Two Greek figures review a nursing-home model, care equipment and staff figures.

Restore the Staffing Standard

In May 2024 the federal government set the first national minimum staffing levels for nursing homes. The 2025 reconciliation law barred their enforcement until September 2034, and the rule was repealed in February 2026, as the nursing home and hospital industries had asked. Democrats in Congress and advocates for residents want the standard back.

6 percent of nursing homes met all four staffing conditions when the rule was written ^[3]
1.2 million people live in certified nursing facilities ^[3]

The proposal

Every nursing home must again provide at least 3.48 nursing hours per resident per day and a registered nurse on site around the clock, on the original phase-in.

The debate

Advocates for residents argue that understaffed homes leave people waiting for hours, and that more staff means fewer deaths and less neglect. Nursing homes and hospitals answer that the workers do not exist and that rural homes would have to close beds. Staffing, they say, should follow clinical judgment rather than a fixed number. The pause passed in July 2025 as part of the reconciliation law, with Republican votes only.

“It is irrefutable that nursing homes with more staff have fewer resident deaths and negative health outcomes than their counterparts.”

argues for: Samuel Brooks, National Consumer Voice for Quality Long-Term Care Comment to the agency, 27 January 2026
“The AHA has repeatedly raised concerns that the requirements could exacerbate workforce shortages, lead to facility closures and jeopardize access to care, especially in rural and underserved communities that often do not have the workforce levels to support these requirements.”

argues against: Stacey Hughes, American Hospital Association Statement, 2 December 2025

The analysis

Restore the Staffing Standard

41% of the weight is for, 59% against.

An argument's weight combines three questions: how big the effect is, how certain it is and how much the good at stake matters. The larger the slice, the more the argument weighs.

For

Against

Restoring the staffing standard is probably a poor proposal on the current analysis, although it would buy something real.

Likely worse

Likely worse: Under the assumptions used, the proposal brings more than it costs in 14 of 100 model runs.

Overall effect: medium

How strongly the proposal works overall, in five steps from very small to very large, comparable across all debates. The ring is always full, even when the overall effect is small; this step says how large it is.

Like the opponents, the analysis judges against the standard, though less over the worker shortage they stress than over the wage bill. Paying the extra staff costs more than they achieve, even though the gains are real: residents live longer and are cared for better each day. The shortage still matters, because half the new staff would come from hospitals and home care, whose patients lose what residents gain; closed beds weigh least. The result rests on valuing a year of life for a very old, frail resident at half the general rate. At the full rate the two sides come close to cancelling.

In detail

All arguments

Arguments for3 arguments

Residents live longer with more nurses

7.9points

The rule would add roughly 102,000 nurses and aides across 15,000 facilities, about a tenth more care time per resident. What happens when nursing home staffing moves has been measured under a natural experiment in Denmark. Residents died sooner when nurses left.

Value 10 · LifeImpact 1.4Plausibility 5.5

▸ Show reasoning & sources ▾ Hide reasoning & sources

The stream is time alive for people who are near the end of it. This site weights human life at the top of its scale and does not lower that weight because the people are old; what age changes is how much life is at stake in each case, and that belongs to the Impact. The residents are mostly over eighty-five, many with dementia, and the difference an additional nurse makes is not dramatic medicine — it is someone noticing a change in breathing, a fall, a pressure sore before it becomes sepsis. What is priced here is only the extra time alive; comfort and dignity are a separate argument. The grief of the families is included at a tenth of the residents' own gain, the allowance this site books alongside every stream of deaths. The value is the highest the scale allows, because the stream is time alive and nothing else is folded into it.

