Abstract
Objectives
The nursing home (NH) population has become increasingly diverse, yet many facilities remain de facto racially segregated. This study examines whether a high proportion of Black, Indigenous, and People of Color (BIPOC) residents is associated with nursing staff levels.
Design
We constructed a longitudinal cohort of NHs (2013–2019) by linking Certification and Survey Provider Enhanced Reports, LTCFocUS.org, Medicare Cost Reports, and Payroll-Based Journal data. Separate multivariable random-effects linear regressions were conducted.
Setting and Participants
14,075 Medicare- and Medicaid-certified NHs in the U.S.
Methods
The proportion of BIPOC residents was categorized as the 10% of nursing homes serving the highest minority residents in each state each year (High-BIPOC), and the remaining 90% (Low- BIPOC). Total nursing staff levels in hours per resident day (HPRD) included both hours paid (2013–2019) and hours worked (2017–2019). The total staff included registered nurses, licensed practical nurses, and certified nurse aides.
Results
The unadjusted difference in total staff levels between High-BIPOC and Low-BIPOC NHs increased from −0.23 HPRD (4.19 vs. 4.42) in 2013 to −0.35 HPRD (3.94 vs. 4.29) in 2019 for hours paid. The difference in hours worked increased from −0.19 (3.55 vs. 3.74) in 2017 to −0.23 (3.50 vs. 3.73) in 2019. The difference became smaller but remained significant after controlling for covariates (−0.037 HPRD for hours paid, and −0.038 for hours worked). Analyses of individual staff types found lower levels of registered nurses and certified nurse aides (but not licensed practical nurses) among high-BIPOC nursing homes. Findings were robust to treating racial and ethnic composition as a continuous variable or excluding payer mix from the models.
Conclusions and Implications
NHs with high concentrations of minority residents reported lower nursing staff levels. Improving staffing in NHs serving primarily marginalized racial and ethnic groups remains a policy priority.
Keywords: Nursing homes, racial and ethnic disparities, staff levels
Brief summary:
In this longitudinal cohort study using 2013–2019 data, we demonstrated that a higher concentration of racial and ethnic marginalized residents in nursing homes was associated with lower total and individual staffing levels.
INTRODUCTION
Over the past two decades, nursing homes have served a more racially and ethnically diverse population. From 1999 to 2008, the proportion of Blacks (from 9.6% to 11.3%), Hispanics (from 2.3% to 3.8%), and Asians (from 1.0% to 1.6%) increased, while the proportion of non-Hispanic White residents decreased from 86.7% to 82.9%.1 This demographic shift continued between 2011 and 2017.2 The growing number of Black, Indigenous, and People of Color (BIPOC) residents, however, has not changed the highly segregated nature of nursing home care.3 Approximately 50% of BIPOC residents reside in only 10% of U.S. nursing homes.4–7
Compared to White residents, newly admitted Black and Hispanic residents report higher levels of functional and cognitive impairments,8 leading to a greater need for nursing and personal care. Despite this increasing need, nursing homes with high concentrations of BIPOC residents tend to have fewer available financial resources.4 The percentage of Medicaid residents was 57% in facilities with <5% BIPOC residents vs. 77% in facilities with ≥35% BIPOC residents.9 Medicaid reimburses nursing homes at lower rates than Medicare or private payers.10 A greater care need coupled with fewer financial resources makes it harder for minority-serving nursing homes to recruit and retain nursing staff, resulting in racial and ethnic disparities in staffing. Inadequate staff levels could compromise nursing homes’ quality of care7,9,11–13 and hurt residents’ quality of life.14
Empirical evidence suggests that nursing homes with high concentrations of BIPOC residents have lower staffing levels.2,9 These studies, however, often rely on data sources like Certification and Survey Provider Enhanced Reports (CASPER) that collected self-reported staffing information from nursing homes.15 Since early 2010s, the Centers for Medicare and Medicaid Services (CMS) started collecting more reliable staffing data based on accounting information in the Medicare Cost Reports16 and, in 2017, employee payroll records in the Payroll-Based Journal (PBJ) system17. No study has examined differences in staffing levels by racial and ethnic compositions using longitudinal Medicare Cost Reports or PBJ data.
To mitigate these knowledge gaps, we used 2013–2019 national nursing home data to describe the time trends in nursing staff levels in nursing homes with high vs. low concentrations of racial and ethnic minority residents. We then performed separate longitudinal regressions to examine the associations of a facility’s racial and ethnic composition with total and individual nursing staff levels.
METHODS
Data Sources
We linked the 2013–2019 CASPER data with the LTCFocUS.org data and Medicare Cost Reports, all at the facility level. This study focused on data from 2013 to 2019 to reflect normal operations of nursing homes before the COVID-19 pandemic. The CASPER data were collected as part of initial and annual recertification inspections of all Medicare- and Medicaid-certified nursing homes, with these inspections occurring every 9 to 15 months.18,19 CASPER included detailed facility characteristics, such as chain affiliation, ownership, and payer mix.20 The LTCFocUS.org data provided annual summarized information on nursing homes from CASPER and Minimum Data Set assessments.2 The Medicare Cost Reports were mandatory annual cost reports for Medicare-certified facilities, which collected the number of hours paid to nursing staff.16 We also obtained the 2017–2019 PBJ data for the number of hours worked by nursing staff based on auditable payroll records.17 PBJ data were unavailable before 2017. Cost Reports and PBJ data provided the best available data on nursing home staffing.16,20,21 The 2010 Rural-Urban Commuting Area Codes identified the rurality of nursing homes.22 This study was exempted by the University of Texas Medica Branch Institutional Review Board. This study followed the Strengthening the Reporting of Observational Studies in Epidemiology (STROBE) reporting guideline for cohort studies.
