What Do We Know About Hospice Enrollment for Nursing Home Residents with Cognitive Disabilities? A Systematic Review

Abstract

Background:

At least 42% of nursing home (NH) residents live with cognitive disability, commonly associated with neurocognitive disorders (e.g., dementia), defined as difficulty concentrating, learning new things, remembering, or making decisions. The population of individuals with cognitive disabilities who qualify for hospice (end-of-life [EOL] care that reduces symptoms and promotes quality of life) in NHs is increasing. However, little is known about hospice enrollment for this population. This information is essential to clarify disparities in hospice access and guide efforts to improve EOL care for this population.

Objective:

To systematically review available evidence on hospice enrollment among US NH residents with cognitive disabilities.

Methods:

This review followed the Preferred Reporting Items for Systematic Reviews and Meta-Analyses guidelines. A comprehensive search of PubMed, CINHAL, Web of Science, and Google Scholar was conducted to identify studies. Peer-reviewed, English-language articles published before February 28, 2025, were included. The Joanna Briggs Institute Analytical Cross-Sectional Studies Appraisal Tool was used to assess study quality.

Results:

We screened 479 titles and abstracts and reviewed 36 full-text articles. Twelve studies met the inclusion criteria. In seven of the 12 studies, a substantial proportion of NH decedents with cognitive disabilities were not enrolled in hospice. Factors associated with increased likelihood of hospice enrollment at the time of death included female sex (n = 2), White race (n = 2), terminal diagnosis other than cognitive disability (n = 1), and increased prevalence of dementia in NH regardless of the individual’s dementia status (n = 1).

Discussion:

This review is the first to systematically examine hospice enrollment among NH residents with cognitive disabilities. Hospice enrollment varied across studies but was consistently lower for NH decedents with cognitive disabilities compared to those without. Future research should include the broader cognitive disability community to assess hospice enrollment and other EOL outcomes, clarify targets for intervention, and address barriers to care.

Keywords

adult and older adult cognitive disability dementia end-of-life care hospice

Introduction

Cognitive disabilities affect approximately 13.9% of adults in the United States.¹ These disabilities—defined as serious difficulty making decisions, learning new things, remembering, or concentrating—can result from a range of conditions, including Alzheimer’s disease and related dementias (ADRDs), traumatic brain injuries, intellectual/developmental disabilities, and strokes.^1,2 With the exception of individuals with ADRDs, individuals with cognitive disabilities are underrepresented in end-of-life (EOL) and hospice research.^1,3,4

Even among patients with ADRDs, their involvement in research is likely an underrepresentation due to under-detection and/or late diagnosis.⁵ In 2022, the Centers for Medicare and Medicaid Services reported 11.1% of decedents in the United States had an official diagnosis of ADRDs, yet estimates suggest that the real percentage of Medicare decedents with ADRDs is 41%.^3,6,7 This difference indicates that using official diagnoses of ADRDs in research, via diagnostic codes and electronic health records, may result in unintentionally excluding a large segment of the population of individuals with cognitive disabilities.^5,8 The use of the term cognitive disability better captures the population of interest, and therefore improves research on EOL outcomes for a population susceptible to poor/sub-optimal EOL outcomes.^4,9,10

Hospice is a covered benefit under Medicare and the Medicaid benefit in many states that provides specialized EOL care to individuals with an anticipated life expectancy of ≤6 months.^11,12 Interdisciplinary teams that focus on symptom management, comfort, and quality of life deliver hospice care and services.^11–13 These services have reduced unnecessary and burdensome care, improved quality of life for patients and families alike, and reduced health care spending.^12,14,15

While hospice enrollment is recommended for all qualifying patients, 67% of all nursing home (NH) residents who die (i.e., 804,000 individuals in 2023) are not enrolled in hospice.^12,16,17 It is not known how many of these residents have cognitive disabilities; however, conservative estimates suggest that at least 42% of NH residents in the United States have some form of cognitive disability.¹⁸ Given that hospice is considered best practice for EOL care, because it minimizes unwanted treatments and improves quality of life near EOL, it is essential to understand barriers to enrollment and use of high-quality hospice care for this population.^19–21

Given the high prevalence of residents with cognitive disabilities in NHs and the well-known benefits of hospice care,^16–19 it is critical to understand the multilevel factors that influence access in this population. However, little is known about specific resident-level (e.g., diagnosis, race/ethnicity), facility-level (e.g., staffing, ownership), and community-level characteristics (e.g., regional poverty) impacting hospice utilization. This information is essential for clarifying disparities in hospice access and guiding efforts to improve EOL care. In this systematic review, we synthesized the evidence on hospice enrollment for individuals with cognitive disabilities in NHs and the different factors influencing this outcome.

