Older Adults With ADRD Likely to Be Discharged to Lower-Quality Nursing Homes

Hospitalized older adults with Alzheimer disease and related dementias (ADRD) are more likely to be discharged to lower-quality skilled nursing facilities compared with patients without ADRD, according to study findings published in JAMA Open Network.

Despite the high number of patients with ADRD in nursing homes, limited attention has been focused on how the quality of care in nursing homes differs among those with and those without ADRD.  It’s estimated that more than 40% of patients in nursing homes have ADRD or some form of cognitive impairment. Over several decades, improvements have been made in the quality of nursing home care, but this has not been the case for individuals with ADRD.

Researchers explored whether individuals with ADRD are systematically admitted to lower-quality nursing homes.

For the study, patients were classified as having ADRD based on the whether the International Statistical Classification of Diseases and Related Health Problems, 10th Revision (ICD-10) code that indicates dementia was listed in any of the 25 discharge diagnosis fields of the hospital claim.

Data were obtained from individual-level, health care facility–level, and community-level sources. The individual-level data were derived from the Medicare Beneficiary Summary File (Medicare enrollment data), the Medicare Provider and Analysis Review (MedPAR) claims files, and Minimum Data Set nursing home resident assessments. Patients’ residences were described with use of the Social Deprivation Index (SDI), which measures economic disadvantage based on an individual’s zip code, to describe patients’ residences. The SDI ratings ranged from 1, which represented the least disadvantaged area, to 100, which represented the most disadvantaged area.

The current study was conducted among Medicare beneficiaries who had been hospitalized from January 2, 2017, through December 31, 2019, with data analysis carried out between January 15, 2022, and May 30, 2022.

All of the study participants had been discharged from a general acute hospital to a Medicare-certified skilled nursing facility. Publicly reported 5-star staffing ratings, grouped into 3 levels, were utilized to rank the quality of all available SNFs, as follows:

• 1 to 2 stars: low-quality rating
• 3-stars: average-quality rating
• 4 to 5 stars: high-quality rating

The key explanatory variable in the conditional logit analysis was quality of a skilled nursing facility. Additional skilled nursing facility attributes were included as covariates, such as:

• occupancy rate,
• number of beds,
• share of Medicare financing,
• level of the therapy staff with a specialization in postacute care,
• for-profit ownership in the skilled nursing facility,
• presence of a dementia special care unit, and
• distance from zip code–tabulated area to the potential skilled nursing facility.

A total of 2,619,464 patients, 22% of whom were diagnosed with ADRD, were included in the full study sample. Overall, mean patient age was 81.3±8.6 years; 61% were women; and 87% were White.

The researchers found that patients with ADRD were admitted more often to larger skilled nursing facilities (mean number of beds, 129.0±75.7), compared with those without ADRD (mean number of beds, 122.7±73.9). Further, individuals with ADRD vs those without ADRD were more likely to be sent to for-profit skilled nursing facilities (71% vs 67%, respectively); less likely to enter hospital-based facilities (3% vs 6%, respectively); and were admitted to skilled nursing facilities with a lower level of postacute care, as confirmed by the level of Medicare financing (mean, 22.5 vs 25.7 share of beds, respectively).

Additionally, patients with ADRD were more likely to enter a skilled nursing facility with a low-quality rating (23% vs 27%, respectively) and were less likely to enter an skilled nursing facility with a high-quality rating (47% vs 53%, respectively).

The log-odds of being discharged to a skilled nursing facility with an average-quality rating was shown to be significantly higher than being discharged to a skilled nursing facility with a low-quality rating (0.23 in individuals with ADRD vs 0.29 in those without ADRD; difference, –0.06; P <.001).

Further, the log-odds of being discharged to a skilled nursing facility with a high-quality rating was significantly higher than being discharged to a skilled nursing facility with a low-quality rating (0.31 in individuals with ADRD vs 0.47 in those without ADRD; difference, –0.16; P <.001). This indicates that although a higher-level rating was linked to an increased odds of skilled nursing facility selection for both patients with ADRD and those without ADRD, the association was stronger for those without ADRD.

The study has several limitations, such as the findings may not be generalizable to certain patient populations, including individuals aged younger than 65, Medicare advantage enrollees, and those who recently had resided within a nursing home — that is, prior to hospitalization. Moreover, the prevalence of ADRD in the study probably represents an underestimation, since confirmation was based on the use of in-hospital billing codes.

“Enhancing incentives to provide postacute care to patients with ADRD may be necessary to improve access to better facilities,” the researchers emphasized, “which may ultimately reduce the importance of ADRD as a marker of disparity.”