State variation in staffing and characteristics of nursing homes most impacted by new federal standards

Abstract

Introduction

New federal minimum staffing standards aim to increase staffing in nursing homes. As the future of the new rule remains uncertain, this study compares the new standard to existing state staffing standards and examines the characteristics of nursing homes who may meet the new requirement.

Methods

We utilized the 2023 CMS Payroll-Based Journal Data to calculate staffing ratios and nurse turnover. We summarized staffing levels by nursing home size, ownership type, chain affiliation, and rurality. We compared state staffing policies using data from the Medicaid and Children's Health Insurance Program (CHIP) Payment and Access Commission (MACPAC). Statistical analyses included Student's t-tests and z-tests.

Results

Nursing homes in 11 states had staffing levels below the new federal minimum. Across states, compliance with the new federal requirements ranged from 31% in states with the lowest staffing requirements to 65% in states with the highest. Higher state staffing standards were associated with better quality ratings and lower turnover.

Conclusion

Our paper explores state efforts toward reaching staffing goals. While higher state staffing standards correlate with better nursing home quality ratings, many homes may struggle to meet the new federal requirements.

Introduction

In 2024, the Centers for Medicare and Medicaid Services (CMS) established the first federal minimum staffing ratios in nursing homes (NHs).¹ States have widely varying policies on resident-staffing, but the new national minimums strengthen the requirements set by the Nursing Home Reform Act. We found that average reported staffing hours per resident day (HPRD) and states' minimum staffing requirements varied substantially (Figure 1). In 2023, the average reported staffing HPRD was below the new CMS minimum threshold of 3.48 nurse HPRD in 11 states. Additionally, most existing state minimum staffing requirements are far below the new federal standard, with some state regulations specifying no specific staff-resident ratio while other states require staffing levels well above the new CMS minimums.² Only 6 states (CA, FL, IL, MA, NY, RI) and Washington, D.C. have mandates for nurse-resident-staffing ratios above 3.48 HPRD.²

Figure 1.

Nursing hours per resident day by state, existing state standards, and the new CMS requirement. Source: State policies were retrieved from the Medicaid and CHIP Payment and Access Commission (MACPAC). Nurse hours-per-resident day were calculated from the Centers for Medicare & Medicaid Services Payroll-Based Journal nurse staffing and resident census data. Direct care staffing in NHs is provided by certified nursing assistants (CNAs), licensed practical nurses (LPNs), and registered nurses (RNs). The analysis includes 14 900 nursing homes present in at least one quarter of 2023 PBJ data.

Low staffing in NHs has been a persistent problem for decades. The new CMS standards were shaped by years of research demonstrating significant associations between direct care staffing, resident health, and care quality outcomes.^3-7 Little is known about how the new standard will impact NHs across states.

Previous CMS standards did not consider resident census, resident complexity, or local market conditions, requiring NHs only to ensure 24-hour care by either an registered nurse (RN) or licensed practical nurse (LPN) and at least 1 RN for 8 consecutive hours each day.⁸ The new CMS minimum staffing standards will require NHs to provide 3.48 HPRD of total nurse staffing, 0.55 HPRD of RN staffing, and 2.45 HPRD of certified nurse aide (CNA) staffing. Unlike many state mandates, LPNs are excluded from the federal staffing standard, which may lead to NHs lowering LPN staffing levels. Additionally, many NHs may struggle to meet these standards due to direct care worker shortages and recruitment challenges.⁹ For financially struggling facilities, often with a higher proportion of Medicaid residents, the new standards could worsen existing challenges with staff hiring. CMS may take enforcement actions if a NH fails to meet the new federal standards such as the levying of civil money penalties, termination of the provider agreement between CMS and the NH, and denial of payment for new admissions.¹⁰ States may enforce their own staffing standards with their own penalties; for instance, in Illinois, noncompliance is met with a financial penalty tied to the cost of wages for the missing nurse hours.¹¹ Finally, we note that many states do not set staffing mandates or set mandates which match federal requirements.