About 1.2 million people live in certified nursing facilities, and 102,000 additional nurses and aides would raise care time per resident by roughly a tenth ^[3]^[4]. What that does to survival was measured in Denmark, where a parental-leave programme unexpectedly pulled a tenth of nursing home nurses out of work: mortality among residents aged 85 and over rose 13 percent and stayed up ^[5]. Read as a rate of exchange, each percent of staffing is worth 1.3 percent of mortality. That figure comes from a sudden loss of experienced staff, which is more disruptive than a gradual addition, so 0.6 is used here instead, in a range from 0.15 to 1.3. A tenth more staffing then lowers mortality by 6 percent. Against roughly 360,000 resident deaths a year that is 21,600 deaths postponed, and postponed is the right word: the people concerned have on average about eighteen months of life ahead, so the gain is about 32,000 years of life a year, in a range from 11,000 to 65,000; the families' share adds a tenth on top. The Impact is the largest gain in this debate and still a modest one, because what is bought is months rather than years, for people who have few of either.

▸ Show calculation ▾ Hide calculation

	Deaths among nursing home residents each year [3]	1.2 million residents, roughly three in ten of whom die in a year	360,000 deaths a year
×	Share postponed by a tenth more staffing Setting, range 0.15 to 1.3: a tenth fewer nurses raised mortality 13 percent in the Danish measurement; a sudden loss disrupts more than a gradual addition, and the study measured nurses where this rule mostly adds aides, so under half of that rate is used [5]	10 % staffing × 0.6 = 6 %	21,600 deaths a year
×	Years of life behind each postponed death Setting, range 0.5 to 3 years: residents at this stage have about eighteen months of life ahead on average [3]	1.5 years	32,400 years of life
×	Value of the years of life a year in full health counts 40,000 euro-equivalent on this site; a year at this stage is counted as half of one, for age and frailty, which is the adjustment and not a different rate	20,000 euro-equivalent each	648 million euro-equivalent
+	Grief of the families the allowance this site books alongside every stream of deaths: a tenth of the loss to the person who dies	10 % of 648 million euro-equivalent	713 million euro-equivalent
÷	Normalised Impact scale of this evaluation	500 million euro-equivalent a point	1.43

Score 1.43 Impact × 10 Value × 5.5 Plausibility ÷ 10 = 7.9 of 100

The Danish study is the cleanest evidence anyone has on this question. A parental-leave programme opened in 1994 and was taken up heavily by nurses, which cut nursing home nurse employment by a tenth for reasons that had nothing to do with any resident's health ^[5]. The counterfactual is the same facilities before the programme and comparable facilities with fewer eligible nurses, and the confounder that would normally ruin such a comparison — that badly run homes both lose staff and lose residents — is removed, because who took leave depended on having a young child rather than on the state of the home. Reverse causation cannot arise for the same reason. Three things stand between that finding and this one. The Danish measurement concerns qualified nurses, whom the study found could not be replaced, while three quarters of the hires this rule requires are aides, whom facilities in that same shortage did replace — so the hour that is added here is not the hour that was measured. Denmark's nursing homes were far better staffed than American ones to begin with, which argues that the American return should if anything be larger, and the study measured staff leaving rather than staff arriving, which argues the other way because a sudden loss disrupts more than a slow gain does. The last two are why the rate of exchange used here is under half of what was measured; the first is a matter of whether the measurement carries at all, and it holds the plausibility half a point lower. The Plausibility is a little above the middle: the design is as clean as this question allows, and the hour it measured — a qualified nurse's — is not the hour most of this rule adds.

Evidence: Study · plausibility at most 8 Evidence of effect: Quasi-experimental · at most 8

Less of what understaffing does daily

3.9points

Most of what a staffing floor changes never appears in a mortality statistic. It is a resident who is turned before a pressure sore forms, helped to eat before the tray goes cold, taken to a toilet rather than left. Across 1.2 million people, small daily differences add up.

Value 9 · HealthImpact 1.0Plausibility 4.5

▸ Show reasoning & sources ▾ Hide reasoning & sources

The stream is the quality of the days themselves: pain that is noticed, hygiene that is maintained, a person who is spoken to. It belongs to the class this site uses for life and health, one step below the top of it because what is lost here is recoverable while a life is not. Dignity is not priced separately from health; the two arrive together in a nursing home and separating them would invent a distinction the evidence cannot carry. The extra time alive is counted in the argument above and is not repeated here. Nothing about how many people are affected enters at this point. The value sits one step below the maximum: the stream is health and dignity together, and both can be regained.

