This cross-sectional study examines admissions of dually eligible Medicare and Medicaid beneficiaries to high-quality skilled nursing facilities (SNFs) in states that do and do not fully reimburse Medicare SNF coinsurance.
Key Points
Question
Are Medicaid lesser-of rules, which allow lower Medicare cost-sharing reimbursements to health care organizations serving dually eligible beneficiaries, associated with reduced access to high-quality skilled nursing facilities (SNFs)?
Findings
In this cross-sectional study of 5.1 million discharges from SNFs financed by traditional Medicare, differences in facility resource levels and admission rates to high-rated SNFs were 43% and 54% larger, respectively, in states with lesser-of rules compared with states that fully reimburse Medicare SNF coinsurance.
Meaning
The findings support addressing state Medicare spending burden without shifting costs to health care organizations who care for dually eligible individuals to enhance access to care for this population.
Abstract
Importance
Individuals dually eligible for Medicare and Medicaid have historically faced barriers to accessing high-quality skilled nursing facilities (SNFs) for postacute care. One factor potentially contributing to this disparity is that dually eligible individuals yield lower margins than Medicare-only beneficiaries due to the widespread adoption of lesser-of rules by state Medicaid programs; these rules allow states to provide little to no reimbursement for Medicare coinsurance among these patients.
Objective
To examine the association of lesser-of rules with disparities in access to high-quality SNFs between dually eligible and Medicare-only beneficiaries.
Design, Setting, and Participants
This cross-sectional study included discharges from a hospital to an SNF financed by traditional Medicare in the US from January 1, 2018, to December 31, 2022. Discharges of individuals already residing in an SNF before hospitalization were excluded. Data were analyzed between July 1 and December 22, 2025.
Exposure
States’ lesser-of vs full coinsurance payment policy for dually eligible individuals.
Main Outcomes and Measures
Primary outcomes were admitting SNF dependence level on Medicaid long-term care reimbursement, a proxy for facility resources, and admitting SNF 5-star rating. Adjusted mean differences between dually eligible and non–dually eligible patients were calculated and compared across lesser-of and full-payment states via regressions of each outcome on the interaction between patient dual eligibility and state lesser-of status.
Results
The sample included 5 126 814 SNF discharges (mean [SD] patient age, 79.7 [10.6] years; 60% female). In lesser-of states, SNFs that admitted dually eligible individuals demonstrated higher Medicaid dependence compared with facilities admitting Medicare-only beneficiaries (β [SE], 8.51 [0.03] percentage points [pp]; P < .001), a metric that was 43% smaller in full-payment states (β [SE], −3.63 [0.07] pp; P < .001). For dually eligible individuals in lesser-of states, admittance to high-quality SNFs was lower compared with Medicare-only beneficiaries (β [SE], 9.42 [0.07] pp; P < .001), a metric lower by 54% in full-payment states (β [SE], −5.04 [0.16] pp; P < .001).
Conclusions and Relevance
This cross-sectional study found that disparities in access to higher-quality SNFs between dually eligible and Medicare-only beneficiaries were larger when states used lesser-of rules for determining Medicare coinsurance reimbursements. Our findings underscore the need for policies that address states’ Medicare spending without shifting costs to health care organizations serving the vulnerable dually eligible population.
Introduction
Individuals who are dually eligible for Medicare and Medicaid are a high-risk population characterized by high rates of chronic disease, physical and cognitive impairments, and social stressors.1,2,3 Despite their complex care needs, dually eligible individuals have historically faced barriers to accessing high-quality health care. This issue has been well documented in the skilled nursing facility (SNF) setting; dually eligible individuals are more likely to receive care from SNFs with lower staffing levels, more frequent health violations, and an elevated risk for termination from the Medicare and Medicaid programs.4 Although the payer-based disparity in SNF quality is associated with the large differential in Medicaid and private payment rates for long-term care,5 dually eligible individuals are also less frequently admitted to high-quality SNFs for Medicare-financed postacute care.6,7,8,9
Postacute care in SNFs is a high-risk setting where the stakes for delivering quality care are especially raised for clinically complex populations such as dually eligible individuals.10,11 However, scholars have theorized that high-quality, high-demand SNFs may avoid admitting dually eligible beneficiaries because these patients yield lower Medicare margins than their non–dually eligible counterparts.7 A major driver of this cost disparity is Medicaid lesser-of rules, which allow many states to reduce or eliminate Medicare cost-sharing payments for dually eligible individuals despite their nominal responsibility for these charges.
Cost-sharing accounts for a nontrivial share of Medicare SNF revenues. Beginning on the 21st day of an SNF benefit period, traditional Medicare beneficiaries are responsible for a daily coinsurance of $217 (as of 2026; the coinsurance increased from $168 in 2018 to $195 in 2022).12,13 For dually eligible individuals in states with lesser-of rules, however, Medicaid reimburses SNFs for the lesser value of either (1) the $217 daily coinsurance or (2) the Medicaid reimbursement rate for the service minus the amount Medicare has already paid. For example, if the Medicaid per-diem rate for SNF care on day 21 is $320 and Medicare reimburses $300 before beneficiary cost-sharing, the SNF would receive only a $20 coinsurance from Medicaid rather than the full $217 owed by a non–dually eligible traditional Medicare beneficiary. Because Medicaid payment rates are generally low and frequently fall below Medicare rates even without coinsurance, SNFs across 40 states with lesser-of rules incur consistent bad debt from uncollected coinsurance when providing postacute care to dually eligible individuals.4,14,15,16,17,18 Although bad debts are partially recoverable from Medicare, due to several limitations, the reimbursement mechanism is unlikely to offset the financial burden on SNFs.