To understand which types of NHs are most at risk of facing staffing gaps and penalties under the new rule, this study examines state and federal staffing requirements and facility characteristics. Using data from CMS and the Medicaid and Children’s Health Insurance Program (CHIP) Payment and Access Commission (MACPAC), we documented variation in staffing levels across NH ownership type and size.²

Study data and methods

We calculated resident-nurse staffing ratios using the 2023 CMS Payroll-Based Journal (PBJ) data, which provides comprehensive daily staffing levels by staff type and resident census levels.¹² Average annual nurse staffing levels of NHs were divided by the average daily resident census to determine the reported HPRD. We calculated the average reported daily nurse HPRD for all NHs reported in the PBJ in 2023 (14 900 NHs, excluding 6 NHs in Puerto Rico and Guam) in each state. We then linked these data with measures in the 2023 CMS Certification and Survey Provider Enhanced Report (CASPER) to determine staffing levels by NH size (small, medium, and large) and NH ownership type (for-profit, nonprofit), and multi-facility chain affiliation. Small NHs were defined as those with an average daily census of fewer than 50 residents. Medium and large NHs were defined as those with 50 to 150 residents and greater than 150 residents, respectively.

We used the United States Department of Agriculture's (USDA) Rural-Urban Commuting Area (RUCA) Codes to determine the rurality of NHs based on facility zip codes.¹³ NHs were divided into rural and non-rural categories based on whether their zip code fell into an “urbanized area” or “urban cluster”. When categorizing NHs by profit-status, rurality, and size, we excluded government-owned NHs as they make up a small proportion (5%) of the market.¹⁴

Additionally, we captured turnover events in the PBJ that occurred between July 1, 2020 and March 31, 2023. We identified turnover based on gaps in days worked by eligible employees (Appendix). We calculate a measure of turnover for 13 875 NHs.

MACPAC's compilation of state policies for NH staffing was used to identify policies implemented through 2021 and to examine how state regulations compare to the new federal standards.² We note that some states do not set minimum nurse staffing levels, while others repeat federal requirements for RNs, while specifying no required hours for LPNs or CNAs.² For example, 12 states (AL, HI, KY, MO, NE, NV, NH, NC, ND, SD, UT, and VA) provide no explicit HPRD requirements, leaving NHs subject to only federal HPRD requirements.² In these cases, we imputed values using MACPAC's estimated standard of 0.3 HPRD based on a hypothetical 100-bed facility and the staff hours the state requires.² Additionally, 5 states (IL, IA, KS, WI, WY) and Washington, D.C. adjust resident-staffing ratios based on resident acuity. For example, in Illinois, NHs are required to provide a minimum of 3.8 HPRD for residents needing skilled care and 2.5 HPRD for residents needing intermediate care. In our analysis, we used the highest staffing ratio from a state's regulation as the state requirement. Additionally, we compared staffing ratios by state staffing standards between states with <1 mandated HPRD (16 states), 1-2.5 mandated HPRD (14 states), 2.5-3.47 mandated HPRD (14 states), and 3.48+ mandated HPRD (6 states and DC).

Student's t-tests were used for comparisons of NHs by reported HPRDs and nurse turnover by NHs meeting new federal requirements. z-Tests were used to compare the proportion of NHs with higher vs lower overall and staffing five-star ratings on Care Compare based on the most recent ratings in 2023 (14 917 NHs).¹⁵ Additionally, we examined how state staffing standards compare to those in the new CMS rule.