About 1.2 million people live in these facilities, and each of them would receive roughly a tenth more care time ^[3]. What that time buys, short of survival, is the ordinary business of a nursing home: turning, feeding, toileting, noticing. A gain of 0.02 resident-years of improvement per resident per year is used here — about a week of the difference between a bad day and an ordinary one, spread across the year — in a range from 0.005 to 0.05. Across 1.2 million residents that is 24,000 resident-years a year moved from a bad day to an ordinary one. What such a move is worth is the gap in health between the two: an ordinary day at this stage counts as half a day in full health, the same weight the survival argument above uses for these residents, and a bad day, in pain or unattended, is set at close to nothing. The gap is therefore half a year in full health per resident-year, a setting rather than a measurement, which gives 12,000 years in full health. What the figure does not include is any effect on residents' families, who visit less anxiously when a home is properly staffed, because nothing measures that. The Impact is smaller than the survival gain but of the same order, which is what one would expect: more hands change many days a little and a few days entirely.

▸ Show calculation ▾ Hide calculation

	People living in certified nursing facilities [3]		1.2 million residents
×	Resident-years of improvement per resident Setting, range 0.005 to 0.05: about a week of the difference between a bad day and an ordinary one, spread across a year	0.02 a year	24,000 resident-years
×	Value of the move from a bad to an ordinary day Setting: an ordinary day at this stage counts as half a day in full health, the weight the survival argument above uses for these residents, and a bad day as close to nothing; the gap is half a year in full health, at 40,000 euro-equivalent a year	20,000 euro-equivalent each	480 million euro-equivalent
÷	Normalised Impact scale of this evaluation	500 million euro-equivalent a point	0.96

Score 0.96 Impact × 9 Value × 4.5 Plausibility ÷ 10 = 3.9 of 100

The direction of this claim is not seriously disputed by anyone, including the industry, whose objection is to the cost rather than to the benefit. What is missing is a measurement. Dozens of studies find that better-staffed homes have fewer pressure ulcers, fewer falls and less use of restraints, but almost all of them compare homes with each other at a point in time, and homes that staff well differ in a dozen other ways — ownership, payer mix, local labour markets — any of which could produce the same pattern without staffing causing it. The Danish study, which does have a clean comparison, measured care delivery as well as mortality and found both moving together, which supports the chain without sizing this end of it ^[5]. The counterfactual for that study is the same facilities before the leave programme; the confounder of badly run homes is removed by the fact that take-up depended on having a young child; reverse causation cannot arise. But nobody has measured what a tenth more staffing does to a resident's quality of life, and the figure used here is built rather than found. The Plausibility is below the middle, not because anything argues against the claim, but because the chain from more hands to better days is nowhere measured at its own end.

Evidence: Mechanism · plausibility at most 6 Evidence of effect: Mechanistic · at most 6 Placement: Chain closed, unevidenced · plausibility 4–5

Open: Nebraska and other states collect resident-reported quality measures alongside payroll-based staffing data. A comparison of homes crossing the threshold against those already above it would replace the setting with a measurement and could carry P to 6.

Fewer trips to hospital that need not happen

0.39points

A quarter of nursing home residents are admitted to hospital each year, and about half of those admissions are considered avoidable with earlier attention. Better staffing catches the infection before the ambulance. Medicare pays for the admissions either way.

Value 5 · Public financesImpact 0.2Plausibility 3.5

▸ Show reasoning & sources ▾ Hide reasoning & sources

The stream is money the public payers do not spend, priced at the middle of the scale like every other dollar on this site. It is a genuine saving rather than a transfer, because the admission does not happen at all: the ambulance is not called, the bed is not occupied, the resources go elsewhere. The harm the resident avoids by not being moved is not counted here — that sits in the quality argument above, and booking it in both places would price the same fact twice. Whether Medicare or a state Medicaid programme carries the bill makes no difference to the weight. The value is the middle of the scale, the level this site uses for public money whatever it is spent on.