In this study, we examined the association of lesser-of rules with payer-based disparities in SNF quality. The federal government originally authorized the use of lesser-of rules to help states balance their Medicaid budgets.19 However, we hypothesized that, by paying SNFs less to care for the clinically complex dual-eligible population, these policies may impede access to high-quality care and exacerbate payer-based disparities in states with lesser-of rules compared with states that fully reimburse SNF coinsurances (ie, full-payment states).
Methods
This cross-sectional study was approved by the institutional review board of Brown University, which waived informed consent due to the use of deidentified administrative data. This study followed the Strengthening the Reporting of Observational Studies in Epidemiology (STROBE) reporting guideline.
Data Sources
We combined several sources of administrative data on Medicare beneficiaries and SNFs. Individual- and encounter-level data included the Medicare Beneficiary Summary File, Medicare Standard Analytic Files, Medicare Provider Analysis and Review database, and Minimum Data Set. We supplemented these data with SNF characteristics and quality measures from the Certification and Survey Provider Enhanced Reports (CASPER) database, LTCfocus.org, annual SNF cost reports, and Medicare’s Care Compare website. We used the American Community Survey to obtain zip code tabulation area (ZCTA)–level poverty rates.
Study Population
We identified all hospitalized traditional Medicare beneficiaries directly discharged to an SNF from January 1, 2018, to December 31, 2022. To focus our analyses on postacute care and the start of a new SNF benefit period, we excluded admissions for individuals who were already residing in an SNF during the 60-day period preceding the claim admission date using the residential history file method.20 We also excluded individuals enrolled in Medicare Advantage (MA) due to the lack of complete MA SNF claims data as well as potential differences in SNF benefit structures and the application of lesser-of rules to the MA population. Race and ethnicity were ascertained from the Research Triangle Institute race and ethnicity variable in the Medicare Master Beneficiary Summary File, and were included to characterize the demographic composition of SNFs across the study cohorts. Categories included American Indian or Alaska Native, Asian, Black, Hispanic, White, and other (individuals not represented by the other 5 racial and ethnic cohorts [eg, multiracial individuals]) or unknown. We excluded residents of US territories.
Outcomes
Our primary outcomes were designed to estimate the rate at which dually eligible individuals had access to high-quality, high-resource SNFs. First, the Care Compare overall 5-star rating is a composite measure of SNF staffing levels, inspection results, and performance with respect to multiple patient-level outcomes. Quasi-experimental evidence suggests that admission to facilities with higher star ratings is associated with reduced risk of adverse outcomes.11,21 To enhance the interpretability of our analyses, we created a binary indicator for admission to a 4- or 5-star facility to represent high SNF quality. Second, we used the admitting facility proportion of residents financed by Medicaid recorded in annual CASPER surveys. This measure reflects a facility’s level of dependence on Medicaid long-term care reimbursement and is an established proxy for facility resources.4 As the COVID-19 pandemic prevented consistent CASPER survey collection from 2020 onward, we utilized the Medicaid dependence level recorded in the nearest completed CASPER survey for 2021 and 2022 discharges.
In secondary analyses, we investigated care patterns that may be associated with the reduction in reimbursement generated by lesser-of rules, including rates of within 20-day SNF discharge and 30-day hospital readmission. SNFs in lesser-of states are financially incentivized to discharge dually eligible individuals before cost-sharing begins on the 21st day of care, either to community settings or other facilities such as hospitals.
Main Explanatory Variables
The main explanatory variables were dual eligibility status and residence in a state with a full-payment or lesser-of rule. We assigned dual eligibility based on individuals’ Medicaid eligibility status in the calendar month preceding SNF admission. Because exact eligibility dates are unknown, lagging by 1 month more accurately assesses an individual’s dual eligibility prior to SNF admission. For our primary analyses, we classified individuals as dually eligible regardless of Medicaid benefit levels (ie, whether they had full benefits with comprehensive Medicaid coverage or partial benefits with Medicaid assistance limited to Medicare premiums or cost-sharing). We addressed Medicaid benefit levels in our sensitivity analyses.
Lesser-of status was derived from reports published by the Medicaid and Children's Health Insurance Program Payment and Access Commission in 2018 and 2025.22 During our study period, 9 states (Alabama, Arkansas, Hawaii, Idaho, Maine, New York, South Dakota, Vermont, and Wisconsin) and the District of Columbia had full-payment policies. All other states were considered to have lesser-of status. Over the course of the study, lesser-of status was consistent for all states with the exception of the District of Columbia, which adopted a full-payment policy in 2019 and, accordingly, was classified as having lesser-of status in 2018 only.
Control Variables
Demographic controls were derived from the Medicare Beneficiary Summary File and included age, age squared, sex, race and ethnicity, and current Medicare entitlement reason. To capture differences in case mix, we controlled for index hospital length of stay, intensive care unit stay, dementia diagnosis,23 and diagnosis-related group (DRG) fixed effects, all obtained from the Medicare Provider Analysis and Review file. DRGs are a classification scheme for inpatient stays based on principal diagnosis and are severity weighted according to secondary diagnoses. For certain descriptive purposes, we reported the prevalence of DRGs that commonly precipitate after acute care.