Study results

Current staffing levels relative to the new CMS staffing standards varied by NH characteristics (Table 1). For-profit NHs, on average, had lower total reported nurse HPRD across small NHs 95% confidence interval (CI) (−0.9279, −0.7528), medium NHs (−0.5172, −0.4419) and large NHs (−0.5671, −0.3537); rural NHs (−0.6067, −0.3905) and non-rural NHs (−0.7631, −0.6849); and chain-associated NHs (−0.5765, −0.4836). Across these characteristics, nonprofit NHs staffed more RNs and CNAs, but typically staffed LPNs at approximately the same level as their for-profit counterparts (Table 1). On average, NHs failed to meet the CNA HPRD requirement for each size and ownership category, although small nonprofit NHs were the closest (2.44 reported HPRD). On average, for-profit NHs met the 0.55 RN HPRD threshold across the facility characteristics. Specifically, small, medium, and large for-profit NHs had 0.76, 0.52, and 0.49 RN reported HPRD, respectively. Chain-associated for-profit NHs had 0.57 RN reported HPRD.

Table 1.

Nurse hours-per-resident day by nurse category and nursing home characteristics.

	For-profit	CI	Nonprofit	CI	Difference	CI	P-value
Registered nurse (RN) HPRD
Small nursing home (<50 residents)	0.7593 (0.5771)	(0.7366, 0.7819)	1.2001 (0.9623)	(1.1496, 1.2506)	−0.4408	(−0.4892, −0.3924)	<0.0001
Medium (50-150 residents)	0.5216 (0.2627)	(0.5156, 0.5275)	0.7342 (0.3522)	(0.7181, 0.7503)	−0.2126	(−0.2271, −0.1982)	<0.0001
Large (150+ residents)	0.4926 (0.2178)	(0.4761, 0.5091)	0.6925 (0.4042)	(0.6323, 0.7526)	−0.1998	(−0.2442, −0.1554)	<0.0001
Rural	0.5964 (0.2860)	(0.5710, 0.6217)	0.8205 (0.3705)	(0.7797, 0.8613)	−0.2241	(−0.2696, −0.1786)	<0.0001
Non-rural	0.5745 (0.3779)	(0.5671, 0.5819)	0.9335 (0.7382)	(0.9075, 0.9595)	−0.359	(−0.3786, −0.3394)	<0.0001
Chain association	0.5742 (0.3596)	(0.5657, 0.5828)	0.8751 (0.5948)	(0.8455, 0.9048)	−0.3009	(−0.3237, −0.2781)	<0.0001
Licensed practical nurse (LPN) HPRD
Small nursing home (<50 residents)	0.8991 (0.4864)	(0.8800, 0.9182)	0.8710 (0.5267)	(0.8434, 0.8987)	0.0281	(−0.00473, 0.0609)	0.0934
Medium (50-150 residents)	0.9024 (0.3014)	(0.8956, 0.9093)	0.8682 (0.3408)	(0.8526, 0.8838)	0.0342	(0.0184, 0.0500)	<0.0001
Large (150+ residents)	0.7967 (0.2786)	(0.7756, 0.8178)	0.7831 (0.2760)	(0.7420, 0.8241)	0.0137	(−0.0326, 0.0599)	0.5619
Rural	0.7397 (0.2938)	(0.7137, 0.7658)	0.6726 (0.3543)	(0.6335, 0.7116)	0.0672	(0.0221, 0.1122)	0.0035
Non-rural	0.9025 (0.3543)	(0.8956, 0.9094)	0.8848 (0.4265)	(0.8698, 0.8999)	0.0177	(0.00266, 0.0326)	0.021
Chain association	0.8858 (0.3196)	(0.8782, 0.8934)	0.8380 (0.3736)	(0.8193, 0.8566)	0.0478	(0.0296, 0.0660)	<0.0001
Certified nurse aide (CNA) HPRD
Small nursing home (<50 residents)	2.0078 (0.6480)	(1.9823, 2.0332)	2.4354 (0.7485)	(2.3962, 2.4747)	−0.4277	(−0.4726, −0.3828)	<0.0001
Medium (50-150 residents)	1.888 (0.4751)	(1.8772, 1.8988)	2.1891 (0.5520)	(2.1639, 2.2144)	−0.3011	(−0.3262, −0.2760)	<0.0001
Large (150+ residents)	1.8967 (0.4361)	(1.8636, 1.9297)	2.1709 (0.4593)	(2.1026, 2.2392)	−0.2742	(−0.3475, −0.2009)	<0.0001
Rural	1.7886 (0.4663)	(1.7472, 1.8300)	2.1303 (0.6111)	(2.0630, 2.1976)	−0.3417	(−0.4163, −0.2671)	<0.0001
Non-rural	1.9228 (0.5229)	(1.9127, 1.9330)	2.3055 (0.6490)	(2.2826, 2.3284)	−0.3827	(−0.4050, −0.3603)	<0.0001
Chain association	1.853 (0.4808)	(1.8416, 1.8644)	2.13 (0.5772)	(2.1012, 2.1588)	−0.277	(−0.3046, −0.2494)	<0.0001
Total nurse HPRD
Small nursing home (<50 residents)	3.6662 (1.2274)	(3.6180, 3.7144)	4.5066 (1.5123)	(4.4273, 4.5859)	−0.8404	(−0.9279, −0.7528)	<0.0001
Medium (50-150 residents)	3.3120 (0.7210)	(3.2956, 3.3283)	3.7915 (0.8056)	(3.7547, 3.8284)	−0.4796	(−0.5172, −0.4419)	<0.0001
Large (150+ residents)	3.1860 (0.6163)	(3.1393, 3.2327)	3.6464 (0.7332)	(3.5373, 3.7555)	−0.4604	(−0.5671, −0.3537)	<0.0001
Rural	3.1247 (0.6373)	(3.0681, 3.1813)	3.6233 (0.9287)	(3.5210, 3.7256)	−0.4986	(−0.6067, −0.3905)	<0.0001
Non-rural	3.3998 (0.8838)	(3.3826, 3.4170)	4.1238 (1.2164)	(4.0810, 4.1667)	−0.724	(−0.7631, −0.6849)	<0.0001
Chain association	3.3130 (0.7962)	(3.2941, 3.3320)	3.8431 (1.0145)	(3.7925, 3.8938)	−0.5301	(−0.5765, −0.4836)	<0.0001