Study Cohort
We assembled a longitudinal cohort of all Medicare- and Medicaid-certified nursing homes in the U.S. for 2013–2019 (Supplemental Figure 1). We initially identified 16,107 unique nursing homes that completed the CASPER surveys. We excluded 1,150 nursing homes that could not be linked to the Medicare Cost Reports, most of which were hospital-based facilities. We then dropped 837 nursing homes without information on the racial and ethnic composition of residents. Finally, we excluded 45 nursing homes with aberrant staffing level data (i.e., zero staff, beyond +/− four standard deviations, or > 24 HPRD).16,20,21 This resulted in a final longitudinal cohort of 84,722 facility-year observations from 14,075 unique nursing homes, with each facility having up to 7 observations from 2013 to 2019.20
Exposure/Measures
The independent variable is the racial and ethnic composition of residents in a facility.2 LTCFocUS.org calculated the composition among all residents who resided in a nursing home on the first Thursday in April using Minimum Data Set assessments.2 BIPOC residents included those who were Blacks, Hispanics, Asians, Native Americans, and Pacific Islanders, excluding non-Hispanic Whites.23,24 In the primary analyses, we ranked nursing homes based on the proportion of BIPOC residents in each state for each year. Our BIPOC concentration variable defined “High-BIPOC” as the top 10% of nursing homes serving the largest proportion of BIPOC residents in each state for each year.25 The bottom 90% of nursing homes were classified as “Low-BIPOC.” This approach acknowledged the fact that both racial and ethnic compositions in nursing homes and state staffing regulations vary substantially by state.24,26 By using state-specific cutoffs, we aimed to understand how the most racially and ethnically diverse nursing homes in each state operate relative to other nursing homes within the same state.24,27 A sensitivity analysis using the top 25% as the cutoff produced similar results.
Staffing outcomes included two measures of total nursing staff levels in hours per resident day (HPRD): hours paid (2013–2019) and hours worked (2017–2019).16 The total nursing staff included registered nurses (RNs), licensed practical nurses (LPNs), and certified nursing assistants (CNAs). RNs provide comprehensive patient care, coordinate treatment plans, and supervise other nursing staff.28 LPNs provide nursing care under the supervision of an RN, who along with CNAs provide direct care in support of daily living activities, like bathing and feeding.29,30 The hours paid from Medicare Cost Reports included hours for direct care and meal breaks, paid time off, and vacation time. The hours worked from PBJ excluded any time (paid or unpaid) for meal breaks, paid time off, and vacation time.16 Hours paid were available for the whole study period. Hours worked, although only available since 2017 in the PBJ data, better reflects the number of hours staff care for residents.
We controlled for nursing home covariates potentially associated with the staffing outcomes: chain affiliation, ownership, number of beds, occupancy rate, case-mix acuity index, percentage of residents with dementia, percentage of residents with depression, percentage of residents with serious mental illness, percentage of Medicaid residents, percentage of Medicare residents, and rurality.5,12,14,31–33
Statistical Analysis
We first described the variations in the proportion of BIPOC residents over time and by facility. We then presented the cutoff values of the top 10% in each state in 2019. We tested whether nursing home characteristics in 2019 differed by BIPOC concentration using t-tests for continuous variables and Chi-squared tests for categorical variables. We plotted the time trends in the two staffing measures (hours paid 2013–2019, and hours worked 2017–2019) by High vs. Low BIPOC concentration. We also assessed the bivariate relationship of racial and ethnic composition with two staffing measures using 2019 data.
To examine the longitudinal associations of BIPOC concentration with staffing outcomes, we employed separate panel random-effects linear regressions for each outcome. The random-effect models accounted for the correlation within the same nursing home and the variability between nursing homes. Fixed-effects regressions were deemed inappropriate because the independent variable —BIPOC concentration— varied minimally within a facility over time.34 Each nursing home had a maximum of 7 observations for the total nursing staff level (hours paid) from 2013 to 2019 and had a maximum of 3 observations for the total nursing staff level (hours worked) from 2017 to 2019. These models adjusted for year-fixed effects, state-fixed effects, and the facility covariates described previously. We also examined potential heterogenous associations of BIPOC concentration with different types of staff, by repeating the random-effects linear regressions on individual staffing levels of RNs, LPNs, and CNAs for hours paid and hours worked.
We performed two sets of sensitivity analyses. To assess the robustness of our primary findings, the first set of analyses repeated the longitudinal models on total staff levels with the continuous racial and ethnic composition variable. Because nursing homes caring for BIPOC residents relied more on Medicaid payments, the second set of analyses examined the impact of removing payer mix on our findings by estimating the longitudinal models without adjusting for the percentage of Medicare residents and the percentage of Medicaid residents.
All statistical analyses were performed using SAS 9.4 (SAS Institute Inc., Cary, NC) and Stata 16.0 (StataCorp LLC, College Station, TX).
RESULTS
The proportion of BIPOC residents in nursing homes remained stable from 2013 to 2019, ranging from 11.38% to 11.73%. There were large variations in the proportion across facilities (Supplemental Figure 2). The cutoff values for the top 10% of nursing homes serving the largest proportion of BIPOC residents varied substantially across states, ranging from 6% of BIPOC residents to over 70% (Supplemental Table 1). Table 1 compares the characteristics of the top 10% of nursing homes vs. the bottom 90% of facilities in 2019. High-BIPOC facilities were more likely to be for-profit (85.8% vs. 72.1%), had more beds (127.8 beds vs. 109.9 beds), had a higher percentage of Medicaid residents (70.5% vs. 59.6%), and were more likely located in urban areas (95.4% vs. 89.8%). In the bivariate analyses, racial and ethnic composition was negatively associated with total nursing staff levels (Supplemental Figure 3), with a Pearson correlation coefficient of −0.12 for hours paid and −0.16 for hours worked.