Statistical Analysis
Data were analyzed between July 1 and December 22, 2025. We began our admission-level analyses by graphically presenting crude means of primary outcomes for dually eligible and Medicare-only beneficiaries across neighborhoods, approximated by ZCTAs, in lesser-of and full-payment states. We then estimated adjusted differences in primary and secondary outcomes observed between dually eligible and Medicare-only beneficiaries associated with lesser-of payment policies via ordinary least-squares. Specifically, we regressed each outcome on dual eligibility status, residence in a lesser-of state (reference group) vs a full-payment state, and the interaction between these 2 terms, controlling for the aforementioned covariates and year and state fixed effects. Although state fixed effects were almost perfectly correlated with lesser-of status, our focus was not on the binary lesser-of status coefficient. Instead, our parameters of interest were the coefficients on dual eligibility and the interaction term. Specifically, the dual-eligibility coefficient represents the mean difference in outcomes between dually eligible and Medicare-only SNF residents in lesser-of states, while the interaction term coefficient indicates how payer-based disparities in outcomes differ in full-payment states (ie, a difference in disparities estimate).
We conducted multiple sensitivity analyses. First, we provided estimates excluding state fixed effects. Second, we supplanted state fixed effects with ZCTA fixed effects to specifically assess within-neighborhood payer-based disparities across state types. Third, we limited dually eligible beneficiaries to those in the Qualified Medicare Beneficiary Program, the group consistently covered by full-payment rules. Fourth, we excluded lesser-of states that did not border a full-payment state to enhance comparability. Fifth, we excluded admissions that occurred in 2020 because SNFs were especially affected by the COVID-19 pandemic that year. All analyses were conducted using Stata, version 18.0 (Stata Corp LLC). Null hypotheses were tested assuming a 2-tailed α of .05.
A consequence of lesser-of rules is the generation of bad, or unrecoverable, debt for SNFs providing postacute care to dually eligible individuals. We calculated bad debt burden for all SNFs located in lesser-of states in 2022, defined as total bad debts attributable to dually eligible individuals (specifically itemized in the cost reports) divided by facility bed size from CASPER. We then characterized SNFs according to quartile of bad debt burden, with the highest quartile indicating the greatest bad debt burden.
Results
The sample included 5 126 814 discharges (mean [SD] patient age, 79.7 [10.6] years; 60% female and 40% male) to SNFs between 2018 and 2022. A total of 1% of discharges were of American Indian or Alaska Native individuals; 2%, Asian; 9%, Black; 4% Hispanic; 83%, White; and 2%, other or unknown. Table 1 lists the characteristics of individuals admitted to SNFs in lesser-of and full-payment states by dual eligibility status. In both lesser-of and full-payment states, dually eligible individuals were younger and less likely to be White, had longer hospital stays, and resided in areas with higher poverty rates compared with individuals with Medicare coverage only. Differences in characteristics across payers were broadly comparable in lesser-of and full-payment states. However, compared with dually eligible individuals in lesser-of states, those in full-payment states were closer in age to Medicare-only beneficiaries (mean [SE] difference, −8.5 [0.01] years vs −6.20 [0.03] years), more likely to have dementia (mean [SE] difference, −3.4 [0.1] percentage points [pp] vs 0.5 [0.1] pp), and less likely to be Hispanic (mean [SE] difference, 9.1 [0.0] pp vs 5.3 [0.1] pp).
Table 1. Characteristics of Postacute Care Admissions to Skilled Nursing Facilities in States With Lesser-of and Full-Payment Policies by Dual Eligibility Status, 2018-2022.
| Characteristic | Total discharges, % (N = 5 126 814) | Lesser-of state | Full-payment state | ||||
|---|---|---|---|---|---|---|---|
| Discharges, % | Mean (SE) difference, pp | Discharges, % | Mean (SE) difference, pp | ||||
| Medicare only (n = 3 319 920) | Dually eligible (n = 825 593) | Medicare only (n = 455 343) | Dually eligible (n = 155 708) | ||||
| Age, mean (SD), y | 79.7 (10.6) | 81.3 (9.2) | 72.8 (12.9) | −8.5 (0.0) | 81.7 (9.2) | 75.5 (12.9) | −6.2 (0.0) |
| Sex | |||||||
| Female | 60 | 60 | 61 | 0.8 (0.1) | 60 | 63 | 2.5 (0.1) |
| Male | 40 | 40 | 39 | −0.8 (0.1) | 40 | 37 | −2.5 (0.1) |