t-Tests were produced to test comparisons. Standard deviations are reported in parentheses. 14 302 nursing homes present in the 2023 PBJ data are used to construct these estimates. We excluded 598 government-owned nursing homes and 6 nursing homes in Puerto Rico or Guam. The estimates include 13 479 non-rural (10 283 for-profit, 3196 nonprofit); 823 rural (496 for-profit, 327 nonprofit); 848 large (672 for-profit, 176 nonprofit); 9306 medium (7467 for-profit, 1839 nonprofit); 4148 small (2640 for-profit, 1508 nonprofit); and 8411 chain-affiliated (6841 for-profit, 1570 nonprofit) nursing homes. Source: Nurse hours-per-resident day are calculated from the Centers for Medicare & Medicaid Services (CMS) Payroll-Based Journal nurse staffing and resident census data. Provider ownership data are retrieved from CMS Certification and Survey Provider Enhanced Reports (CASPERs).

HPRD, hours per resident day; RN, registered nurse; CNA, certified nurse aide; LPN, licensed practice nurse.

The proportion of NHs that met the new federal requirements in 2023 ranged from 31% in states with the lowest staffing requirements (<1 total nurse HPRD) to 65% in the states with the highest staffing requirements (≥3.48 total nurse HPRD) (Figure 2). States with staffing standards equal to or greater than the new federal standard (CA, D.C., FL, IL, MA, NY, and RI) had the highest total staffing levels, and higher levels of CNA and LPN staffing. Despite the higher required staffing levels, NHs in 3 of these states (IL, NY, RI), on average, did not meet the new federal standard of 0.55 HPRD for RN care.

Figure 2.