Methods

We developed and registered an a priori protocol in PROSPERO (CRD42024593889) and followed the Preferred Reporting Items for Systematic Reviews and Meta-Analyses guidelines.²² We created a comprehensive search strategy in consultation with a research librarian (see Supplementary Data) and searched four databases (without limits) through February 28, 2025: the Cumulative Index of Nursing and Allied Health Literature (CINAHL), PubMed, Web of Science, and Google Scholar. Studies were eligible for inclusion if they were observational or interventional in design, conducted in the United States, were English language studies published in peer-reviewed journals, included adult (≥18 years) participants receiving care in NH settings, and examined hospice enrollment as an outcome. Eligible studies had to include individuals with cognitive disabilities, which we defined as having a diagnosis of ADRD, stroke, brain injuries, intellectual/developmental disability, or identification of cognitive disability through formal assessment (e.g., cognitive performance scale). Studies were excluded if they were: qualitative; conducted outside the United States; not published in peer-reviewed journals; focused on care settings other than NHs; not inclusive of individuals with cognitive disabilities, or did not assess hospice enrollment.

Article screening

All identified records were uploaded into Covidence²³ for duplicate removal and workflow management. Two reviewers independently screened all titles and abstracts, followed by full-text reviews of potentially eligible studies. Discrepancies were resolved through team discussion.

Critical appraisal of study quality

Study quality was assessed using the Joanna Briggs Institute (JBI) Analytical Cross-Sectional Studies Tool²⁴ (see Supplementary Data), an 8-item checklist evaluating participant selection, the validity of exposures and reliability of outcome measures, approach to addressing confounders, and appropriateness of statistical analysis. Each criterion was scored as “Yes” (1 point), “No” (0 points), “Unclear” (0 points), or “Not Applicable.”^24,25 scores on the tool were summed, and overall quality was calculated as the proportion of “Yes” responses.²⁴ We established ≤50% as the cutoff for “high risk of bias.” Two reviewers independently assessed each study, the results were compared, and any differences were resolved through team discussion.

Data extraction

Data were extracted using a standardized form that was pilot tested prior to use. Extracted variables included: article citation; study aims and outcomes; NH care setting, defined population meeting review criteria for cognitive disability (e.g., ADRD) inclusion criteria (e.g., the definition of cognitive disability, NH care setting); exclusion criteria (e.g., hospice enrollment was not assessed, not meeting population or setting requirements); data source; sample size; data collection period; study location; statistical approach; study findings, including resident demographics (i.e., race, sex, age, primary hospice diagnoses, and comorbidities) and NH-level characteristics (e.g., geographic location). Two authors independently extracted data using the standardized form, and discrepancies were resolved through team discussion. Study data were synthesized using summary tables and grouped by participant diagnosis, sex, setting, and location to facilitate comparisons.

Results

The initial search yielded 479 records after removing duplicates. Following title and abstract screening, 36 full texts were assessed for eligibility, and 12 studies^7,26–35 met all inclusion criteria (Fig. 1). A complete list of excluded studies with reasons for exclusion is provided (See Supplementary Data).

FIG. 1.

Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) 2020 flow diagram for new systematic reviews, which included searches of databases and registers only.

Quality appraisal

All included studies met at least 75% of the JBI quality appraisal criteria as described above and were included in the qualitative synthesis (Table 1). Eleven of the 12 studies met 100% of the criteria.^7,27–36 Four studies were descriptive in nature,^26,27,31,35 and thus received a “Not Applicable” rating for the criterion related to strategies for confounding. One study was downgraded due to uncertainty around valid/reliable outcome measurement.²⁶

Table 1.

Joanna Briggs Institute Checklist

JBI checklist element	Aldridge2022	Kiely2010	Gozalo2015	Li2013	Monroe2013	Gozalo2008	Miller2012	Bartley2023	Reinhardt2020	Miller2010	Unroe2015	Buchanan2004
Analytic Cross-Sectional Studies
Were the criteria for inclusion in the sample clearly defined?	Y	Y	Y	Y	Y	Y	Y	Y	Y	Y	Y	Y
Were the study subjects and setting described in detail?	Y	Y	Y	Y	Y	Y	Y	Y	Y	Y	Y	Y
Was the exposure measured in a valid and reliable way?	Y	Y	Y	Y	Y	Y	Y	Y	Y	Y	Y	Y
Were objective, standard criteria used for measurement of the condition?	Y	Y	Y	Y	Y	Y	Y	Y	Y	Y	Y	Y
Were confounding factors identified?	Y	Y	Y	Y	Y	Y	N/A	Y	Y	Y	N/A	Y
Were strategies to deal with confounding factors stated?	Y	Y	Y	Y	Y	Y	Y	N/A	Y	Y	Y	N/A
Were the outcomes measured in a valid and reliable way?	Y	Y	Y	Y	Y	Y	Y	U	Y	Y	Y	Y
Was an appropriate statistical analysis used?	Y	Y	Y	Y	Y	Y	Y	Y	Y	Y	Y	Y
Total	8/8 100%	8/8 100%	8/8 100%	8/8 100%	8/8 100%	8/8 100%	7/7 100%	6/7 85.7%	8/8 100%	8/8 100%	7/7 100%	7/7 100%

JBI, Joanna Briggs Institute; N/A, not applicable; U, unclear; Y, yes.