| Race and ethnicity | |||||||
| American Indian or Alaska Native | 1 | 0 | 1 | 0.9 (0.0) | 0 | 1 | 0.5 (0.0) |
| Asian | 2 | 1 | 5 | 3.8 (0.0) | 2 | 4 | 1.9 (0.0) |
| Black | 9 | 7 | 18 | 11.5 (0.0) | 7 | 17 | 10.4 (0.1) |
| Hispanic | 4 | 3 | 12 | 9.1 (0.0) | 2 | 7 | 5.3 (0.1) |
| White | 83 | 88 | 63 | −25.5 (0.0) | 87 | 69 | −18.7 (0.1) |
| Other or unknowna | 2 | 2 | 2 | 0.2 (0.0) | 2 | 2 | 0.6 (0.0) |
| OASI entitlement | 93 | 97 | 76 | −21.1 (0.0) | 97 | 81 | −15.7 (0.1) |
| Hospital length of stay, mean (SD), d | 7.6 (8.2) | 7.2 (6.8) | 8.8 (11.2) | 1.6 (0.0) | 8.0 (8.4) | 10.1 (13.7) | 2.1 (0.0) |
| ICU admission | 30 | 31 | 34 | 3.5 (0.1) | 22 | 22 | 0.7 (0.1) |
| Dementia diagnosis | 24 | 25 | 22 | −3.4 (0.1) | 25 | 26 | 0.5 (0.1) |
| Diagnosis-related group | |||||||
| Joint replacement | 6 | 6 | 4 | −2.2 (0.0) | 7 | 4 | −2.5 (0.1) |
| Heart failure | 4 | 4 | 4 | −0.1 (0.0) | 4 | 4 | −0.1 (0.1) |
| Pneumonia | 2 | 2 | 2 | −0.2 (0.0) | 2 | 2 | −0.1 (0.0) |
| COPD | 1 | 1 | 2 | 0.6 (0.0) | 1 | 1 | 0.5 (0.0) |
| Acute kidney injury | 3 | 3 | 3 | 0.2 (0.0) | 3 | 3 | 0.1 (0.0) |
| Urinary tract infection | 3 | 3 | 3 | −0.2 (0.0) | 3 | 3 | 0 (0.1) |
| Stroke | 3 | 3 | 3 | 0 (0.0) | 3 | 3 | −0.1 (0.1) |
| Septicemia | 8 | 8 | 9 | 1.5 (0.0) | 8 | 9 | 1.2 (0.1) |
| Other | 69 | 69 | 69 | 0.3 (0.1) | 69 | 70 | 0.9 (0.1) |
| ZCTA poverty rate, mean (SD) | 12.6 (7.9) | 11.8 (7.4) | 15.8 (9.0) | 4.0 (0.0) | 11.6 (7.4) | 15.3 (8.8) | 3.7 (0.0) |
Abbreviations: COPD, chronic obstructive pulmonary disease; ICU, intensive care unit; OASI, old age and survivor’s insurance; pp, percentage points; ZCTA, zip code tabulation area.
Included individuals not represented by the 5 other racial and ethnic cohorts, such as multiracial individuals, as defined by the Research Triangle Institute racial and ethnic code in the Medicare Master Beneficiary Summary File.
Figure 1 shows the mean Medicaid dependence level for SNFs that admitted dually eligible and Medicare-only beneficiaries in lesser-of and full-payment states, organized by beneficiary neighborhood poverty level. Admitting SNFs for individuals living in higher-poverty areas were more likely to be Medicaid dependent regardless of dual eligibility or lesser-of status. However, the SNFs to which Medicare-only beneficiaries were admitted had lower overall Medicaid-dependence levels than the SNFs to which dually eligible individuals from the same neighborhood were admitted. Although the payer-based disparity in this facility resource measure was present in both lesser-of and full-payment states, the magnitude was smaller in full-payment states. The data on admissions to SNFs with higher star ratings showed a corresponding pattern (Figure 2). Across neighborhoods, dually eligible individuals were less likely to be admitted to SNFs with 4- or 5-star ratings than Medicare-only beneficiaries were. However, payer-based differences in the use of highly rated facilities were larger in lesser-of states compared with full-payment states.
Figure 1. Line Graph of Admitting Skilled Nursing Facility Resource Level and Neighborhood Poverty Level by Dual Eligibility Status in Lesser-of and Full-Payment States.
The means and 95% CIs (shading) were calculated using local polynomial regression. FPL indicates federal poverty level.
Figure 2. Line Graph of Admitting Skilled Nursing Facility Quality Rating and Neighborhood Poverty Level by Dual Eligibility Status in Lesser-of and Full Payment States.
The means and 95% CIs (shading) were calculated using local polynomial regression. FPL indicates federal poverty level.
Table 2 lists adjusted estimates of payer-based disparities in all outcomes derived from state fixed-effects regressions (unadjusted differences are listed in eTable 1 in Supplement 1). In lesser-of states, SNFs admitting dually eligible individuals had overall Medicaid dependence levels that were higher than those admitting Medicare-only beneficiaries (β [SE], 8.51 [0.03] pp; P < .001). In full-payment states, this payer-based difference in admitting facility Medicaid dependence level was 43% smaller (β [SE], −3.63 [0.07] pp; P < .001). In lesser-of states, the adjusted probability of admission to highly rated SNFs was lower for dually eligible individuals compared with Medicare-only beneficiaries (β [SE], −9.42 [0.07] pp; P < .001). In full-payment states, this payer-based disparity was 54% lower (β [SE], 5.04 [0.16] pp; P < .001). Point estimates from analyses that focused on the probability of admission to only 5-star facilities followed a similar pattern (eTable 2 in Supplement 1).