About a quarter of the 1.2 million residents are admitted to hospital in a year, and roughly half of those admissions are the kind that earlier attention would have prevented — a urinary infection, dehydration, a fall that was waiting to happen ^[3]. That is about 150,000 avoidable admissions a year. The same rate of exchange used for survival applies here: a tenth more staffing, at the rate of 0.6, removes 6 percent of them, or 9,000 admissions. An avoidable admission from a nursing home costs the public payers roughly 14,000 dollars, counting the stay and the transport. The result is about 125 million dollars a year, in a range from 30 to 270 million. Neither the resident's own experience of the admission nor the risk it carries is counted here. The Impact is the smallest gain in this debate: the admissions are expensive but the share the staffing floor removes is small.

▸ Show calculation ▾ Hide calculation

	Avoidable hospital admissions from nursing homes each year [3]	1.2 million residents × a quarter admitted × half avoidable	150,000 admissions a year
×	Share removed by a tenth more staffing the same rate of exchange as the survival argument, taken from the Danish measurement of readmissions [5]	10 % staffing × 0.6 = 6 %	9,000 admissions a year
×	Cost to the public payers Setting, range 8,000 to 18,000 euro-equivalent (about 9,300 to 21,000 dollars): the stay and the transport, converted at 1.16 dollars to the euro	12,000 euro-equivalent each (about 14,000 dollars)	108 million euro-equivalent
÷	Normalised Impact scale of this evaluation	500 million euro-equivalent a point	0.22

Score 0.22 Impact × 5 Value × 3.5 Plausibility ÷ 10 = 0.4 of 100

The chain has three links and only the middle one is measured. That understaffed homes send more residents to hospital is well documented in comparisons between homes, but those comparisons cannot separate staffing from everything else that differs between a good home and a bad one. The Danish study, which can, measured a large rise in hospital readmissions when nurses were pulled out, which supports the direction under a clean comparison and is the reason the same rate of exchange is used here as for survival ^[5]. The counterfactual there is the same facilities before the leave programme, the confounder of badly run homes is removed by the eligibility rule, and reverse causation cannot arise. What is not measured is the share of American nursing home admissions that additional staffing would actually prevent, nor the cost of one. The counter-argument that better-staffed homes send residents to hospital more often, because someone notices sooner, is real and pulls against the claim; it is why the figure is not set higher. The Plausibility is low because the size of this effect has not been measured at either end of the chain and the counter-mechanism is unanswered.

Evidence: Mechanism · plausibility at most 6 Evidence of effect: Mechanistic · at most 6 Placement: Chain open · plausibility 3–3.5

Open: Payroll-based staffing data can be matched to Medicare claims at facility level; a comparison of homes crossing the threshold against those already above it would settle both the share and the direction, and could carry P to 6.

Arguments against3 arguments

The wage bill for 102,000 more staff

13points

The agency put the cost of its own rule at 43 billion dollars over ten years, and the repeal at 5.5 billion a year saved. Nursing homes take about three fifths of their revenue from Medicaid and Medicare, so most of the bill lands on public budgets. Half the staff are assumed to be new to care work, and their hours are the cost counted here; the other half move from hospitals and home care, and what their hours produced there is the argument after this one.

Value 5 · Public financesImpact 4.7Plausibility 5.5

▸ Show reasoning & sources ▾ Hide reasoning & sources

The stream is money, priced at the middle of the scale. For the staff who are new to care work it is a real cost and not a transfer: they give up hours they would otherwise have spent in another job or outside paid work, and the wage is the price of those hours. For the staff who move across from hospitals and home care the wage is not counted here; what their hours produced there is the argument after this one. Public and private money carry the same weight here, which matters because about three fifths of the bill reaches the facilities through Medicaid and Medicare and the rest through private fees. That the workers are themselves low-paid does not turn the wage into a gain — they are paid roughly what their time is worth elsewhere, and the difference is too small to book. The value is the middle of the scale, the level this site uses for money whatever its source.