Study characteristics

Table 2 reports detailed study characteristics. Studies were published between 2004²⁷ and 2023²⁶ and analyzed data from 1998²⁹ to 2018.²⁶ Authors from six studies used national datasets (i.e., included NHs across all 50 states),^{7,27,28,30–32} and of these, the District of Columbia was included in two.^31,32 Authors from other studies focused on specific geographic regions, including a single safety-net health care system in Indiana,³⁵ nine NHs within a 150 mile radius of Nashville, Tennessee,³³ 22 NHs within a 60 mile radius of Boston, Massachusetts,³⁶ 11 skilled nursing facilities in Rochester, Minnesota,²⁶ a single NH in New York State,³⁴ and an undisclosed number of NHs across the state of Florida.²⁹ Sample sizes ranged from 55³³ to 3,848,114³² corresponding to the variation in geographic location (e.g., larger sample sizes for studies of national datasets).^{7,27,28,30–32}

Table 2.

Study Characteristics for Included Studies

Author, year	Study aim(s)	Data source	Design and statistical approach	Sample, setting, and study period
Aldridge et al, 2022⁷	Examine differences in hospice use patterns between those with comorbid dementia and those with a primary hospice diagnosis of dementia	HRS linked with Medicare claims	Retrospective cohortBivariate analyses via chi-square and t-testsMultivariable logistic regression to estimate adjusted odds ratios between dementia status and each outcomeAssessed potential interaction effects of specific primary hospice diagnoses(cancer and heart failure)and comorbid dementia, and also dementia and low socioeconomic status	Individuals who participated in the HRS study (i.e., US adults aged 51+ years), died between 2004 and 2016, and received hospice care.National Sample ( N = 3123):Female 58.6%NH Resident 43.2%Race: - White 82.5% - Black 11.5% - Hispanic 6.0%
Kiely et al, 2010³⁶	Identify resident and HCP factors associated with hospice referral.Examine the association between the receipt of hospice services and treatment of pain and dyspnea.Examine the association between the receipt of hospice services and the HCP’s evaluation of unmet care needs during the residents’ last 7 days of life	Electronic Health Records	Prospective cohort studyDescriptive information, including means and standard deviations for continuous variables, and counts and percentages for categorical variables.	Individuals who participated in the CASCADE study (i.e., US adults aged 60+ years), and located in one of 22 Boston-area NHs with greater than 60 beds and within a 60-mile radius of Boston, between 2003 and 2007.Boston-area Sample ( N = 259):Female 89.5%NH Resident 100%Race: - White 89.4% - Black 10.2% - Asian 0.3%
Gozalo el al, 2015²⁸	Examine changes in Medicare expenditures in the last year of life that were associated with hospice expansion between 2004 and 2009	CMS hospice claims	Difference-in-differences cross-temporal matching design, with propensity score models	NH decedents in 2004 or 2009, adults aged 67= years, enrolled in Medicare for the last 2 years of life.National Sample ( N = 786,328):NH Resident 100%2004:Female 65.6%Race: - White 89.6% - Nonwhite 10.4% 2009:Female 64.8%Race: - White 88.5% - Nonwhite 11.5%
Li et al,2013³⁰	To describe the longitudinal patterns and the within- and across-facility differences in hospice use and in-hospital deaths between long-term NH decedent residents with and without dementia	MDS, Medicare beneficiary files	Retrospective analyses of secondary datasets.Chi-squared, and logistic regression models	Long-term NH resident (stayed in NH longer than 3 months or whose NH stay was not paid by Medicare), adults aged 65+ years, had at least one comprehensive assessment, were not in a coma, and did not enroll in managed care in the last month of life, between 2003 and 2007.National Sample ( N = 236,619):Female 69.9%NH Resident 100%Race: - White 87.1% - Nonwhite 12.9%
Monroe et al,2013³³	Examine the association between hospice enrollment, dementia severity, and pain among NH residents who died from advanced cancer	MDS, physical charts, and hospice orders	Retrospective cohortt-test, Mann–Whitney tests, and chi-squared tests, also multiple logistic regression	NH resident with dementia and terminal cancer, located in one of nine NHs within 150 miles from Nashville, Tennessee, died from one of the top 10 cancers of the overall population in 2004Tennessee-area Sample ( N = 55):Female 54.5%NH Resident 100%Race: - White 70.9% - Nonwhite 29.1%
Gozalo et al,2008²⁹	To examine the effect of the Medicare hospice benefit on Medicare and Medicaid expenditures by dual-eligible Medicare-Medicaid NH residents	Medicare claims, Medicaid claims, MDS, NH provider, and hospice provider files	Retrospective cohortRegression models with inverse probability of treatment weighting to control for confounding and selection bias	NH resident in the state of Florida who died in a NH or an acute care or hospice inpatient setting after transfer from NH between July and December 1999. Aged 65+ years, and Medicare eligible, as well as Medicaid eligible for at least the last month of life and did not participate in a managed care program.Florida Sample ( N = 5774):Female 66%NH Resident 100%Race: - White 84% - Nonwhite 16%
Miller et al,2012³¹	Compare hospice use and sites of death for NH residents who did and did not receive Medicare SNF care in the last 90 days of life; and, for residents with SNF care in the last 90 days, examine the independent effect of hospice enrollment on the likelihood of hospital death	CMS, MDS, hospice claims, Medicare enrollment data, OSCAR database, ARF (Area Resource File)	Retrospective cohortChi-squared, ANOVA, logistic regression with generalized estimating equations to estimate the independent effects	Medicare beneficiary in a US NH who died in 2006, had advanced dementia, and received SNF care in the last 90 days of life, and died in a hospital.National Sample ( N = 99,370):Female 66%NH Resident 100%Race: - White 84% - Black 12% - Hispanic 3% - Other 1%
Bartley et al,2023²⁶	Compare care delivery in the last year of life for people living with dementia in the community (home or assisted living facilities) vs. those in skilled nursing facilities (SNFs)	Electronic Health Records	Retrospective cohortt-tests, chi-squared tests to compare demographic and clinical variablesKruskal–Wallis tests to compare medians.Univariate and multivariate Poisson regression used to evaluate associations between variables of interest and hospitalizations in the last year of life	Adults age 65+ with a dementia diagnosis, empaneled in a specific health care network, who died in the community or one of 11 SNFs in Rochester, Minnesota from 2013 through 2018Minnesota-area Sample ( N = 1203):Female 66.4%NH Resident 48.3%Race: - White 96.4% - Asian 1.5% - Black 0.8% - Did not disclose 0.2% - Native Hawaiian/Pacific Islander 0.2% - Other 0.7%
Reinhardt,2020³⁴	Identify predictors in hospice use in NHs	MDS, clinical nodes	Retrospective cohortBivariate statistical analyses to examine the associations among study variables.Multiple logistic regression model to identify variables that significantly predicated hospice use.	NH decedent in single hospice in NH in New York State in 2013 to 2014.New York NH Sample ( N = 300):Female 68%NH Resident 100%Race: - White 47% - Black 31% - Latino 21%
Miller at al,2010³²	To document how the use of hospice (i.e., enrollment and length of stay) for US NH residents dying with advanced and mild-to-moderately severe dementia has changed over time	CMS, Medicare claims, MDS	Retrospective cohortLongitudinal analyses used proportions and meansCross-sectional analyses	NH decedent with a diagnosis of dementia in the last 12 months of life, who died from 1999 to 2006 in a Medicare/Medicaid-certified NH.National Sample ( N = 3,848,114):Female 63–73.4%NH Resident 100%Race: - White 85.4–90.5% - Black 6.4–10% - Hispanic 2.3–2.9% - Other 0.8–1.6%
Unroe et al,2015³⁵	To describe characteristics and utilization of hospice among NH and non-NH patients	Medicare and Medicaid claims, MDS	Retrospective cohortDescriptive statistics, chi-squared, and t-tests to compare demographic and clinical characteristics.Utilization outcomes were analyzed using modelsGeneralized linear model using a negative binomial distribution was used for group comparisons of utilization outcomes, costs included means, standard deviations, and medians.Cost comparisons were performed using a Wald test	Hospice patient aged 65+ years between 1999 and 2008, who were served by a safety net health system in IndianaIndiana-area Sample ( N = 3771):Female 58.5%NH Resident 30.3%Race: - White 62.2% - Nonwhite Thirty-seven.8%
Buchanan et al,2004²⁷	Develop and analyze comprehensive profiles of NH residents with end-stage disease in hospice care at admission and compare them to profiles of end-stage residents at admission not in hospice	MDS	Retrospective cohortChi-squared	NH resident with end-stage disease in 2000National Sample ( N = 18,211):Female 53.5–58.5%NH Resident 100%Race: - White 81–85.6% - Black 8.8–11.4% - Hispanic 3.9–4.2% - Asian/Pacific Islander 1.2–2.2% - American Indian/Alaska Native 0.6–1.2%