Table 1.
Nursing Home Characteristics by Racial and Ethnic Composition, 2019
| Variables | All nursing homes (N=11,960) | High-BIPOC nursing homesa (N=1,225) | Low-BIPOC nursing homesb (N=10,735) | P valuec |
|---|---|---|---|---|
|
| ||||
| Mean (SD) | Mean (SD) | Mean (SD) | ||
|
| ||||
| Nursing home characteristics | ||||
| Chain affiliation, N (%) | 7,203 (60.23%) | 741 (60.49%) | 6,462 (60.20%) | 0.84 |
| Ownership, N (%) | ||||
| For-profit | 8,791 (73.50%) | 1,051 (85.80%) | 7,740 (72.10%) | <0.001 |
| Government | 592(4.95%) | 36 (2.94%) | 556(5.18 %) | |
| Not-for-profit | 2,577 (21.55%) | 138(11.27%) | 2,439 (22.72 %) | |
| Number of beds | 111.73 (56.76) | 127.8 (63.99) | 109.9 (55.58) | <0.001 |
| Occupancy rate (0–100) | 79.19 (15.43) | 79.90 (14.38) | 79.12 (15.55) | 0.07 |
| Case-mix acuity index | 10.45 (1.26) | 10.71 (1.46) | 10.42 (1.2923 | <0.001 |
| % Residents with dementia | 44.85 (16.25) | 41.22 (16.22) | 45.27 (16.21) | <0.001 |
| % Residents with serious mental illness | 33.95 (17.74) | 34.31 (17.66) | 33.91 (17.75) | 0.45 |
| % Residents with depression | 36.58 (23.24) | 31.24 (22.41) | 37.19 (23.26) | <0.001 |
| % Medicare residents (0–100) | 12.48 (11.18) | 10.40 (9.29) | 12.72 (11.35) | <0.001 |
| % Medicaid residents (0–100) | 60.73 (22.08) | 70.48 (19.22) | 59.62 (22.11) | <0.001 |
| Rurality, N (%) | ||||
| Urban | 10,804 (90.33%) | 1,168 (95.35%) | 9,636 (89.76%) | <0.001 |
| Rural | 1,156(9.67 %) | 57 (4.65%) | 1,099 (10.24%) | |
| Outcome measures | ||||
| Total nursing staff hours per resident day (hours worked) | 3.73 (0.70) | 3.54 (0.63) | 3.75 (0.71) | <0.001 |
| Total nursing staff hours per resident day (hours paid) | 4.36 (1.61) | 4.07 (1.41) | 4.39 (1.63) | <0.001 |
Notes: BIPOC, Black, Indigenous, and People of Color; SD, Standard Deviation.
High-BIPOC nursing homes are defined as the top 10% of nursing homes serving the largest proportion of minority residents in each state.
The bottom 90% of nursing homes are classified as Low-BIPOC.
T-tests for continuous variables, Chi-Square tests for categorical variables.
Figure 1 examines the trends in total nursing staff hours per resident day (hours paid and hours worked) in nursing homes by BIPOC concentration from 2013 to 2019. Overall, High-BIPOC nursing homes reported lower staffing levels (hours paid) than Low-BIPOC nursing homes. In 2013, staffing levels (hours paid) were 4.19 HPRD in High-BIPOC nursing homes vs. 4.42 HPRD in Low-BIPOC nursing homes. In 2019, total staffing levels (hours paid) decreased to 3.94 HPRD in High-BIPOC nursing homes and to 4.29 HPRD in Low-BIPOC nursing homes. Thus, the difference increased from 2013 to 2019 (−0.23 HPRD vs. −0.35 HPRD). From 2017 to 2019, the difference in total staffing levels (hours worked) slightly increased from −0.19 HPRD in 2017 to −0.23 HPRD.
Figure 1. Trends in Total Nursing Staff Levels in Nursing Homes by Racial and Ethnic Composition, 2013–2019.
The nursing staff levels are measured in hours per resident day. Data on hours paid for 2013–2019 are from the Medicare Cost Reports; data on hours worked for 2017–2019 are from the Payroll-Based Journal (PBJ). The number of nursing homes included each year ranges from 11,834 to 12,302. High-BIPOC is defined as the top 10% of facilities serving the largest proportion of Black, Indigenous, and People of Color (BIPOC) residents in each state for each year. The bottom 90% of nursing homes are classified as Low-BIPOC. The 95% confidence intervals are not presented because of small variations.
To assess the association of BIPOC status with total staffing levels, we performed two separate random-effects linear regressions on hours paid using 2013–2019 Medicare Cost Report data and on hours worked using 2017–2019 PBJ data (Table 2). Compared to low-BIPOC nursing homes, high-BIPOC nursing homes reported lower hours paid (β = −0.037 HPRD, 95% CI: −0.08, −0.004) and lower hours worked for total nursing staff (β = −0.038 HPRD, 95% CI: −0.06, −0.02). In the analyses of individual staff levels, High-BIPOC nursing homes reported lower hours paid for RNs (β = −0.02 HPRD, 95% CI: −0.03, −0.001) and CNAs (β = −0.03 HPRD, 95% CI: −0.05, −0.01) (Table 3), but not LPNs. Similar findings on hours worked were found in Table 3.
Table 2.