The agency's own analysis of the 2024 rule put the cost at 43 billion dollars over ten years, and the repeal rule that reversed it in February 2026 restated the figure as about 5.5 billion dollars a year once the phase-in is complete ^[1]^[2]. The industry's estimate, built from the 102,000 hires the rule requires at prevailing wages, is 6.5 billion ^[4]. The lower official figure is the wage bill: 5.5 billion dollars, in a range from 4.8 to 6.5 billion. Not all of it is a cost to anyone but the payer. Half of the new staff, in a range from a quarter to three quarters, are assumed to be new to care work, and their hours are a real resource that the wage measures: about 2.75 billion dollars. The other half move across from hospitals and home care; the wage they are paid is a transfer between payers and workers whose dollar carries the same weight, and what their hours produced where they came from is counted, in health rather than in money, in the argument after this one. The dollar carries the standard weight of one whether it comes from a state Medicaid budget, from Medicare or from a resident's family, because all three sit at or near the middle of the income distribution once the mix is taken together. The Impact is by a wide margin the largest figure in this debate, and it is about twice the value of everything the staffing floor delivers.

▸ Show calculation ▾ Hide calculation

	Additional nurses and aides the rule requires [4]	77,000 aides and 24,000 registered nurses	102,000 staff
=	Annual wage bill, as the agency scores it the agency's own analysis of the 2024 rule; the repeal rule restated it as an annual saving [1][2]	5.5 billion dollars, against the industry's 6.5 billion	5.5 billion dollars
÷	In euro-equivalent Range 4.1 to 5.6 billion euro-equivalent (4.8 to 6.5 billion dollars): the agency's figure at the bottom, the industry's 6.5 billion dollars at the top	1.16 dollars to the euro	4.74 billion euro-equivalent
×	Share of the hires who are new to care work Setting, range 25 to 75 percent, shared with the argument after this one: only the hours of new entrants are a resource the wage measures; the wage of those who move from hospitals and home care is a transfer, and what their hours produced there is counted in health in that argument [4]	50 %	2.37 billion euro-equivalent
×	Weight of a dollar across the payers about three fifths reaches facilities through Medicaid and Medicare and the rest through private fees; the mix sits at the middle of the income distribution	1.0	2.37 billion euro-equivalent
÷	Normalised Impact scale of this evaluation	500 million euro-equivalent a point	4.74

Score 4.74 Impact × 5 Value × 5.5 Plausibility ÷ 10 = 13 of 100

The number of additional staff follows from arithmetic rather than from behaviour: the rule names a minimum in hours per resident per day, payroll data show what each facility actually staffs, and the gap between the two is measured rather than predicted ^[3]. The counterfactual is current staffing as recorded in the payroll-based system, which every certified facility must report. What is projected is the wage at which those hours are bought, and here two independent estimates — the agency's and the industry's — differ by about a fifth; what is set rather than measured is how many of the hires are new to care work, which the argument after this one shares and which nobody has looked at. The one genuine uncertainty is behavioural and runs against the figure: a facility that cannot hire may reduce its resident count instead of raising its wage bill, in which case the money is not spent and the benefit is not delivered either. That response is booked as its own contra argument rather than discounted here. What the figure is not is certain: nobody pays this bill unless the facilities hire, which is the very condition the benefits above hang on, so the cost enters or stays away together with them rather than standing as a sure loss against an uncertain gain. The Plausibility is at the top of what a cost projection can carry: the staffing gap is measured, and only the wage at which it is filled is estimated.

Evidence: Projection · plausibility at most 6 Evidence of effect: Mechanistic · at most 6

Hospitals and home care lose the same nurses

3.3points

The rule does not create 102,000 nurses and aides; it requires nursing homes to bid for them. Where they come from decides whether the country gains care or moves it. Hospitals, home care and assisted living are short of exactly the same people.

Value 10 · HealthImpact 0.9Plausibility 3.5

▸ Show reasoning & sources ▾ Hide reasoning & sources

The stream is the same one the pro side counts, in the opposite direction and on other people: survival and health for patients in hospitals, in home care and in assisted living, where the nurses and aides would otherwise have worked. It carries the same weight as the gain it offsets, because a life is a life wherever it is lived. The patients are on average younger and less frail than nursing home residents, which affects how much is at stake in each case rather than what a case is worth, so that difference belongs to the Impact. The wages the movers are paid are not counted as a cost anywhere: they are a transfer, and this argument is what their hours cost where they came from. The value is the highest the scale allows, because the stream is the same life and health that the argument it offsets counts.