HRS, Health and Retirement Study; NH, nursing home; HCP, health care provider; CASCADE, Choices, Attitudes, and Strategies for Care of Advanced Dementia at the EOL; CMS, centers for Medicare and Medicaid Services; MDS, Minimum Data Set; SNF, skilled nursing facility; OSCAR, Online Survey, Certification and Reporting database; ARF, Area Resource File.

In nine studies, investigators reported the mean age of participants,^{7,26,28,30,33,34,36} which ranged from 71.8²⁷ to 87.5 years.²⁶ Investigators of all studies reported sex and race; in all studies, females outnumbered males, ranging from 54.5%³³ to 85.4%.³⁶ Reporting of race varied across studies (e.g., multiple racial groups, White vs. nonwhite). White decedents were the majority in all studies, ranging from 47%³⁴ to 97.7%.²⁶

With regards to diagnoses/conditions, in five studies, researchers focused on NH residents with dementia;^{7,26,31,32,36} in one study, researchers focused on NH residents with dementia and terminal cancer.³³ In the remaining six studies, researchers looked at decedents in general.^{27–30,34,35}

Variables measurement

In four studies, investigators used cognitive performance scale scores to determine the presence of, and severity of, cognitive disabilities.^27,28,32,33 Investigators of one study used the global deterioration scale scores to determine the presence of, and severity of, cognitive disabilities.³⁶ In one study, investigators used a clinically validated algorithm based on cognitive and functional assessment scores to determine the severity of cognitive disabilities.⁷ In addition to these assessments, in all 12 studies, investigators used diagnoses via medical charts (i.e., ADRDs) to determine the presence of cognitive disabilities.