Random Effects Regressions Examining the Association of Racial and Ethnic Composition with Total Nursing Staff Levels
| Variables | Total nursing staff hours per resident day (hours paid), 2013–2019a | Total nursing staff hours per resident day (hours worked), 2017–2019b |
|---|---|---|
|
| ||
| Coef. (95% CI) | Coef. (95% CI) | |
|
| ||
| Independent variable | ||
| High-BIPOC (vs. low-BIPOC) | −0.037 (−0.079 to −0.017) | −0.038 (−0.058 to −0.017) |
| Nursing home characteristics | ||
| Chain affiliation | −0.080 (−0.102 to −0.058) | −0.084 (−0.098 to −0.071) |
| Ownership (Ref: For-profit) | ||
| Government | 0.149 (0.097 to 0.200) | 0.239 (0.208 to 0.269) |
| Not-for-profit | 0.329 (0.291 to 0.366) | 0.311 (0.292 to 0.330) |
| Number of beds | −0.0001 (−0.001 to 0) | −0.001 (−0.001 to −0.001) |
| Occupancy rate (0–100) | −0.011 (−0.012 to −0.010) | −0.007 (−0.007 to −0.006) |
| Case-mix acuity index | 0.009 (0.001 to 0.017) | 0.039 (0.034 to 0.044) |
| % Residents with dementia | −0.00004 (−0.001 to 0.001) | 0.001 (0.001 to 0.001) |
| % Residents with serious mental illness | −0.001 (−0.001 to 0) | −0.002 (−0.002 to −0.001) |
| % Residents with depression | 0.0001 (−0.000 to 0.001) | 0.0002 (−0.000 to 0) |
| % Medicare residents (0–100) | 0.008 (0.006 to 0.0081) | 0.007 (0.006 to 0.007) |
| % Medicaid residents (0–100) | −0.004 (−0.005 to −0.003) | −0.007 (−0.007 to −0.006) |
| Rural (Ref: Urban) | −0.16 (−0.245 to −0.075) | −0.066 (−0.095 to −0.036) |
| Observations | 84,722 | 35,651 |
| Unique Nursing homes | 14,075 | 13,649 |
Notes:
Data on hours paid for 2013–2019 are from the Medicare Cost Reports, with each nursing home having up to 7 facility-year observations;
data on hours worked for 2017–2019 are from the Payroll-Based Journal (PBJ), with each nursing home having up to 3 facility-year observations;
CI, confidence interval; Coef., coefficient; High-BIPOC, the top 10% of facilities serving the largest proportion of Black, Indigenous, and People of Color (BIPOC) residents in each state for each year; Low-BIPOC, the bottom 90% of nursing homes in each state for each year.
Year- and state-fixed effects are not presented.
Table 3.
Random Effects Regressions Examining the Association of Racial and Ethnic Composition with Individual Nursing Staff Levels
| Staff hours per resident day (hours paid), 2013–2019a | Staff hours per resident day (hours worked), 2017–2019 b | |||||
|---|---|---|---|---|---|---|
|
| ||||||
| Variables | Registered Nurse | Licensed Practical Nurse | Certified Nursing Assistant | Registered Nurse | Licensed Practical Nurse | Certified Nursing Assistant |
|
| ||||||
| Coef. (95% CI) | Coef. (95% CI) | Coef. (95% CI) | Coef. (95% CI) | Coef. (95% CI) | Coef. (95% CI) | |
|
| ||||||
| Independent variable | ||||||
| High-BIPOC (vs. low-BIPOC) | −0.018 (−0.029 to −0.001) | 0.009 (−0.001 to 0.019) | −0.027 (−0.049 to −0.005) | −0.020 (−0.028 to −0.012) | 0.007 (−0.001 to 0.016) | −0.021 (−0.036 to −0.006) |
| Observations | 84,722 | 84,722 | 84,722 | 35,951 | 35,951 | 35,951 |
| Unique Nursing homes | 14,075 | 14,075 | 14,075 | 13,649 | 13,649 | 13,649 |
Notes: Data on hours paid for 2013–2019 are from the Medicare Cost Reports, with each nursing home having up to 7 facility-year observations;
data on hours worked for 2017–2019 are from the Payroll-Based Journal (PBJ), with each nursing home having up to 3 facility-year observations;
CI, confidence interval; Coef., coefficient; High-BIPOC, the top 10% of facilities serving the largest proportion of Black, Indigenous, and People of Color (BIPOC) residents in each state for each year; Low-BIPOC, the bottom 90% of nursing homes in each state for each year.
Covariates (chain affiliation, ownership, number of beds, occupancy rate, case-mix acuity index, % residents with dementia, % residents with serious mental illness, % residents with depression, and rurality), as well as year- and state-fixed effects are not presented.
Finally, sensitivity analyses confirmed the negative association of racial and ethnic composition with total nursing staff levels when using the continuous BIPOC variable (Supplemental Table 2). Another set of sensitivity analyses examined whether removing payer mix from the random effects regressions affected the association of BIPOC status with staff levels (Supplemental Table 3). Overall, removing payer mix resulted in minimum changes in the estimates for total and individual staff levels (both hours paid and hours worked).