Table 2. Association of Lesser-of Rules With Disparities in Admitting SNF Quality and Patient Outcomesa.
| Outcome | Adjusted mean (SE) | β (SE), ppb | ||||||
|---|---|---|---|---|---|---|---|---|
| Lesser-of state | Full-payment state | Difference between dually vs non–dually eligible beneficiaries in lesser-of states | Difference in payer-based gap between full-payment and lesser-of states | |||||
| Non–dually eligible | Dually eligible | Non–dually eligible | Dually eligible | |||||
| Medicaid dependence level, % | 44.94 (0.07) | 53.45 (0.08) | 45.12 (0.50) | 50.00 (0.50) | 8.51 (0.03) | −3.63 (0.07) | ||
| Probability of admission to 4- or 5-star SNF | 58.54 (0.13) | 49.12 (0.14) | 55.94 (0.86) | 51.56 (0.86) | −9.42 (0.07) | 5.04 (0.16) | ||
| Discharge within 20 d of SNF admission | 51.35 (0.15) | 45.53 (0.15) | 52.02 (0.98) | 44.11 (0.98) | −5.82 (0.07) | −2.08 (0.15) | ||
| 30-d Hospital readmission | 19.24 (0.11) | 19.59 (0.12) | 18.52 (0.75) | 18.06 (0.75) | 0.35 (0.05) | −0.81 (0.12) | ||
Abbreviations: pp, percentage points; SNF, skilled nursing facility.
Results were derived from a regression of given outcomes on dual-eligibility status, residence in a lesser-of or full-payment state, the interaction between dual-eligibility and state payment status, state fixed effects, and other covariates described in the Methods section.
Point estimates were multiplied by 100 to reflect pp differences. P < .001 for all differences.
In lesser-of states, dually eligible individuals were less likely than Medicare-only beneficiaries to be discharged within 20 days of SNF admission (β [SE]; −5.82 [0.07] pp; P < .001). In full-payment states, this disparity was 36% larger (β [SE], −2.08 [0.15] pp; P < .001). Adjusted 30-day readmission rates were modestly higher for dually eligible individuals than Medicare-only beneficiaries in lesser-of states (β [SE], 0.35 [0.05] pp; P < .001). In full-payment states, the payer-based gap in 30-day readmission rates was lower (β [SE], −0.81 [0.12] pp; P < .001). Payer-based differences in readmission rates were larger without covariate adjustment, but removing covariate adjustment did not substantially impact the estimate of disparity between lesser-of and full-payment states (eTable 1 in Supplement 1).
Point estimates derived from ZCTA fixed-effects models were broadly consistent with the primary analysis estimates (eTable 2 in Supplement 1). One difference was that the interaction term coefficient, or difference in size of the disparity estimate between full-payment and lesser-of states, in the model estimating admission to a highly rated SNF was considerably smaller, though still statistically significant. Coefficients from sensitivity analyses were also consistent with the main results (eTable 2 in Supplement 1).
Table 3 shows the characteristics of SNFs in lesser-of states by level of bad debt burden attributable to treating dually eligible patients. SNFs with a higher bad debt burden, and thus more unrecovered coinsurances, were larger, were more likely to be for-profit, had a lower share of private payers, disproportionately served Black residents, and disproportionately served residents with schizophrenia or bipolar disorders. The share of Medicare-financed residents, a proxy for post–acute care specialization, was comparable between SNFs in the highest and lowest bad debt burden quartiles. SNFs with a higher bad debt burden had more deficiency citations in 2022, and 1637 of 3312 facilities in the highest quartile of bad debt (49%) had low star ratings of 1 or 2. Also, 1643 SNFs in the highest-burden quartile (50%) were located in a Southern state, and 1028 (31%) were located in rural counties.
Table 3. Characteristics of SNFs in States With Lesser-of Rules by Bad Debt Burden Attributable to Dually Eligible Individualsa.
| Characteristic | SNFs | ||||
|---|---|---|---|---|---|
| Total (N = 13 251) | Bad debt burden quartileb | ||||
| 1 (n = 5056) | 2 (n = 1570) | 3 (n = 3313) | 4 (n = 3312) | ||
| Bad debt per bed, mean (SD), $ | 596.8 (1041.8) | 0 (2.8) | 66.1 (46.5) | 438.9 (186.5) | 1917.5 (1363.9) |
| Total beds per facility, mean (SD), No. | 104.4 (53.5) | 93.1 (53.7) | 113.5 (53.9) | 110.3 (53.5) | 111.4 (49.8) |
| Occupancy rate, mean (SD), % | 73.1 (17.0) | 72.0 (18.5) | 70.5 (16.9) | 72.3 (16.4) | 76.9 (14.5) |
| For-profit facility, No. (%) | 9488 (72) | 3063 (61) | 1015 (65) | 2465 (74) | 2943 (89) |
| Financing source, mean (SD), % of residents, | |||||
| Medicaid | 59.0 (24.1) | 52.8 (29.0) | 58.0 (21.4) | 62.5 (20.0) | 65.4 (17.7) |
| Medicare | 13.6 (14.1) | 15.1 (17.6) | 11.1 (11.8) | 11.0 (10.6) | 15.2 (11.2) |
| Private insurance | 27.4 (19.8) | 32.1 (24.1) | 30.9 (18.0) | 26.5 (16.3) | 19.4 (12.7) |
| Overall star rating, No. (%) | |||||
| Low (1-2) | 5002 (38) | 1460 (29) | 562 (36) | 1343 (41) | 1637 (50) |
| Average (3) | 2599 (20) | 912 (18) | 304 (20) | 700 (21) | 683 (21) |
| High (4-5) | 5488 (42) | 2596 (52) | 692 (44) | 1239 (38) | 961 (29) |
| Total deficiency citations in 2022, mean (SD), No. | 7.1 (6.3) | 6.9 (6.0) | 6.8 (6.1) | 7.2 (6.5) | 7.3 (6.6) |
| Race and ethnicity, mean (SD), % of residents, | |||||
| Black | 13 (17.8) | 9.1 (14.1) | 10.4 (16.3) | 13.3 (17.8) | 19.8 (21.3) |
| Hispanic | 4.9 (11.0) | 7.4 (12.7) | 6.4 (15.1) | 2.0 (5.4) | 3.2 (8.9) |