Investigators of five studies focused on hospice enrollment for decedents with dementia.^{7,26,31,32,36} In five studies, investigators looked at hospice enrollment for all decedents (i.e., those with and without dementia, as well as other diagnoses).^{28–30,34,35} In one study, investigators examined hospice enrollment for all decedents diagnosed with end-stage diseases, including dementia.²⁷ And in the final study, investigators examined hospice enrollment for individuals with cancer and dementia diagnoses.³³

Study findings

Hospice enrollment

Table 3 reports detailed study findings. In 11 studies, investigators reported the percentage of NH decedents with cognitive disabilities who were enrolled in hospice care;^{7,26–33,35,36} of those, researchers from seven found that fewer than half were enrolled in hospice at the time of death.^7,29–33,36 Investigators from two studies concluded that only a minority of residents with severe cognitive disability received hospice care,^31,36 and researchers from one study found that hospice-enrolled decedents were less likely to have severe cognitive disability.³³ One study³⁴ found a significant association between dementia diagnosis and hospice enrollment, but did not report a specific enrollment percentage. Researchers from all the included studies found that a significant percentage of NH decedents with cognitive disabilities were not enrolled in hospice at the time of their deaths.

Table 3.

Key Findings

Author, publication date	Key findings
Aldridge et al., 2022⁷	45% of hospice enrollees have dementia.Almost one-third of individuals who enroll in hospice are suffering from comorbid dementia, while dying of another serious illness.In adjusted analyses, those with comorbid dementia had higher odds than those with no dementia of enrolling with hospice for more than six months (aOR = 1.52; 95% CI: 1.11–2.09) and lower odds of three or fewer days in hospice (aOR = 0.73; 95% CI: 0.56–0.94).Having a primary diagnosis of dementia was associated with increased odds of hospice enrollment greater than six months, hospice disenrollment, and hospice disenrollment following six months of hospice.Comorbid dementia was associated with increased odds of hospice enrollment greater than six months, and hospice disenrollment following six months of hospice.
Kiely et al., 2010³⁶	Hospice was provided to 22.3% of all residents over 18 months, and 29.9% of those who died.A minority of NH residents with advanced dementia received hospice care.Factors associated with hospice referral included: non-white, eating problems, health care proxy’s perception of the resident’s prognosis was <6 months, and better health care proxy mental health.20.5% of residents were referred to hospice.The rate of hospice utilization observed in this NH population (22.3%) supports the widely held view that dementia patients are largely under-severed by hospice services.
Gozalo et al., 2015²⁸	36.63% of NH residents with dementia enrolled in hospice.63.Thirty-seven% of NH residents with dementia did not receive hospice services.
Li et al., 2013³⁰	36.5% of residents with dementia were enrolled in hospice. 34.4% of residents without dementia were enrolled in hospice. 35.7% of total residents were enrolled in hospice.Hospice use for decedents with dementia increased from 25.1% to 36.5% from 2003 to 2007, compared to an increase from 26.5% to 34.4% for decedents without dementia.
Monroe et al., 2013³³	45% of residents with cognitive impairment who died from cancer were in hospice at EOL.Residents in hospice were less likely to show severe cognitive impairment (20% vs. 50%, p = 0.50).CPS scores for NH residents enrolled in hospice were statistically significantly lower on average than those for residents not enrolled in hospice (hospice mean 3.00, DC 1.35) (not in hospice mean 4.Thirty-seven, SD 1.71).50% of those not enrolled in hospice showed severe cognitive impairment (CPS 5–6) whereas just 20% of those in hospice showed that level of impairment.None of the NH residents with very severe cognitive impairment (CPS =6) were enrolled in hospice.
Gozalo et al., 2008²⁹	44% of long-stay NH residents with dementia received hospice. 43% of long-stay NH residents with dementia did not receive hospice.45% of short-stay NH residents with dementia received hospice. 47% of short-stay NH residents with dementia did not receive hospice.Severe cognitive impairment, female gender, and older age were associated with higher hospice-related expenditures in long-stay residents with non-cancer diagnoses.
Miller et al., 2012³¹	70% of residents did not get hospice but had severe impairment.24% of residents on hospice had terminal diagnosis documented, whereas 14% of those not on hospice had a terminal diagnosis documented.Higher proportions of non-hospice residents had a cognitive performance scale score (CPS) of 6-more severe cognitive disability.
Bartley et al., 2023²⁶	56.5% of dementia patients in SNF received hospice, 43.5% did not receive hospice.Home/ALF residents were more likely to be enrolled in hospice than those living in SNFs (68.3% vs. 56.5%).Median time in hospice care was 30.0 days for those living at home/ALFs and 26.5 days for those living in SNFs.
Reinhardt, 2020³⁴	Poorer cognitive status, weight loss, and having “goals of care” conversation were significantly correlated with being on hospice in NHs.Dementia was significantly correlated with hospice enrollment. (r = 0.19; p < 0.001)Those with poorer cognitive status were 1.5x more likely to be on hospice than those with better cognitive status. (aOR 1.55 [1.03. 2/32]; p < 0.05)
Miller et al., 2010³²	42.5% of residents with advanced dementia were enrolled in hospice. Thirty-seven.9% of residents with mild-to-moderate dementia were enrolled in hospice.Hospice access for NH decedents with advanced dementia tripled between 1999 and 2006; 14.5% in 1999 and 42.5% in 2006. Rates also tripled for decedents with mild-to-moderate dementia, from 13.2% in 1999 to Thirty-seven.9% in 2006.In both dementia groups, residents with NH hospice were more frequently female, non-Hispanic white, and higher proportions had higher levels of education, cancer diagnoses, and other cardiovascular disease, renal failure, do-not-resuscitate or do-not-hospitalize orders, and long NH stays.In both groups residents with additional diagnoses of cancer or other cardiovascular diseases were more likely to access hospice compared to those without these diseases.
Unroe et al., 2015³⁵	82.2% of NH hospice enrollees had dementia.Only 24.1% of NH hospice enrollees had dementia as their primary hospice diagnosis.NH hospice patients were older, and more likely to be women, have dementia, and be dual-eligible (Medicare and Medicaid).
Buchanan et al., 2004²⁷	64.1% of residents enrolled in hospice had mild to very severe impairment.Hospice residents at admission were more likely to be female, older, white, and widowed than other end-stage residents at admission.Hospice residents at admission were significantly more likely to have poorer cognitive function than other end-stage residents at admission.At least 51% of hospice residents had at least moderate cognitive impairment at admission, compared to 45% of other end-stage residents.