DISCUSSION
Using longitudinal Medicare Cost Reports and PBJ data, this study found that nursing homes with a higher concentration of racial and ethnic minority residents consistently reported lower nursing staff levels. In the longitudinal models, disparities in nursing staff levels became smaller but remained significant after controlling facility characteristics. These results are robust to the staffing measures used (hours paid or hours worked), the measures of racial and ethnic composition (bivariate or continuous), or whether the payer mix was controlled or not. Our findings are consistent with previous studies using CASPER data or cross-sectional PBJ data, providing further evidence of racial and ethnic disparities in nursing home staffing.2,9
One concerning finding is that disparities in nursing staff levels (both hours paid and hours worked) worsened in 2019, before the COVID-19 pandemic hit nursing homes. Nursing homes lost 220,000 employees in 202035,36 and lag hospitals and home health agencies in recovering their staff.18,37 Over 20% of facilities still reported shortages in nursing staff and aides in September 2022.38 Almost half of nursing homes now rely on costly agency/contract workers to fill the staffing gaps.21 This can especially hurt minority-serving nursing homes with limited financial resources. Therefore, disparities in staffing levels may become worse during and after the pandemic.13,18,39
The analyses of individual types of nursing staff found lower levels of RNs and CNAs among nursing homes with high concentrations of BIPOC residents. Staff levels for LPNs did not differ by racial and ethnic composition. Nursing homes may use LPNs to substitute for more skilled RNs, which could lower the quality of care being delivered as LPNs practice near the frontiers of their scope of practice.28 Lower levels of CNAs in High-BIPOC facilities may reduce direct personal care provided to residents. Disparities in RN and CNA levels may change in the coming years because of the federal staffing mandates. In May 2024, CMS finalized the rule for the first-ever federal minimum staffing standards for nursing homes.40,41 The proposal included a minimum of 0.55 HPRD for RNs, 2.45 HPRD for CNAs, and 3.48 HPRD for total nursing staff. Many nursing homes would hire more staff or use agency staff to fulfill the new requirements.18,21,42 This could lead to reduced disparities in nursing staff levels because High-BIPOC nursing homes have to increase their staff levels more to meet the mandates. For example, in 2019, we found that only 40% of nursing homes with high concentrations of BIPOC residents would have met the RN standard vs. 53% of other facilities. Despite the promises, policymakers should ensure that minority-serving facilities have the necessary resources to recruit additional staff.43
Reliance on Medicaid payments could have explained the lower staffing levels in nursing homes with a high concentration of BIPOC residents.32 However, excluding % Medicaid residents and % Medicare residents did not change the direction of the association of BIPOC status with staff levels. This could be due to the inclusion of a random effect, or this may suggest that other structural factors beyond financial resources also contribute to poorer staffing levels in high-BIPOC nursing homes.25 For example, High-BIPOC nursing homes are more likely to be for-profit, which is associated with lower staffing.44 Also, nursing homes with mostly BIPOC residents may have poorer working environments that hurt staff levels.28,45
The fundamental question is how can we minimize (if not eliminate) the enduring racial and ethnic disparities in nursing home staffing? A solution would be to address systemic racism in the design of healthcare systems, which leads to racial and spatial segregation3,5,6,8,12,14,25,46 and makes it harder for historically marginalized groups to access high-quality nursing homes.2,8 Our health system and policymakers should reduce barriers (geographic, financial, and informational) faced by historically marginalized older adults to access high-quality facilities. Because nursing homes serving these populations are more likely to be located in severely deprived neighborhoods, policies such as payment enhancement and workforce development grants that could be micro-targeted to those facilities should improve staff levels.46 Practice initiatives may also strengthen staff recruitment, training, and retention in nursing homes with high concentrations of BIPOC residents.8,25,47 One such example is to improve pre-licensure training experiences in nursing homes through academic and practice partnerships.48 Programs that improve the organizational culture and working environment should also increase retention and reduce turnover of nursing staff.49 Culturally-specific programs in nursing homes with high concentrations of BIPOC residents may also help. Those programs can improve nursing staff’s competency in meeting the needs of diverse residents, hence reducing staff burnout and turnover.47
LIMITATIONS
Staffing measures may have errors, including facilities reporting zero staff, > 24 staff hours per resident day, and beyond +/− four standard deviations.16 These aberrant staff levels are theoretically impossible for nursing home to operate, hence are likely due to errors.16,20,21 Excluding these facilities should have minimized the impact of aberrant staffing levels on our analyses. We also did not examine differences in other aspects of nursing home staffing, such as turnover and staff instability. Because of the observational study design, we cannot ascertain a causality of racial and ethnic composition with staffing. The year- and state-fixed effects in the models removed the impact of time trends and variations in state policies, providing conservative estimates. Finally, the underlying mechanisms leading to racial and ethnic disparities need to be further examined.
CONCLUSIONS AND IMPLICATIONS
Using longitudinal data from 2013–2019, we found persistent racial and ethnic disparities in nursing home staff levels. The tolls of racial and ethnic segregation in nursing homes cannot be ignored. Staff shortages during the COVID-19 pandemic may have further enlarged existing racial and ethnic disparities in staffing.13,18,21,50 Additional research is needed to identify policy and practice initiatives that can mitigate nursing home disparities, including understanding the role of facility culture and working environment in contributing to disparities in staff levels.
Supplementary Material
Supplemental Figure 1. Consort Selection Diagram for Nursing Homes
Supplemental Figure 2. Distribution of the Racial and Ethnic Composition in Nursing homes, 2019
Supplemental Figure 3. Bivariate Association of Racial and Ethnic Composition with Total Nursing Staff Levels in Nursing Homes, 2019
Data on hours paid are from the Medicare Cost Reports. Data on hours worked are from the Payroll-Based Journal. The red lines represent a simple linear regression between the proportion of Black, Indigenous, and People of Color residents in a facility and staff hours per resident day (hours paid and hours worked). The Pearson correlation coefficient was −0.12 for hours paid and −0.16 for hours worked (both P<0.01).
Supplemental Table 1. State Variations in the Cutoffs of the Top 10% Nursing Homes with the Largest Concentration of Racial and Ethnic Minority Residents, 2019
Supplemental Table 2. Random Effects Regressions Examining the Association of Continuous Racial and Ethnic Composition with Total Nursing Staff Levels
Supplemental Table 3. Random Effects Regressions Examining the Association of Racial and Ethnic Composition with Nursing Staff Levels without Adjusting for Payer Mix
Funding sources:
This work was supported by the National Institutes of Health (R01AG081282, R01AG087296, and P30AG024832).