| White | 75.9 (23.5) | 75.2 (24.6) | 78 (24.3) | 79.8 (20.6) | 72.2 (23.6) |
| ADRD, mean (SD), % of residents | 44.6 (16.7) | 45.2 (19.4) | 48.5 (15.3) | 44.1 (14.7) | 42.1 (14.0) |
| Schizophrenia or bipolar disorders, mean (SD), % of residents | 14.4 (14.6) | 13.0 (15.9) | 13.3 (13.7) | 14.3 (13.9) | 17.3 (13.3) |
| State region, No. (%)c | |||||
| New England | 713 (5) | 633 (13) | 70 (4) | 0 | 0 |
| Mideast | 1301 (10) | 338 (7) | 177 (11) | 468 (14) | 318 (10) |
| Great Lakes | 2601 (20) | 567 (11) | 352 (22) | 925 (28) | 757 (23) |
| Great Plains | 1871 (14) | 577 (11) | 308 (20) | 691 (21) | 295 (9) |
| Southeast | 3126 (24) | 470 (9) | 163 (10) | 850 (26) | 1643 (50) |
| Southwest | 1645 (12) | 1072 (21) | 315 (20) | 143 (4) | 112 (3) |
| Rocky Mountain | 423 (3) | 170 (3) | 62 (4) | 123 (4) | 68 (2) |
| Pacific | 1574 (12) | 1229 (24) | 123 (8) | 103 (3) | 119 (4) |
| Located in rural county, No. (%)d | 3544 (27) | 1099 (23) | 415 (27) | 1002 (30) | 1028 (31) |
Abbreviations: ADRD, Alzheimer disease and related dementias; SNFs, skilled nursing facilities.
The aggregated facility-level data were derived from cost reports, the Care Compare website, Certification and Survey Provider Enhanced Reporting, and LTCfocus.org.
First quartile, $0 per bed; second quartile, $1 to $156 per bed; third quartile, $157 to $816 per bed; and fourth quartile, $817 to $4551 per bed.
According to the Bureau of Economic Analysis.
Rural classification was based on applying rural-urban continuum codes developed by the US Department of Agriculture to SNF counties.
Discussion
In this study, we found that disparities in access to high-quality SNFs between dually eligible and non–dually eligible Medicare beneficiaries were greater when states used lesser-of rules for determining cost-sharing reimbursements. In states that fully reimbursed SNF coinsurance amounts, payer-based differences in admissions to highly rated facilities were 54% lower and the disparity in admitting facility Medicaid dependence level was 43% lower. These results suggest that SNFs in full-payment states are less segregated by payer mix than those in lesser-of states; this may result in more resources being available to improve care quality and outcomes.
In our analysis of secondary outcomes, we found that dually eligible individuals, who have more complex care needs, were less likely to be discharged within 20 days compared with Medicare-only beneficiaries in all states. However, the size of this disparity was reduced by 36% in lesser-of states compared with full-payment states, suggesting that loss of point-of-service payments for dually eligible individuals on the 21st day of SNF care may result in higher rates of early discharge than may be appropriate in this population. This finding is consistent with past work documenting high rates of discharge before 21 days among vulnerable patient populations.24 Higher disparities in 30-day readmission rates in lesser-of states may also reflect increased incentives to discharge dually eligible individuals earlier.
The federal government officially authorized lesser-of rules in 1997 to assist states with balancing their budgets.19 The results of this study suggest that this practice may negatively impact dually eligible individuals’ access to high-quality care, which can lead to further costs downstream. For SNFs, a direct consequence of lesser-of rules is the accumulation of bad debt from unpaid or underpaid coinsurances. Although bad debts are recoverable through a Medicare reimbursement mechanism, this is unlikely to offset the consequences of lesser-of policies for at least 3 reasons. First, Medicare reimburses only 65% of bad debts, and several proposals have aimed to further reduce or even eliminate bad debt reimbursement entirely.25 Second, because bad debts are reimbursed after the fiscal year ends, SNFs lose revenue at the point of service, constraining their operational capacity. Third, there are administrative costs to seeking bad debt reimbursement from Medicare. Notably, in our study, SNFs most burdened by bad debt had exceedingly low quality ratings, high numbers of deficiency citations, and disproportionately served racial and ethnic minority groups.
Although lesser-of rules have been widely used across states for many years, their role in health care access and delivery has been underresearched, especially outside primary care settings.16,26,27 To our knowledge, this study is the first to focus on lesser-of rules in the SNF setting and extends the literature on payer-based disparities in SNF quality. In past research, low Medicaid long-term care reimbursement rates have been consistently associated with lower SNF quality for dually eligible individuals.4,28,29 Regarding postacute care, scholars have theorized that SNFs potentially avoid admitting dually eligible individuals due to the risk of them becoming long-stay residents, which causes a sharp drop in revenues as the payer switches from Medicare to Medicaid.7,9,30 Our results are consistent with this theory, as dually eligible individuals were admitted to lower-quality SNFs in both lesser-of and full-payment states. However, our analysis identified lesser-of rules as an additional contributor because payer-based disparities were magnified in states with these policies.