The rule requires 102,000 additional nurses and aides at a moment when every part of American health and social care is short of them ^[4]. Some will be people entering or re-entering the workforce, drawn by the wages the rule forces up; some will simply move across from a hospital, a home care agency or an assisted living facility. Half is used here for the share who move rather than join, in a range from a quarter to three quarters — the wide range reflects that nobody has looked. Where those workers came from, patients lose the same kind of attention that nursing home residents gain, though at a somewhat lower rate because hospital and home care patients are on average less frail: 80 percent of the nursing home return is used, in a range from 50 to 100 percent. The offset is therefore about 40 percent of the survival and quality gains counted above, or 0.48 against their 1.19. The wage those movers are paid is not counted again in the argument before this one: it is a transfer, and this is what it buys. The Impact is a substantial fraction of the benefit it offsets, which is the point: a staffing rule for one setting is a staffing cut for another unless the workers are new.

▸ Show calculation ▾ Hide calculation

	Survival and quality gained in nursing homes [5]	713 plus 480 million euro-equivalent from the two arguments above	1,193 million euro-equivalent
×	Share of the new staff who move rather than join Setting, range 25 to 75 percent: nobody has measured where staff hired under a mandate come from [4]	50 %	597 million euro-equivalent
×	Return to the same hours where they came from Setting, range 50 to 100 percent: hospital and home care patients are on average less frail than nursing home residents, so the same hour is worth somewhat less there	80 %	477 million euro-equivalent
÷	Normalised Impact scale of this evaluation	500 million euro-equivalent a point	0.95

Score 0.95 Impact × 10 Value × 3.5 Plausibility ÷ 10 = 3.3 of 100

The mechanism is not in dispute — a binding hiring requirement in a tight labour market bids workers away from somewhere — and the counterfactual is the same labour market without the rule. What is entirely unmeasured is the load-bearing number: what share of the 102,000 would be new entrants rather than movers. No study has followed workers across settings after a staffing mandate, and the one American precedent, California's hospital nurse ratio law of 2004, is not close enough in setting to carry the number here. The confounder that would matter is the wage response: if the rule raises wages enough, it draws people into care work who were not in it before, and the offset shrinks. That link is exactly the one that has not been checked, which is what holds this argument where it is rather than anything measured against it. Reverse causation does not arise; the rule precedes the hiring. The Plausibility is low because the number this argument turns on has never been looked for, not because anything has been found against it.

Evidence: Mechanism · plausibility at most 6 Evidence of effect: Mechanistic · at most 6 Placement: Chain open · plausibility 3–3.5

Open: State-level payroll data covering hospitals, home care and nursing homes would show where staff moved after Nebraska and other early adopters raised their floors, and could carry P to 6 or dissolve the argument.

Beds close and families take over

1.6points

A facility that cannot hire has one other way to meet an hours-per-resident floor, which is to have fewer residents. Rural homes with no local labour market to draw on are the ones with the least room to manoeuvre. What the beds carried does not disappear; it moves to families.

Value 9 · Life timeImpact 0.5Plausibility 3.5

▸ Show reasoning & sources ▾ Hide reasoning & sources

The stream is unpaid hours: a daughter who cuts her working week, a spouse in his eighties who lifts someone he cannot lift. This site treats such time as part of a person's life rather than as an inconvenience, which places it near the top of the scale. It is not freely chosen time; nobody takes on the daily care of a parent with dementia as a hobby. What an hour is worth and how many there are are separate questions, both set out in the derivation. The harm to the resident who is moved is mentioned but not priced separately, because the number is small beside the caregiving hours. The value is high because the stream is hours of life spent under compulsion, not a convenience that is lost.