Sponsor’s role:
The funders had no role in the design and conduct of the study; collection, management, analysis, and interpretation of the data; preparation, review, or approval of the manuscript; and decision to submit the manuscript for publication.
Footnotes
Conflicts of Interest: Drs. HX and JRB are supported by the National Institutes of Health (Grants # and #). Dr. BD is supported by the National Institutes of Health (Grant #). Dr. JRB receives funds from various state agencies and the federal government to conduct research on nursing homes and advise on nursing home policy. He has also been retained as an expert witness in legal matters involving the healthcare and other industries. Other authors have no conflicts of interest.
REFERENCES
- 1.Feng Z, Fennell ML, Tyler DA, Clark M, Mor V. The Care Span: Growth of racial and ethnic minorities in US nursing homes driven by demographics and possible disparities in options. Health Aff (Millwood). Jul 2011;30(7):1358–65. doi: 10.1377/hlthaff.2011.0126 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 2.Travers JL, Dick AW, Wu B, et al. A Profile of Black and Latino Older Adults Receiving Care in Nursing Homes: 2011–2017. J Am Med Dir Assoc. Nov 2022;23(11):1833–1837 e2. doi: 10.1016/j.jamda.2022.04.010 [DOI] [PubMed] [Google Scholar]
- 3.Rahman M, Foster AD. Racial segregation and quality of care disparity in US nursing homes. J Health Econ. Jan 2015;39:1–16. doi: 10.1016/j.jhealeco.2014.09.003 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 4.Smith DB, Feng Z, Fennell ML, Zinn JS, Mor V. Separate and unequal: racial segregation and disparities in quality across U.S. nursing homes. Health Aff (Millwood). Sep-Oct 2007;26(5):1448–58. doi: 10.1377/hlthaff.26.5.1448 [DOI] [PubMed] [Google Scholar]
- 5.Mack DS, Jesdale BM, Ulbricht CM, Forrester SN, Michener PS, Lapane KL. Racial Segregation Across U.S. Nursing Homes: A Systematic Review of Measurement and Outcomes. Gerontologist. Apr 2 2020;60(3):e218–e231. doi: 10.1093/geront/gnz056 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6.Sloane PD, Yearby R, Konetzka RT, Li Y, Espinoza R, Zimmerman S. Addressing Systemic Racism in Nursing Homes: A Time for Action. J Am Med Dir Assoc. Apr 2021;22(4):886–892. doi: 10.1016/j.jamda.2021.02.023 [DOI] [PubMed] [Google Scholar]
- 7.Li Y, Harrington C, Temkin-Greener H, et al. Deficiencies In Care At Nursing Homes And Racial/Ethnic Disparities Across Homes Fell, 2006–11. Health Aff (Millwood). Jul 2015;34(7):1139–46. doi: 10.1377/hlthaff.2015.0094 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 8.Rivera-Hernandez M, Kumar A, Epstein-Lubow G, Thomas KS. Disparities in Nursing Home Use and Quality Among African American, Hispanic, and White Medicare Residents With Alzheimer’s Disease and Related Dementias. J Aging Health. Aug 2019;31(7):1259–1277. doi: 10.1177/0898264318767778 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 9.Campbell LJ, Cai X, Gao S, Li Y. Racial/Ethnic Disparities in Nursing Home Quality of Life Deficiencies, 2001 to 2011. Gerontol Geriatr Med. Jan-Dec 2016;2:2333721416653561. doi: 10.1177/2333721416653561 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 10.Zimmerman S, Cesari M, Gaugler JE, et al. The Inevitability of Reimagining Long-Term Care. J Am Med Dir Assoc. Feb 2022;23(2):187–189. doi: 10.1016/j.jamda.2021.12.032 [DOI] [PubMed] [Google Scholar]
- 11.Jutkowitz E, Landsteiner A, Ratner E, et al. Effects of Nurse Staffing on Resident Outcomes in Nursing Homes: A Systematic Review. J Am Med Dir Assoc. Jan 2023;24(1):75–81 e11. doi: 10.1016/j.jamda.2022.11.002 [DOI] [PubMed] [Google Scholar]
- 12.Li Y, Yin J, Cai X, Temkin-Greener J, Mukamel DB. Association of race and sites of care with pressure ulcers in high-risk nursing home residents. JAMA. Jul 13 2011;306(2):179–86. doi: 10.1001/jama.2011.942 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 13.Chen M, Goodwin JS, Bailey JE, Bowblis JR, Li S, Xu H. Longitudinal Associations of Staff Shortages and Staff Levels with Health Outcomes in Nursing Homes. J Am Med Dir Assoc. Nov 2023;24(11):1755–1760 e7. doi: 10.1016/j.jamda.2023.04.017 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 14.Shippee TT, Ng W, Duan Y, et al. Changes over Time in Racial/Ethnic Differences in Quality of Life for Nursing Home Residents: Patterns within and between Facilities. J Aging Health. Dec 2020;32(10):1498–1509. doi: 10.1177/0898264320939006 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 15.Geng F, Stevenson DG, Grabowski DC. Daily Nursing Home Staffing Levels Highly Variable, Often Below CMS Expectations. Health Aff (Millwood). Jul 2019;38(7):1095–1100. doi: 10.1377/hlthaff.2018.05322 [DOI] [PubMed] [Google Scholar]
- 16.Bowblis JR, Brunt CS, Xu H, Grabowski DC. Understanding Nursing Home Spending And Staff Levels In The Context Of Recent Nursing Staff Recommendations. Health Aff (Millwood). Feb 2023;42(2):197–206. doi: 10.1377/hlthaff.2022.00692 [DOI] [PubMed] [Google Scholar]