Improving health care delivery to dually eligible individuals has been an ongoing policy priority. Contemporary reforms have largely focused on the coordination of Medicaid and Medicare benefits, especially through the MA program. Proposals related to shifting dually eligible individuals’ cost-sharing burden from Medicaid to Medicare, which could address the burden of lesser-of rules, have not yet gained traction.31 Our results underscore the importance of these proposals, particularly because states are unlikely to adopt full-payment policies in the current fiscally austere environment for Medicaid.
Limitations
This study has limitations. First, our results cannot be generalized to the MA program. Although lesser-of rules may impact SNF care for MA enrollees, the mechanisms are less clear due to potential differences in SNF benefit structures and bad debt reimbursement processes across MA plans. Second, the use of lesser-of or full-payment rules may be correlated with other Medicaid policies that are more or less favorable to dually eligible individuals in the SNF setting. Although we cannot rule out this possibility, it is worth noting the geographic diversity of Medicaid programs with full-payment policies. Third, although we used a carefully constructed postacute cohort with controls with strong health status, there may have been unobserved payer-based differences in health status across lesser-of and full-payment states that confounded our cross-sectional estimates.
Conclusions
In this cross-sectional study of 5.1 million discharges to SNFs, compared with Medicare-only beneficiaries, dually eligible individuals in lesser-of states experienced higher rates of admission to lower-quality facilities with fewer resources as well as earlier discharge and modestly increased readmission rates. While lesser-of rules are intended to lower state Medicaid expenditures, this policy may have unintended consequences for dually eligible individuals and ultimately lead to additional costs downstream. These findings underscore the need for policies that address state Medicare spending without shifting costs to the health care organizations serving the vulnerable dually eligible population.
eTable 1. Crude outcome rates in states with lesser-of and full payment policies by dual-eligibility status.
eTable 2. Regression estimates derived from alternative specifications.
Data Sharing Statement
References
- 1.Keohane LM, Zhou Z, Stevenson DG. Aligning Medicaid and Medicare Advantage managed care plans for dual-eligible beneficiaries. Med Care Res Rev. 2022;79(2):207-217. doi: 10.1177/10775587211018938 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 2.Kasper J, O’Malley Watts M, Lyons B. Chronic disease and co-morbidity among dual eligibles: implications for patterns of Medicaid and Medicare Service Use and Spending. KFF. 2010. Accessed December 18, 2025. https://www.kff.org/wp-content/uploads/2013/01/8081.pdf
- 3.Bynum JPW, Austin A, Carmichael D, Meara E. High-cost dual eligibles’ service use demonstrates the need for supportive and palliative models of care. Health Aff (Millwood). 2017;36(7):1309-1317. doi: 10.1377/hlthaff.2017.0157 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 4.Mor V, Zinn J, Angelelli J, Teno JM, Miller SC. Driven to tiers: socioeconomic and racial disparities in the quality of nursing home care. Milbank Q. 2004;82(2):227-256. doi: 10.1111/j.0887-378X.2004.00309.x [DOI] [PMC free article] [PubMed] [Google Scholar]
- 5.Loomer L, Gandhi A, Geng F, Grabowski DC. Secret shopper data on private prices in the nursing home industry from 2008 to 2010. Med Care Res Rev. 2021;78(4):449-457. doi: 10.1177/1077558719879022 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6.Rahman M, Gozalo P, Tyler D, Grabowski DC, Trivedi A, Mor V. Dual eligibility, selection of skilled nursing facility, and length of Medicare paid postacute stay. Med Care Res Rev. 2014;71(4):384-401. doi: 10.1177/1077558714533824 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 7.Rahman M, Grabowski DC, Gozalo PL, Thomas KS, Mor V. Are dual eligibles admitted to poorer quality skilled nursing facilities? Health Serv Res. 2014;49(3):798-817. doi: 10.1111/1475-6773.12142 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 8.Sharma H, Perraillon MC, Werner RM, Grabowski DC, Konetzka RT. Medicaid and nursing home choice: why do duals end up in low-quality facilities? J Appl Gerontol. 2020;39(9):981-990. doi: 10.1177/0733464819838447 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 9.Zuckerman RB, Wu S, Chen LM, Joynt Maddox KE, Sheingold SH, Epstein AM. The five-star skilled nursing facility rating system and care of disadvantaged populations. J Am Geriatr Soc. 2019;67(1):108-114. doi: 10.1111/jgs.15629 [DOI] [PubMed] [Google Scholar]
- 10.Werner RM, Coe N, Qi M, Konetzka RT. The value of an additional day of post-acute care in a skilled nursing facility. Am J Health Econ. 2023;9(1):1-21. doi: 10.1086/721706 [DOI] [Google Scholar]
- 11.Kosar CM, Mor V, Trivedi AN, Rahman M. Impact of skilled nursing facility quality on Medicare beneficiaries with dementia: evidence from vacancies. Alzheimers Dement. 2024;20(12):8308-8316. doi: 10.1002/alz.14251 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 12.2018 Medicare parts A & B premiums and deductibles. Fact sheet. Centers for Medicare & Medicaid Services. November 17, 2017. Accessed August 30, 2026. https://www.cms.gov/newsroom/fact-sheets/2018-medicare-parts-b-premiums-and-deductibles