Average nurse hours-per-resident day by state staffing policy. Source: State policies are retrieved from the Medicaid and Chip Payment and Access Commission. Nurse hours-per-resident day are calculated from the Centers for Medicare & Medicaid Services Payroll-Based Journal nurse staffing and resident census data. HPRD, hours per resident day; RN, registered nurse; CNA, certified nurse aide; LPN, licensed practice nurse. t-Tests were produced to test comparisons. The analysis includes 14 900 nursing homes present in at least one quarter of 2023 PBJ data. Six nursing homes in Guam and Puerto Rico are excluded. States with standards < 1 HPRD: AK, AL, AZ, HI, IN, KY, MO, NC, ND, NE, NH. NV, SD, TX, UT, VA. States with standards between 1 and 2.5 HPRD: CO, GA, IA, ID, KS, LA, MI, MN, MT, OR, SC, TN, WV, WY. States with standards between 2.5 and 3.47 HPRD: AR, CT, DE, MD, ME, MS, NJ, NM, OH, OK, PA, VT, WA, WI. States with standards 3.48+: CA, DC, FL, IL, MA, NY, RI. There are 4213 nursing homes in states with staffing policies of <1 HPRD; 3250 nursing homes in states ranging from 1-2.5 HPRD, 3845 in states ranging from 2.5-3.47 HPRD, and 3592 in states with 3.48+ HPRD policies.

Higher state staffing standards were generally associated with better overall NH ratings on Care Compare, with more 4- and 5-star ratings and fewer 1- and 2-star ratings (Figure 3, Appendix Table S1). For instance, in states with staffing requirements of <1 HPRD, 1-2.5 HPRD, and 2.5-3.47 HPRD, the proportion of facilities with 1- or 2-star ratings compared to 4- or 5-star ratings was 14.7 percentage points (pp) (CI: 12.6, 16.8), 9.2 pp (CI: 6.9, 11.6), and 7 pp (CI: 4.8, 9.1) higher, respectively. No difference was observed in states with staffing standards of 3.48 HPRD or greater (Appendix Table S1).

Figure 3.

Nursing home Care Compare star rating by state staffing requirements. Source: State policies were retrieved from the Medicaid and Chip Payment and Access Commission. Star ratings were retrieved from the Centers for Medicare & Medicaid Services provider information files. HPRD, hours-per-resident day; RN, registered nurse; CNA, certified nurse aide; LPN, licensed practice nurse. States with standards < 1 HPRD: AK, AL, AZ, HI, IN, KY, MO, NC, ND, NE, NH. NV, SD, TX, UT, VA. States with standards between 1 and 2.5 HPRD: CO, GA, IA, ID, KS, LA, MI, MN, MT, OR, SC, TN, WV, WY. States with standards between 2.5 and 3.47 HPRD: AR, CT, DE, MD, ME, MS, NJ, NM, OH, OK, PA, VT, WA, WI. States with standards 3.48+: CA, DC, FL, IL, MA, NY, RI. Analysis includes 14 917 nursing homes.

Finally, in Figure 4, we show the percentage of annual nurse turnover based on categories of state staffing requirements (<1 HPRD, 1-2.5 HPRD, 2-5-3.47 HPRD, 3.48+ HPRD). In all state staffing categories, the turnover in NHs that met the new federal standard was lower than those that did not (Appendix Table S2).

Figure 4.

Nurse turnover and state staffing policy. Source: State policies were retrieved from the Medicaid and Chip Payment and Access Commission. Nurse turnover was calculated from the Centers for Medicare & Medicaid Services Payroll-Based Journal nurse staffing data. States with standards < 1 HPRD: AK, AL, AZ, HI, IN, KY, MO, NC, ND, NE, NH. NV, SD, TX, UT, VA. States with standards between 1 and 2.5 HPRD: CO, GA, IA, ID, KS, LA, MI, MN, MT, OR, SC, TN, WV, WY. States with standards between 2.5 and 3.47 HPRD: AR, CT, DE, MD, ME, MS, NJ, NM, OH, OK, PA, VT, WA, WI. States with standards 3.48+: CA, DC, FL, IL, MA, NY, RI. We include estimates of turnover for 13 875 nursing homes in 2023. Of these nursing homes, 3724 are in states with staffing policies of <1 HPRD; 3005 nursing homes are in states with policies ranging from 1-2.5 HPRD; 3668 nursing homes are in states with policies ranging from 2.5-3.47 HPRD; and 3478 are in states with 3.48+ HPRD policies.