aOR, adjusted odds ratio; CI, confidence interval; CPS, Cognitive Performance Scale; ALF, assisted living facility.

Growth in hospice enrollment

Investigators from three studies examined changes in hospice enrollment over time among NH decedents with dementia.^28,30,32 Gozalo et al. (2015) reported a 5.2% increase in hospice use from 2004 to 2009 for NH decedents with dementia, reporting a decline in enrollment for decedents without dementia. Li et al. (2013) also found an increase in hospice enrollment from 2003 to 2007 for NH decedents with dementia (11.4%), compared to those without dementia (7.9%). Miller, Lima, & Mitchell (2010) reported that hospice use tripled from 1999 to 2006 for residents with mild-to-moderate dementia.

Factors associated with hospice enrollment

Findings reported across studies focused mainly on patient-level factors associated with hospice enrollment. Investigators of two studies identified significant associations between hospice enrollment and specific demographic characteristics; in both, White race and female sex were positively associated with hospice use.^27,32 Investigators of one study found that NH decedents with additional diagnoses of cancer or cardiovascular disease were more likely to enroll in hospice compared to those without these comorbidities.³² Additionally, investigators in one study found that decedents enrolled in hospice were more likely to have a formal terminal diagnosis documented in their medical records than those who were not enrolled in hospice.³¹ These individuals had “terminal” documented in their medical charts, rather than alternative options for hospice enrollment like “estimated less than 6 months life-expectancy.” Only one study team (Li et al., 2013) reported factors associated with hospice enrollment at the NH facility level (i.e., prevalence of dementia among residents; positive association with enrollment).

Discussion

To our knowledge, this is the first systematic review to examine hospice use among US NH decedents with cognitive disabilities. While this population extends beyond those with formally diagnosed ADRDs, we found no studies assessing hospice use among NH residents with other important cognitive disability subtypes (e.g., intellectual or developmental disabilities, and acquired brain injuries). Notably, none of the studies included in this review used the term cognitive disability. Given the heightened risk of poor EOL outcomes among this population,^4,9 future research should aim to include a wider range of cognitive conditions to ensure more inclusive and representative findings.

We expanded upon prior research by synthesizing findings related to factors associated with hospice enrollment in NH settings. A few of the studies included in this review identified/reported the association of facility-level factors (e.g., size, ownership, and rurality) with hospice use, warranting further investigation in this area. The literature we synthesized was limited to NH residents with dementia and is thus limited in its generalizability across NH residents with other cognitive disabilities. However, our findings show that hospice enrollment was consistently lower for NH decedents with cognitive disabilities compared to those without, underscoring the critical need for exploration of the barriers and facilitators affecting access/utilization for this population.^7,26–36

As of 2010, data show that only 6% of NH decedents were enrolled in hospice care, despite 24% being eligible.³⁷ While this statistic does not specify cognitive disability status, or the multiple factors that impede access, it highlights a wider gap between eligibility and enrollment observed nationally among NH decedents. Public awareness and perceptions of hospice have evolved since this time; by 2017, 70% of Americans had heard of hospice, and 85% of those endorsed a positive view.³⁸ Given that the studies in this review span from 2004 to 2023, it is possible that hospice enrollment patterns for individuals with cognitive disabilities have shifted over time, and this may be influenced by multiple factors (e.g., evolving NH resident demographics).