- 17.The Centers for Medicare and Medicaid Services. Payroll-Based Journal (PBJ) Data. Accessed 11/6, 2022. https://www.cms.gov/Medicare/Quality-Initiatives-Patient-Assessment-Instruments/NursingHomeQualityInits/Staffing-Data-Submission-PBJ
- 18.Xu H, Intrator O, Bowblis JR. Shortages of Staff in Nursing Homes During the COVID-19 Pandemic: What are the Driving Factors? J Am Med Dir Assoc. Oct 2020;21(10):1371–1377. doi: 10.1016/j.jamda.2020.08.002 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 19.Bowblis JR, Roberts AR. Cost-Effective Adjustments to Nursing Home Staffing to Improve Quality. Medical care research and review : MCRR. Jun 2020;77(3):274–284. doi: 10.1177/1077558718778081 [DOI] [PubMed] [Google Scholar]
- 20.Xu H, Intrator O, Culakova E, Bowblis JR. Changing landscape of nursing homes serving residents with dementia and mental illnesses. Health Serv Res. Jun 2022;57(3):505–514. doi: 10.1111/1475-6773.13908 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 21.Bowblis JR, Brunt CS, Xu H, Applebaum R, Grabowski DC. Nursing Homes Increasingly Rely On Staffing Agencies For Direct Care Nursing. Health Aff (Millwood). Mar 2024;43(3):327–335. doi: 10.1377/hlthaff.2023.01101 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 22.Xu H, Bowblis JR, Caprio TV, Li Y, Intrator O. Rural-Urban Differences in Nursing Home Risk-adjusted Rates of Emergency Department Visits: A Decomposition Analysis. Med Care. Jan 2021;59(1):38–45. doi: 10.1097/MLR.0000000000001451 [DOI] [PubMed] [Google Scholar]
- 23.National Institutes of Health. Race and National Origin. Accessed 12/25, 2024. https://www.nih.gov/nih-style-guide/race-national-origin#:~:text=BIPOC%20stands%20for%20Black%2C%20Indigenous,solidarity%20between%20communities%20of%20color.
- 24.Bowblis JR, Akosionu O, Ng W, Shippee TP. Identifying Nursing Homes With Diverse Racial and Ethnic Resident Compositions: The Importance of Group Heterogeneity and Geographic Context. Medical care research and review : MCRR. Apr 2023;80(2):175–186. doi: 10.1177/10775587221134870 [DOI] [PubMed] [Google Scholar]
- 25.Bowblis JR, Ng W, Akosionu O, Shippee TP. Decomposing Racial and Ethnic Disparities in Nursing Home Quality of Life. J Appl Gerontol. Sep 2021;40(9):1051–1061. doi: 10.1177/0733464820946659 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 26.Park J, Stearns SC. Effects of state minimum staffing standards on nursing home staffing and quality of care. Health Serv Res. Feb 2009;44(1):56–78. doi: 10.1111/j.1475-6773.2008.00906.x [DOI] [PMC free article] [PubMed] [Google Scholar]
- 27.Shippee TP, Ng W, Bowblis JR. Does Living in a Higher Proportion Minority Facility Improve Quality of Life for Racial/Ethnic Minority Residents in Nursing Homes? Innov Aging. 2020;4(3):igaa014. doi: 10.1093/geroni/igaa014 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 28.Mueller CA, Bowers BJ, Kolanowski AM. RNs in nursing homes-It is not always about the numbers. Journal of the American Geriatrics Society. Aug 2024;72(8):2311–2318. doi: 10.1111/jgs.18898 [DOI] [PubMed] [Google Scholar]
- 29.Beynon C, Supiano K, Siegel EO, Edelman LS, Hart SE, Madden C. Collaboration Between Licensed Nurses and Certified Nurse Aides in the Nursing Home: A Mixed Methods Study. Res Gerontol Nurs. Jan-Feb 2022;15(1):16–26. doi: 10.3928/19404921-20211209-05 [DOI] [PubMed] [Google Scholar]
- 30.Campbell A, Layne D, Scott E. Relational Quality of Registered Nurses and Nursing Assistants: Influence on Patient Safety Culture. Healthcare (Basel). Feb 9 2021;9(2)doi: 10.3390/healthcare9020189 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 31.Bowblis JR, Meng H, Hyer K. The urban-rural disparity in nursing home quality indicators: the case of facility-acquired contractures. Health Serv Res. Feb 2013;48(1):47–69. doi: 10.1111/j.1475-6773.2012.01431.x [DOI] [PMC free article] [PubMed] [Google Scholar]
- 32.Fashaw S, Chisholm L, Mor V, et al. Inappropriate Antipsychotic Use: The Impact of Nursing Home Socioeconomic and Racial Composition. https://doi.org/10.1111/jgs.16316. Journal of the American Geriatrics Society. Mar 2020;68(3):630–636. doi: 10.1111/jgs.16316 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 33.Bowblis JR, Vassallo A. The Effect of Closure on Quality: The Case of Rural Nursing Homes. Journal of Competition Law and Economics. 2014;10(4):909–931. doi: 10.1093/joclec/nhu017 [DOI] [Google Scholar]
- 34.Kang Y, David SV, Bowblis JR, et al. Financial Performance is Associated With PPE Shortages in Chain-Affiliated Nursing Homes During the COVID-19 Pandemic: A Longitudinal Study. Inquiry. Jan-Dec 2023;60:469580231219443. doi: 10.1177/00469580231219443 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 35.American Health Care Association. Long term care industry facing worse job loss among all health care providers. 2022. Accessed 1/20/2022. https://www.ahcancal.org/News-and-Communications/Fact-Sheets/FactSheets/BLS-Report-LTC-Job-Losses.pdf