- 13.2022 Medicare parts A & B premiums and deductibles/2022 Medicare part D income-related monthly adjustment amounts. Centers for Medicare & Medicaid Services. November 12, 2021. Accessed August 30, 2026. https://www.cms.gov/newsroom/fact-sheets/2022-medicare-parts-b-premiums-deductibles-2022-medicare-part-d-income-related-monthly-adjustment
- 14.Nyman JA, Connor RA. Do case-mix adjusted nursing home reimbursements actually reflect costs? Minnesota’s experience. J Health Econ. 1994;13(2):145-162. doi: 10.1016/0167-6296(94)90021-3 [DOI] [PubMed] [Google Scholar]
- 15.Swan JH, Harrington C, Clemeña W, Pickard RB, Studer L, deWit SK. Medicaid nursing facility reimbursement methods: 1979-1997. Med Care Res Rev. 2000;57(3):361-378. doi: 10.1177/107755870005700306 [DOI] [PubMed] [Google Scholar]
- 16.Cabral M, Carey C, Miller S. The impact of provider payments on health care utilization of low-income individuals: evidence from Medicare and Medicaid. Am Econ J Econ Policy. 2025;17(1):106-143. doi: 10.1257/pol.20220775 [DOI] [Google Scholar]
- 17.Buxbaum JD, Ne’eman AD, Kosar CM. Eliminating bad debt reimbursement to hospitals serving traditional Medicare beneficiaries. JAMA Netw Open. 2025;8(8):e2526402. doi: 10.1001/jamanetworkopen.2025.26402 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 18.Williams D, Davlyatov G, Bowblis JR, Braun RT. Eliminating Medicare bad debt payments: are critical access and rural hospitals at risk? Health Aff Sch. 2025;3(11):qxaf220. doi: 10.1093/haschl/qxaf220 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 19.Medicaid and CHIP Payment and Access Commission . Medicaid coverage of premiums and cost sharing for low-income Medicare beneficiaries. 2013. Accessed March 24, 2025. https://www.macpac.gov/publication/ch-4-medicaid-coverage-of-premiums-and-cost-sharing-for-low-income-medicare-beneficiaries/
- 20.Intrator O, Hiris J, Berg K, Miller SC, Mor V. The residential history file: studying nursing home residents’ long-term care histories. Health Serv Res. 2011;46(1 pt 1):120-137. doi: 10.1111/j.1475-6773.2010.01194.x [DOI] [PMC free article] [PubMed] [Google Scholar]
- 21.Cornell PY, Grabowski DC, Norton EC, Rahman M. Do report cards predict future quality? the case of skilled nursing facilities. J Health Econ. 2019;66:208-221. doi: 10.1016/j.jhealeco.2019.05.008 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 22.Medicaid and CHIP Payment and Access Commission . Compendium: state Medicaid payment policies for Medicare cost sharing. 2025. Accessed November 24, 2025. https://www.macpac.gov/publication/compendium-state-medicaid-payment-policies-for-medicare-cost-sharing/
- 23.Chronic Conditions Data Warehouse . 30 CCW chronic conditions (2017 forward). Accessed November 24, 2025. https://www2.ccwdata.org/condition-categories-chronic
- 24.Chatterjee P, Qi M, Coe NB, Konetzka RT, Werner RM. Association between high discharge rates of vulnerable patients and skilled nursing facility copayments. JAMA Intern Med. 2019;179(9):1296-1298. doi: 10.1001/jamainternmed.2019.1209 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 25.Congressional Budget Office . Options for Reducing the Deficit: 2025 to 2034. Congressional Budget Office; December 2024. Accessed February 27, 2025. https://www.cbo.gov/system/files/2024-12/60557-budget-options.pdf
- 26.Fung V, McCarthy S, Price M, et al. Payment discrepancies and access to primary care physicians for dual-eligible Medicare-Medicaid beneficiaries. Med Care. 2021;59(6):487-494. doi: 10.1097/MLR.0000000000001525 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 27.Roberts ET, Nimgaonkar A, Aarons J, et al. New evidence of state variation in Medicaid payment policies for dual Medicare-Medicaid enrollees. Health Serv Res. 2020;55(5):701-709. doi: 10.1111/1475-6773.13545 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 28.Grabowski DC, Angelelli JJ, Mor V. Medicaid payment and risk-adjusted nursing home quality measures. Health Aff (Millwood). 2004;23(5):243-252. doi: 10.1377/hlthaff.23.5.243 [DOI] [PubMed] [Google Scholar]
- 29.Grabowski DC, Angelelli JJ. The relationship of Medicaid payment rates, bed constraint policies, and risk-adjusted pressure ulcers. Health Serv Res. 2004;39(4 pt 1):793-812. doi: 10.1111/j.1475-6773.2004.00258.x [DOI] [PMC free article] [PubMed] [Google Scholar]
- 30.Kosar CM, Mor V, Werner RM, Rahman M. Risk of discharge to lower-quality nursing homes among hospitalized older adults with Alzheimer disease and related dementias. JAMA Netw Open. 2023;6(2):e2255134. doi: 10.1001/jamanetworkopen.2022.55134 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 31.Keohane LM, Hwang A. Payment policy and the challenges of Medicare and Medicaid integration for dual-eligible beneficiaries. Health Aff. Published online October 20, 2022. doi: 10.1377/hpb20220923.93608 [DOI] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Supplementary Materials
eTable 1. Crude outcome rates in states with lesser-of and full payment policies by dual-eligibility status.
eTable 2. Regression estimates derived from alternative specifications.
Data Sharing Statement