The requirement is stated as hours per resident per day, which means a facility that cannot hire can comply by reducing its resident count instead. Six percent of nursing homes met all four conditions when the rule was written, and rural facilities, which have the smallest labour markets to draw on, were furthest from them ^[3]. Three percent of beds closing is used here, in a range from one to eight percent — about 36,000 residents. Some will find another facility; some will be cared for at home. Half is used for the share who move to family care, at 20 hours a week each: 18,000 people times 20 hours times 52 weeks is 18.7 million hours a year. A further 19 million hours is allowed for families whose relative moved to a facility further away, and for the search itself, giving about 38 million hours. Moving a frail resident carries its own mortality risk of one to three percent, which would add a few hundred deaths; that is left out because it is small beside the hours and because it is partly avoided by residents who never enter care at all. The Impact is small but not trivial: it is roughly a fifth of the survival gain the rule is meant to produce, landing on different people.

▸ Show calculation ▾ Hide calculation

	Nursing home beds [3]	1.2 million residents	1.2 million beds
×	Share closing because the facility cannot hire Setting, range 1 to 8 percent: six percent of homes met all four conditions when the rule was written, and rural facilities were furthest from them [3]	3 %	36,000 residents
×	Share cared for at home instead Setting, range 30 to 70 percent: the rest find another facility	50 % × 20 hours a week × 52 weeks	18.7 million hours a year
+	Hours of families whose relative moved further away, and of the search itself Setting, range 5 to 40 million hours	19.3 million hours	38 million hours a year
×	Value of forced time the rate this site uses for time a person must spend with nothing in return	6.85 euro-equivalent an hour (about 7.90 dollars)	260 million euro-equivalent
÷	Normalised Impact scale of this evaluation	500 million euro-equivalent a point	0.52

Score 0.52 Impact × 9 Value × 3.5 Plausibility ÷ 10 = 1.6 of 100

The mechanism is a matter of arithmetic — an hours-per-resident floor can be met by lowering the denominator — and the counterfactual is the same facilities without a floor. Whether facilities actually respond that way is the open question. The precedent that comes closest is California's hospital nurse ratio law, after which hospitals did not close in numbers, but hospitals and rural nursing homes are not comparable in their access to capital or staff. No source carries the closure rate, and the three percent used here is a construction. Two counter-mechanisms pull against the argument and only one is answered: the rule's own phase-in gave rural facilities extra years, which the four-year horizon here already reflects, but the possibility that states would raise Medicaid rates rather than let beds close is real and unresolved. Reverse causation does not arise, since the floor precedes the closures. The Plausibility is low because the closure rate this argument turns on has no source behind it and one counter-mechanism is unanswered.

Evidence: Mechanism · plausibility at most 6 Evidence of effect: Mechanistic · at most 6 Placement: Chain open · plausibility 3–3.5

Open: Certified bed counts by county are published quarterly. Tracking them in states that raise their own floors, against states that do not, would put a number on the closure rate and could carry P to 6.

How robust the result is

Under the assumptions used, the proposal brings more than it costs in 14 of 100 model runs.

gains more than it costscosts more than it gains

Each dot stands for 200 of the 20,000 model runs.

A year of life for a nursing home resident is valued at half the general rate used on this site, because the residents are very old and often frail; that one setting carries the balance. With it, the proposal comes out ahead in 14 of 100 runs. At the full rate the two sides would come close to cancelling; at a lower rate the case against would grow. The figures on both sides are the annual amounts at full compliance ^[1]^[3]. Hiring builds up over the first two years, so both sides are overstated early by about the same share, and the balance between them does not depend on the timetable.

Show the distribution of the model runs

20,000 simulated futures

Likely worse · 14 %

In 14 of 100 simulated futures the proposal brings more than it costs.

No effect is known exactly; every argument has a range. The calculation draws a possible size for each argument from its range 20,000 times and counts how often the benefits exceed the costs.

Net effect in points per future. Left of zero the drawbacks outweigh the benefits, right of zero the benefits win. Eight in ten futures fall between −12 and +2.7 points.

Scale and assessment

Scale of this evaluation: Normalised Impact, unitless and calibrated to this topic. For comparison, one point here corresponds to about 500 million euro-equivalent a year.

