RN on duty is more likely to be coded to direct care. The pattern is a useful caution against reading any single PBJ column in isolation, and it is exactly the kind of measurement subtlety that a careful reproduction of these numbers has to account for.

Contract labor

Roughly 7.3% of all registered-nurse hours in the snapshot were supplied by contract or agency staff rather than direct facility employees. Agency reliance is a recognized marker of staffing instability — facilities that cannot retain permanent nurses lean on temporary ones — but it does not, on its own, drive the zero-RN-day count: a zero-RN day is zero whether the missing nurse would have been an employee or a contractor. We note the figure because it sits in the same file and rounds out the staffing picture, not because it explains the gap.

What this study does, and does not, show

This is a descriptive study built on a single quarter of a single federal file. Its limits are as important as its findings.

• It is not a finding of non-compliance. The 2025 Q2 data predates both the effective date of the federal 24-hour, 7-day RN requirement and the rescission of the staffing rule. A zero-RN day in this snapshot was not, in itself, a violation of federal law. These are measures of reported staffing, not legal judgments.
• It is one quarter. April–June 2025 is a single snapshot. Staffing varies seasonally and by quarter; a facility's pattern here may not be its pattern today. The figures will move as CMS publishes later quarters.
• It counts the direct-care RN column. The headline metric excludes RN administrative and director-of-nursing hours by design. The stricter all-RN-zero measure (0.56% of resident-days) is reported alongside it so readers can choose the lens that fits their question.
• It is not risk-adjusted. Zero-RN-day rates are raw frequencies. They are not adjusted for resident acuity, case mix, facility size, or local labor markets. State differences reflect those factors as well as staffing decisions, and the file cannot separate them.
• It does not measure resident outcomes. This study counts staffing inputs, not harm. It does not claim that any specific resident was injured on a zero-RN day, and it draws no causal line from a missing nurse to any outcome. Linking staffing to outcomes requires data this file does not contain.
• It depends on facility self-report. PBJ is payroll-derived and auditable, which makes it far stronger than the old survey self-report, but the hours are still submitted by the facilities themselves. Mis-coding a present RN into the wrong job column would shift these counts, as the weekday pattern above suggests can happen.
• No facility-level claims. State and national aggregates are research outputs. We attach nothing to any provider profile and rate, endorse, or warn against no individual home.

None of these limits change the central, reproducible fact: in the most recent published quarter, a registered nurse was absent from direct care on millions of resident-hours' worth of days, and that absence was concentrated, by facility and by state, in patterns too sharp to be accidental.

Why the timing matters

For most of the program's history, the federal requirement was simply that a registered nurse be on duty eight consecutive hours a day, seven days a week, with broader "sufficient" staffing left undefined. The 2024 rule was the first attempt to put a hard, around-the-clock floor under that language. Its rescission, effective February 2, 2026, returns the country to a regime with no enforceable national minimum on total or RN-specific staffing.

In that regime, transparency is the remaining accountability layer. If no rule requires a nurse, the public record of whether one was present becomes the only check a resident's family, a state regulator, a researcher, or a journalist can apply. That is the case for continuing to publish PBJ regardless of the regulatory weather: the file documents what staffing actually is, independent of what any rule does or does not require. The press attention to this question across health-policy outlets reflects the same logic — with the floor gone, the measurement is what is left.

We will append later quarters to this analysis rather than silently revise it, so the trend after February 2, 2026 can be read against this pre-rescission baseline.

The accountability gap, quantified

It is worth being concrete about what changes when a national floor disappears. Under the rescinded rule, a facility running zero-RN days routinely would eventually have faced a measurable standard it was failing and a federal mechanism to enforce it. After February 2, 2026, the same facility running the same way faces no national standard at all. The 51 facilities that reported no bedside registered nurse on every day of the quarter were not, in this snapshot, breaking a federal staffing law — because the around-the-clock requirement had not taken effect — and they will not be breaking one after the rescission either, because the requirement no longer exists to take effect.

That is the accountability gap this study is meant to size. The figures here — 5.86% of resident-days, 5,062 facilities, a 120-fold state spread, 51 facilities at zero every single day — are not the exception the system is working to eliminate. With the floor gone, they are the equilibrium the system now permits. The only force still pushing against that equilibrium is visibility: the existence of a public, daily, payroll-derived record that anyone can query, and the willingness of researchers, regulators, families, and the press to keep querying it. A number that is published and watched behaves differently from a number that is filed and forgotten. Keeping this one in view is the entire purpose of measuring it.

Medical review

Reviewed by Jennifer Montecillo, MD, medical reviewer. Non-practicing medical reviewer. This study reports staffing figures drawn from a public CMS payroll file; it does not assess, rate, or make clinical claims about any individual facility or resident, and it does not offer medical advice.