Our study has limitations to consider. For example, it is not clear whether the relationships between NHs and staffing levels we identified will continue if the new federal standards are implemented. Our results are based on data from 2023, a period following the end of the Coronavirus disease 2019 (COVID-19) public health emergency. Previous research has indicated that nurse staffing levels remained largely unchanged during the COVID-19 pandemic,^16-19 suggesting that our estimates of NHs failing to meet the new federal standards may persist after their implementation. Furthermore, inspection delays have extended beyond the official end of the public health emergency. This may contribute to inaccurate Care Compare star ratings reported by CMS, as used in our findings.²⁰ Our analysis is descriptive and does not establish causality between state and federal standards, and staffing ratios and turnover. We do not adjust for other NH characteristics, such as facility culture, or other policies that may contribute to state variation in staffing and turnover. These include laws in some states requiring that Medicaid funding for NHs be used to increase compensation for nurses through Medicaid wage pass-through requirements and restrictions on nurse unionization.^21,22

Discussion

Our study found that the mean reported HPRD falls short of the new CMS standards in a minority of states. We also found higher staffing levels and NH quality ratings in states with higher staffing standards. Nonprofit homes were found to consistently report higher overall staffing. Smaller and non-rural NHs were also frequently found to report higher staffing levels.

As the future of the new staffing standard remains uncertain, these findings provide additional detail on state-level staffing standards while aligning with previous studies on the impact of the new federal standard. For instance, previous research has found that the new CMS staffing standard would require nearly half of all NHs to increase their overall staffing levels.²³ Additionally, a recent report found that only 11% of for-profit NHs would currently meet the new standards, compared to 41% of nonprofit homes.⁹

In addition to many NHs potentially failing to meet the new federal requirements, the composition of direct care staff is likely to change under the new rule. Although state scope of practice laws vary in the allowable activities for CNA, LPN, and RNs, there are common activities nationwide.^24,25 Common CNA tasks include assistance with activities of daily living such as bathing and dressing, in addition to measuring vital signs. RN tasks include conducting resident assessments, care planning, and supervision of other nurses, while LPNs typically handle bedside care and are supervised by nurses.²⁶ RNs on average have the highest wages, followed by LPNs and CNAs.²⁷ Many states' standards do not include specific levels for LPNs, CNAs, and RNs, and the new federal rule does not specify a minimum HPRD for LPNs. NHs may reduce LPN staffing to increase CNA or RN levels to reduce total staffing costs. This could incentivize facilities to substitute lower-cost CNA staffing for higher-cost LPNs while maintaining RN staffing near the required threshold. Additionally, NHs may replace support staff for direct care staff to meet the new federal requirements.²⁸ State staffing policies during the COVID-19 pandemic other than mandated levels may also have contributed to variation in staffing levels. For example, Alaska has a statute which reduces the CNA training requirement from 140 to 75 h during any state-declared public health emergency, which may have contributed to the state's relatively high staffing ratios.²⁹ Some states also increased wages for nurses and permanently or temporarily relaxed training requirements due to the pandemic.²⁹

As a new presidential administration has taken office and 21 state attorneys general have filed lawsuits to block the new federal rule, its future remains uncertain.^30,31 Additionally, the 2025 federal budget reconciliation bill has prohibited the Department of Health and Human Services from enforcing the rule until October 1, 2034.³² Regardless of whether the new federal rule is implemented, inadequate staffing in NHs is likely to remain an ongoing concern with states continuing to adjust their own staffing standards.

Supplementary material

Supplementary material is available at Health Affairs Scholar online.

Funding

None.