Researchers from six studies in this review used national datasets,^{7,27,28,30–32} while others focused on specific regions or health systems.^29,33,34 Although geographic variation in hospice referral is well-documented, including disparities both within and across states,³⁹ none of the studies in this review specifically analyzed regional differences in hospice enrollment. Such variation may reflect differences in the availability of health care resources (i.e., distance to and concentration of health care facilities), NH practices and programs that impact hospice access, racial and ethnic composition, and socioeconomic status.³⁹ Importantly, none of the studies examined the impact of community-level factors, such as regional poverty. Understanding these and other sources of disparities is essential to improving equitable hospice access for NH residents with cognitive disabilities and should be the focus of future research.

Although all included studies focused on individuals with ADRDs, some researchers used broader cognitive assessments beyond diagnostic codes. Still, no studies specifically addressed important cognitive disability subtypes, such as intellectual/developmental disabilities or acquired brain injuries. While we sought to identify research that included patients with a variety of cognitive disabilities, this proved more difficult than anticipated. All data identified and included in this review focused on individuals with ADRDs. Given the heightened risk of poor EOL outcomes among this population, future research should aim to include a wider range of cognitive conditions to ensure more inclusive and representative findings.

No interventional studies were identified in this review. As the population continues to age and efforts to address health disparities advance, future research should prioritize developing and testing interventions that improve hospice enrollment among individuals with cognitive disabilities. This population already experiences significant disparities in health care access and quality across the care continuum,^40–44 yet data on their use of hospice remain scarce. This review highlights the need for a deeper understanding of hospice care disparities affecting individuals with cognitive disabilities and the development of targeted strategies to address them.^4,9,40

Additionally, hospice care delivered in other settings (i.e., hospitals, home, assisted living facilities, and group homes) should be examined, as many deaths occur outside of NHs, and many patients with ADRD receive EOL care in assisted living facilities. Furthermore, in the spirit of the principle, “nothing about us, without us,”⁴⁵ individuals affected by cognitive disabilities should be meaningfully engaged in the research process, not only as participants but also as stakeholders and collaborators.

Limitations and potential biases

The definition of cognitive disability used in this review was developed by the authors and, while intended to be comprehensive, may have excluded studies of this population using varied definitions. This review was limited to English language, peer-reviewed studies conducted in US NHs, which may have excluded important research from other settings (e.g., those who died in the hospital following hospice care in NHs, participants from other countries). Additionally, all included studies focused on individuals with ADRDs, limiting the generalizability of the findings to the broader population with cognitive disabilities.

Conclusions

Disparities in EOL care exist for individuals with cognitive disabilities, who are less likely to receive hospice services compared to those without such disabilities. This review found that, among US NH residents, individuals with cognitive disabilities are less likely to be enrolled in hospice than their peers without cognitive disabilities. As this population ages and continues to grow, it is vital to understand and address the underlying mechanisms of reduced hospice enrollment and identify novel solutions to address it. Improving hospice access can ensure the delivery of quality EOL care, reduce unnecessary health care utilization and costs, and alleviate distress for both patients and family caregivers.

This review examines hospice enrollment of people with cognitive disabilities. Seven out of 12 articles identified that a significant number of NH residents with cognitive disabilities were not enrolled in hospice at the time of death. Results show that hospice enrollment varied among NH residents with cognitive disabilities, and that race/ethnicity, sex, and facility-level characteristics may influence access to hospice care.

Footnotes

Author Disclosure Statement

The authors listed above have No Interest to disclose.

Funding Information

V.M.W. is supported by the Columbia University School of Nursing Systems Science and Comparative and Cost-Effectiveness Research Training for Nurse Scientists program (S2CER2; T32 NR014205).

Supplemental Material

References

DHDS. Centers for Disease Control and Prevention, National Center of Birth Defects and Developmental Disabilities, Division of Human Development and Disability. Disability and Health Data System (DHDS) Data. December 18, 2024. Available from: https://www.cdc.gov/dhds/datasets/index.html [Last accessed: March 15, 2024].

