Tests H9.1 (is the binding lever federal or state?) against the concrete case Phase 0 surfaced: the 2024 CMS staffing rule is dead (courts + Congress), so does state-level regulation turn out to be the more durable lever, as the finding in phase0-findings.md §2 speculated?
State staffing mandates: institutionally durable, enforcement is the weaker link
38 states + DC had numeric staffing standards where the old, pre-2024 federal requirement had none. Corrected on Phase 1 verification (2026-08-10): this section previously described the pre-2024 federal baseline as "the old federal floor (0.3 HPRD)." 42 CFR 483.35, read directly from eCFR in its restored post-repeal form, contains no numeric HPRD floor at all — it requires "sufficient nursing staff," licensed nurses on a 24-hour basis, and an RN "for at least 8 consecutive hours a day, 7 days a week." "0.3 HPRD" is a derived equivalence, not a codified floor, and "38 states exceeded it" is therefore a much weaker statement than it reads as. The like-for-like comparison is the one in the next sentence. Specifics: California 3.5 HPRD, New York 3.5 HPRD (plus a 2021 law requiring 40% of revenue go to resident-facing staffing), DC 4.1 HPRD (highest nationally), Florida 3.6 HPRD, Arkansas 3.36 HPRD. Nationally, though, most states still fall short of the rescinded federal 3.48 HPRD standard — 29 states require under 3.5 HPRD, 15 under 2.5. So state law is a genuinely uneven patchwork, not a uniform substitute for the dead federal floor.
Durability is real but not clean. No state staffing mandate has been struck down or repealed — a real contrast with the federal rule's fate. But New York's law shows statutory survival isn't the same as enforcement: regulators identified ~400 violating facilities with essentially no penalties for years, and a fourth industry lawsuit specifically targeting the law's funding/staffing provisions is ongoing, with meaningful fines ($2,000/day) only recently starting to reach ~20 facilities. The honest framing for the whitepaper: state law is the more durable statute, but enforcement capacity is a second, independent variable that has to be tracked state-by-state — H9.1 is supported on the "which level legislates" question and open on the "which level delivers" question.
No state legislation specifically reacting to the 2026 federal rescission is yet visible. The post-rescission window is only about six months and state legislative reporting lags; that is too short to read as evidence that states declined to respond. The state activity identified here (AZ, NY, RI, WA mandates, 2010–2023) predates rescission; the legislative response so far has been federal-level replacement bills (S.3886, H.R.8100) that have not passed.
Outcome evidence exists and is favorable to mandates on the point industry opposition usually raises. A Health Affairs 22-state panel (2010–2023) — State Nursing Home Minimum Staffing Mandates: Increased Staff Levels, Minimal Impact On Finances And Closures, 2010–23, published in the March 3, 2026 issue (this file previously dated it 2025; the 2025 in its DOI is a submission identifier) — found mandates raised direct-care staffing ~5% (CNAs +5.7%, LPNs +7.5%, no RN change), with labor costs rising less than revenue (margins statistically unchanged) and no effect on facility closure rates. That undercuts the closures argument for the range of state mandates studied (up to about 4.1 HPRD); it does not establish that every ratio mandate, including a stricter uniform national floor, has the same effect.
PE ownership and mortality: the anchor holds, with a material scope caveat
Anchor 10's underlying study is Gupta, Howell, Yannelis & Gupta, Review of Financial Studies 37(4) 2024 (not JPE, as the seed's phrasing had assumed venue-agnostic) — 4.2 million unique short-stay Medicare patients, 2005–2017, a within-facility instrumental-variables design built specifically to address two selection problems (which facilities PE targets, and which patients get matched to PE-owned homes). The causal (IV) estimate is +11% short-stay-plus-90-day mortality, notably larger than the OLS estimate (+0.3 pp, about 2% of the mean) — the authors' interpretation is that simple correlation understates the effect because PE homes selectively admit healthier patients, not that it overstates it. Also: +8% billed spending per stay, +6% for the stay plus 90 days; ~22,500 implied excess deaths and ~172,400 lost life-years over the sample.
Corrected at primary tier on Phase 1 verification (2026-08-10, verification-log). The paragraph above previously read ">7 million Medicare patients," "the naive OLS estimate (+10%)," and "+19% billed spending, ~50% higher antipsychotic use." The fetched full text of NBER WP 28474 (rev. Aug 2023 — the version published as RFS 37(4)) supports none of those. The antipsychotic figure is not in the paper at any vintage; its three patient-well-being measures are mobility (−6.2% OLS / −3% IV), ulcer development (+8.5% / 0%) and pain intensity (+10.5% / +8.3%). Scope, which this file did not previously state: the +11% is a local average treatment effect for compliers — patients who go to a PE facility because it is nearest. The authors' marginal-treatment-effects analysis recovers a similar average but "reveals substantial heterogeneity in treatment effects, including small beneficial effects for some patients," and they conclude that "PE has nuanced effects, with adverse outcomes for a subset of patients."
No direct econometric rebuttal was located. The material complication is scope-specific: a JAMA COVID-era comparison found PE facilities' pandemic mortality in line with the industry average. Its exclusion as inconclusive does not overturn the IV estimate, but it prevents a claim that the evidence is uncontested. Two systematic reviews (BMJ 2023, cross-national; a 2025 US-specific review) reinforce the general direction. Report the exact IV figures with medium-high confidence, not as a settled universal effect.
Effect on the protocol
§6/§9's framing shifts from "assess an emerging federal standard's design" to "the federal standard is dead; assess whether state law and PE-ownership accountability are adequate substitutes." The answer so far: state staffing law is a real, durable lever in states that have it, but coverage is uneven and enforcement lags the statute in at least one closely-watched case (NY); the PE-mortality evidence is strong enough to support facility-ownership disclosure or PE-specific oversight as a candidate architecture independent of the staffing-ratio question, since the mortality effect operates through mechanisms (staffing cuts, mobility declines) that ratio mandates only partially address.
Sources
Federal Register (Dec 3, 2025 repeal, eff. Feb 2, 2026); Center for Medicare Advocacy; Duane Morris legal alert; MACPAC/LegalClarity state-mandate summary; Health Affairs 2025.01223 (22-state panel); Center Square/McKnight's NY enforcement reporting; Congress.gov (S.3886, H.R.8100); Gupta/Howell/Yannelis/Gupta, Review of Financial Studies 37(4) 2024; NBER WP 28474; BMJ 2023 systematic review (via Healthcare Dive/Columbia Mailman); 2025 ScienceDirect US-specific systematic review; JAMA COVID-mortality comparison (via usrtk.org).
Confidence tiers
State mandate list and outcome evidence: medium-high (Health Affairs peer-reviewed panel; state-by-state HPRD figures are secondary/aggregator-sourced, not independently checked against each state's current code). Enforcement-gap finding: medium (single closely-documented case, NY — not confirmed as representative of all mandate states). PE-mortality finding: medium-high.