Reproduce this

Every figure above traces to the CMS Payroll-Based Journal 2025 Q2 snapshot, the immutable quarter CMS published for the period 2025-04-01 through 2025-06-30. The exact SQL that regenerates the headline counts ships with this study in the reproducibility block below, and the daily source file is documented on the PBJ staffing dataset page. The full methodology standard for our research desk is at /methodology, and the provenance posture for every source family we ingest is catalogued at /sources. For adjacent analysis built on the same nursing-home data graph, see our companion studies on the deficiency and harm-rate gap across states and on staffing after the federal mandate.

Further reading and primary sources

• CMS Payroll-Based Journal Daily Nurse Staffing — the source dataset, at data.cms.gov.
• CMS nursing-home provider data topics — facility-level extracts, at data.cms.gov.
• The 2024 minimum-staffing final rule (since rescinded) — the regulatory text, at federalregister.gov.
• Independent policy analysis of nursing-home staffing — KFF and MACPAC both track the staffing-rule debate and its evidence base.

The underlying CMS data is a U.S. Government Work in the public domain. Reuse of this study is permitted with attribution to Fonteum Research and a link back to this page.

-- Zero-RN days in US nursing homes (CMS PBJ, CY2025 Q2).
-- Snapshot: pbj_daily_nurse_staffing, snapshot_quarter = '2025Q2'
--           (the immutable CMS Payroll-Based Journal quarter covering
--            2025-04-01 .. 2025-06-30, downloaded from data.cms.gov).
--
-- Definitions (locked):
--   facility-day  := one nursing home on one calendar day.
--   included      := mds_census > 0  (residents present that day).
--   zero-RN day   := Hrs_RN = 0      (RN direct-care job code; this column
--                    EXCLUDES RN administrative and RN director-of-nursing
--                    hours, which PBJ records separately as Hrs_RNadmin and
--                    Hrs_RNDON).
--
-- Re-running this query against the 2025Q2 snapshot reproduces every headline
-- figure in the study, exactly.

-- ── Headline figures ─────────────────────────────────────────────────────────
with day_rn as (
  select
    provnum,
    state,
    work_date,
    coalesce(hrs_rn, 0) as rn_direct_hours
  from public.pbj_daily_nurse_staffing
  where snapshot_quarter = '2025Q2'
    and mds_census > 0
)
select
  count(*)                                                          as facility_days,        -- 1,322,254
  count(*) filter (where rn_direct_hours = 0)                       as zero_rn_days,          -- 77,542
  round(100.0 * count(*) filter (where rn_direct_hours = 0)
        / count(*), 2)                                              as pct_zero_rn_days,      -- 5.86
  count(distinct provnum)                                           as facilities,            -- 14,537
  count(distinct provnum) filter (where rn_direct_hours = 0)        as facilities_any_zero,   -- 5,062
  round(100.0 * count(distinct provnum) filter (where rn_direct_hours = 0)
        / count(distinct provnum), 1)                               as pct_facilities_any     -- 34.8
from day_rn;

-- ── State-level rate (facility-days with zero RN direct-care hours) ───────────
-- Louisiana 27.94% (highest) .. Rhode Island 0.23% (lowest) — a 121x range.
-- with day_rn as ( ... as above ... )
-- select state,
--        count(*)                                          as facility_days,
--        count(*) filter (where rn_direct_hours = 0)       as zero_rn_days,
--        round(100.0 * count(*) filter (where rn_direct_hours = 0)
--              / count(*), 2)                              as pct_zero_rn_days
-- from day_rn
-- group by state
-- having count(distinct provnum) >= 20
-- order by pct_zero_rn_days desc;

-- ── Per-facility distribution of zero-RN days across the 91-day quarter ───────
-- 9,475 facilities had 0 zero-RN days; 51 reported zero RN direct-care hours on
-- every single reported day.
-- with per_fac as (
--   select provnum, count(*) filter (where coalesce(hrs_rn,0) = 0) as zero_days
--   from public.pbj_daily_nurse_staffing
--   where snapshot_quarter = '2025Q2' and mds_census > 0
--   group by provnum
-- )
-- select
--   count(*) filter (where zero_days = 0)            as fac_none,        -- 9,475
--   count(*) filter (where zero_days between 1 and 6)  as fac_1_6,        -- 2,650
--   count(*) filter (where zero_days between 7 and 29) as fac_7_29,       -- 1,497
--   count(*) filter (where zero_days between 30 and 90) as fac_30_90,     -- 864
--   count(*) filter (where zero_days = 91)           as fac_every_day    -- 51
-- from per_fac;

Source: public.pbj_daily_nurse_staffing where snapshot_quarter = '2025Q2'
Facility-day included only when mds_census > 0 (residents present that day)
Zero-RN day := Hrs_RN = 0 — the RN direct-care job code, which excludes RN administrative (Hrs_RNadmin) and director-of-nursing (Hrs_RNDON) hours
Facility identity keyed on provnum (CMS Certification Number); state from the PBJ state field