- 36.Buerhaus PI, Staiger DO, Auerbach DI, Yates MC, Donelan K. Nurse Employment During The First Fifteen Months Of The COVID-19 Pandemic. Health Aff (Millwood). Jan 2022;41(1):79–85. doi: 10.1377/hlthaff.2021.01289 [DOI] [PubMed] [Google Scholar]
- 37.Cantor J, Whaley C, Simon K, Nguyen T. US Health Care Workforce Changes During the First and Second Years of the COVID-19 Pandemic. JAMA Health Forum. Feb 2022;3(2):e215217. doi: 10.1001/jamahealthforum.2021.5217 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 38.Chen M, Goodwin JS, Bailey JE, Bowblis JR, Li S, Xu H. Staff Shortages and Decreased Physical Function Among Nursing Home Residents during the COVID-19 Pandemic. presented at: AcademyHealth 2022 Annual Research Meeting; June 2022; Washington DC. [Google Scholar]
- 39.Brazier JF, Geng F, Meehan A, et al. Examination of Staffing Shortages at US Nursing Homes During the COVID-19 Pandemic. JAMA Netw Open. Jul 3 2023;6(7):e2325993. doi: 10.1001/jamanetworkopen.2023.25993 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 40.Harris E CMS Finalizes Staffing Minimum Requirements for Nursing Homes. JAMA. Jun 11 2024;331(22):1884. doi: 10.1001/jama.2024.8427 [DOI] [PubMed] [Google Scholar]
- 41.The Center for Medicare and Medicaid Services. Minimum Staffing Standards for Long-Term Care Facilities and Medicaid Institutional Payment Transparency Reporting. 2024. Accessed 5/20/2024. https://www.federalregister.gov/documents/2024/05/10/2024-08273/medicare-and-medicaid-programs-minimum-staffing-standards-for-long-term-care-facilities-and-medicaid
- 42.Grabowski DC, Bowblis JR. Minimum-Staffing Rules for U.S. Nursing Homes - Opportunities and Challenges. N Engl J Med. Nov 2 2023;389(18):1637–1640. doi: 10.1056/NEJMp2310544 [DOI] [PubMed] [Google Scholar]
- 43.Bowblis JR, Applebaum R, Brunt CS, Xu H, Grabowski DC. Nursing Home Workforce: The Authors Reply. Health Aff (Millwood). Sep 2024;43(9):1341. doi: 10.1377/hlthaff.2024.00502 [DOI] [PubMed] [Google Scholar]
- 44.Comondore VR, Devereaux PJ, Zhou Q, et al. Quality of care in for-profit and not-for-profit nursing homes: systematic review and meta-analysis. BMJ. Aug 4 2009;339:b2732. doi: 10.1136/bmj.b2732 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 45.Kennedy KA, Applebaum R, Bowblis JR, Straker JK. Organizational Factors Associated With Retention of Direct Care Workers: A Comparison of Nursing Homes and Assisted Living Facilities. Gerontologist. Jun 2 2021;61(4):530–539. doi: 10.1093/geront/gnaa130 [DOI] [PubMed] [Google Scholar]
- 46.Falvey JR, Hade EM, Friedman S, et al. Severe neighborhood deprivation and nursing home staffing in the United States. Journal of the American Geriatrics Society. Mar 2023;71(3):711–719. doi: 10.1111/jgs.17990 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 47.Shippee TP, Davila H, Ng W, et al. Evidence to Inform Policy and Practice: Mechanisms to Address Racial/Ethnic Disparities in Nursing Home Quality of Life. Innov Aging. 2022;6(4):igac037. doi: 10.1093/geroni/igac037 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 48.Morgan B, Foley K, Hodgson N, et al. Revisiting the Teaching Nursing Home: Qualitative Analysis of Nursing Students’ Clinical Experience. Innov Aging. Nov, Dec 20 2022;6(Suppl 1):858. doi: 10.1093/geroni/igac059.3070. eCollection 2022 Nov. [DOI] [Google Scholar]
- 49.American Nurses Credentialing Center. 2020 Pathway to Excellence in Long-Term Care® Practice Standards and Elements of Performance 2020.
- 50.Werner RM, Konetzka RT, Grabowski DC, Stevenson DG. Reforming Nursing Home Financing, Payment, and Oversight. N Engl J Med. May 19 2022;386(20):1869–1871. doi: 10.1056/NEJMp2203429 [DOI] [PMC free article] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Supplementary Materials
Supplemental Figure 1. Consort Selection Diagram for Nursing Homes
Supplemental Figure 2. Distribution of the Racial and Ethnic Composition in Nursing homes, 2019
Supplemental Figure 3. Bivariate Association of Racial and Ethnic Composition with Total Nursing Staff Levels in Nursing Homes, 2019
Data on hours paid are from the Medicare Cost Reports. Data on hours worked are from the Payroll-Based Journal. The red lines represent a simple linear regression between the proportion of Black, Indigenous, and People of Color residents in a facility and staff hours per resident day (hours paid and hours worked). The Pearson correlation coefficient was −0.12 for hours paid and −0.16 for hours worked (both P<0.01).
Supplemental Table 1. State Variations in the Cutoffs of the Top 10% Nursing Homes with the Largest Concentration of Racial and Ethnic Minority Residents, 2019
Supplemental Table 2. Random Effects Regressions Examining the Association of Continuous Racial and Ethnic Composition with Total Nursing Staff Levels
Supplemental Table 3. Random Effects Regressions Examining the Association of Racial and Ethnic Composition with Nursing Staff Levels without Adjusting for Payer Mix