Size class:medium

Effect medium (−3) · 40 % of the effect is positive

Areas show individual argument weights. The positive share of the effect nets out money transfers between the two sides; the probability of a positive net effect is determined separately.

How we score →

Review and sources

AI evaluation · not yet reviewed by a human

This evaluation was produced and sourced by an AI model; a human review is still pending. Figures and conclusions may still change. The review log is at the foot of the page.How review works →

Last reviewed by Claude Opus 5.5 · October 29, 2026 · 7× AI, not yet reviewed by a human

October 29, 2026AI reviewClaude Opus 5.5record updated
Page texts for the new page layout (draft G) with quotes from primary sources; numbers unchanged.
October 21, 2026AI reviewClaude Fable 5.1record updated
Money amounts in argument texts stated in dollars per style-en.md; scores, chains and balance unchanged
October 20, 2026AI reviewClaude Fable 5.1record updated
Summary and scale note rewritten to the style of 20 Oct 2026; numbers unchanged
October 14, 2026AI reviewClaude Opus 5.5record updated
Prüfbericht 11.10. (H10): Halbierung bei pro-2 als Abstand zwischen schlechtem und gewöhnlichem Tag begründet statt „years of the same lives“; Staffelung des Neustarts 2027 im Maßnahmentext offengelegt; Zahlen unverändert.
September 14, 2026AI reviewClaude Opus 5re-scored
Gesamtprüfung 08.09.: Elastizität 6 % statt 7,2 %, Angehörigen-Zuschlag nachgetragen, P pro-1 6→5,5 (Nurses vs. Aides), Doppelzählung Lohn/abgeworbene Pflege aufgelöst (con-1 i 4,74→2,37); r 0,30 → 0,40.
September 6, 2026AI reviewClaude Opus 5record updated
i_spanne und normalisierung an allen 6 (nur globale Anker, kein Transfer). Kernpunkt: con-1 (Lohnsumme) war definitorisch gebucht, faellt aber nur an, wenn die Heime einstellen — jetzt mechanistisch und in der Eintrittsgruppe mit der Nutzenseite, P 5,5 unter dem Deckel 6,0. Allein das hebt P(D>0) von 0,00 auf 0,35, Kategorie Schlechter → Ausgeglichen. Zwei offene Punkte: con-1 und con-2 messen dieselben Opportunitaetskosten zweimal, und pro-2 zieht wegen kette_geschlossen trotz Gruppe immer ein — mit der vorgeschlagenen Vererbungsregel laege der Record bei 0,18 statt 0,35.
September 6, 2026AI reviewClaude Opus 5First evaluation
Created for the English side: staffing-to-mortality rate taken from the Danish parental-leave experiment at under half its measured value.

Evaluations are produced with AI support and reviewed on a schedule for new developments; human passes are marked separately.How we review →

Show all 5 sourcesHide sources

Federal Register: Medicare and Medicaid Programs; Minimum Staffing Standards for Long-Term Care Facilities and Medicaid Institutional Payment Transparency Reporting. federalregister.gov
Federal Register: Medicare and Medicaid Programs; Repeal of Minimum Staffing Standards for Long-Term Care Facilities. federalregister.gov
KFF: A Closer Look at the Final Nursing Facility Rule and Which Facilities Might Meet New Staffing Requirements. kff.org
American Health Care Association: Staffing Mandate Analysis. ahcancal.org
Friedrich and Hackmann, Review of Economic Studies 88(5): The Returns to Nursing: Evidence from a Parental-Leave Program. academic.oup.com

More debates in this area

Very likely better · 99 %

Size: large

Restore the Enhanced Credits

Bring back the larger Marketplace premium credits that lapsed at the end of 2025, for three more years.

Unclear · 49 %

Size: large

Medicaid Work Requirements

From January 2027, adults in the Medicaid expansion group must show 80 hours a month of work, study or volunteering, half-time schooling, or earnings of at least 580 dollars a month to keep their coverage.

Likely better · 88 %

Size: very large

Medicare For All

Replace private health insurance with a single federal programme covering everyone for everything, with no premiums, deductibles or co-payments at the point of care.

All debates in Health Coverage and Costs →