Consumer and Governmental Affairs Bureau. Individuals with Cognitive Disabilities: Barriers to and Solutions for Accessible Information and Communication Technologies. Federal Communications Commission, 2016:1–37. Available from: https://docs.fcc.gov/public/attachments/DOC-341628A1.pdf

HHS. Cognitive Impairment: A Call for Action, Now! U.S. Department of Health and Human Services, Centers for Disease Control and Prevention. 2011:1–4. Available from: https://www.cdc.gov/aging/pdf/cognitive_impairment/cogimp_poilicy_final.pdf

Johansson

. Time to increase the involvement of persons with intellectual disability in palliative care and research. Evid Based Nurs, 2021; 24(2):48; doi: 10.1136/ebnurs-2019-103188

Chen

, Power

, Grodstein

, et al. Correlates of missed or late versus timely diagnosis of dementia in healthcare settings. Alzheimer’s &Amp; Dementia, 2024:alz.14067; doi: 10.1002/alz.14067

CMS. Centers for Medicare & Medicaid Services. Chronic Conditions Data Warehouse: Alzheimer’s Disease Disparities in Medicare Fee-For-Service Beneficiaries. January 2022. Available from: https://www2.ccwdata.org/web/guest/medicare-tables-reports

Aldridge

, Hunt

, Husain

, et al. Impact of comorbid dementia on patterns of hospice use. J Palliat Med, 2022; 25(3):396–404; doi: 10.1089/jpm.2021.0055

Bradford

, Kunik

, Schulz

, et al. Missed and delayed diagnosis of dementia in primary care: Prevalence and contributing factors. Alzheimer Dis Assoc Disord, 2009; 23(4):306–314; doi: 10.1097/WAD.0b013e3181a6bebc

Voss

, Loxton

, Anderson

, Watson

. “It was one of those complicated cases”: Health practitioners’ perspectives and practices of providing end-of-life care for people with profound intellectual and multiple disability. BMC Palliat Care, 2021; 20(1):177; doi: 10.1186/s12904-021-00873-5

10.

Landes

, Wilmoth

, McDonald

, Smith

. Racial-ethnic inequities in age at death among adults with/without intellectual and developmental disability in the United States. Prev Med, 2022; 156:106985; doi: 10.1016/j.ypmed.2022.106985

11.

Dhingra

, Lipson

, Dieckmann

, et al. Institutional special needs plans and hospice enrollment in nursing homes: A national analysis. J Am Geriatr Soc, 2019; 67(12):2537–2544; doi: 10.1111/jgs.16103

12.

Miller

, Lima

, Gozalo

, Mor

. The growth of hospice care in U.S. nursing homes. J Am Geriatr Soc, 2010; 58(8):1481–1488; doi: 10.1111/j.1532-5415.2010.02968.x

13.

Bekelman

, Black

, Shore

, et al. Hospice care in a cohort of elders with dementia and mild cognitive impairment. J Pain Symptom Manage, 2005; 30(3):208–214; doi: 10.1016/j.jpainsymman.2005.04.003

14.

Candy

, Holman

, Leurent

, et al. Hospice care delivered at home, in nursing homes and in dedicated hospice facilities: A systematic review of quantitative and qualitative evidence. Int J Nurs Stud, 2011; 48(1):121–133; doi: 10.1016/j.ijnurstu.2010.08.003

15.

Zheng

, Mukamel

, Friedman

, et al. The effect of hospice on hospitalizations of nursing home residents. J Am Med Dir Assoc, 2015; 16(2):155–159; doi: 10.1016/j.jamda.2014.08.010

16.

AMA. Hospice and Palliative Care. American Medical Association; 2017:1–2. Available from: https://www.ama-assn.org/sites/ama-assn.org/files/corp/media-browser/public/about-ama/councils/Council%20Reports/council-on-medical-service/issue-brief-hospice-palliative-care.pdf

17.

Dying in America: Improving quality and honoring individual preferences near the end of life. National Academies Press, 2015:18748; doi: 10.17226/18748

18.

Mukamel

, Saliba

, Ladd

, Konetzka

. Dementia care is widespread in US nursing homes; Facilities with the most dementia patients may offer better care. Health Aff (Millwood), 2023; 42(6):795–803; doi: 10.1377/hlthaff.2022.01263

19.

Miller

, Mor

, Wu

, et al. Does receipt of hospice care in nursing homes improve the management of pain at the end of life? J Am Geriatr Soc, 2002; 50(3):507–515; doi: 10.1046/j.1532-5415.2002.50118.x

20.

Munn

, Hanson

, Zimmerman

, et al. Is hospice associated with improved end-of-life care in nursing homes and assisted living facilities? J Am Geriatr Soc, 2006; 54(3):490–495; doi: 10.1111/j.1532-5415.2005.00636.x

21.

Unroe

, Sachs

, Dennis

, et al. Effect of hospice use on costs of care for long‐stay nursing home decedents. J Am Geriatr Soc, 2016; 64(4):723–730; doi: 10.1111/jgs.14070

22.

Page

, McKenzie

, Bossuyt

, et al. The PRISMA 2020 statement: An updated guideline for reporting systematic reviews. BMJ, 2021; 372:n71; doi: 10.1136/bmj.n71

23.

Covidence systematic review software. 2025. Available from: www.covidence.org

24.

Moola

, Munn

, Tufanaru

, et al. Chapter 7: Systematic reviews of etiology and risk. In: Joanna Briggs Institute Reviewer’s Manual. The Joanna Briggs Institute; 2017.

25.

Phillips

, Barker

. Systematic reviews: Structure, form and content. J Perioper Pract, 2021; 31(9):349–353; doi: 10.1177/1750458921994693

26.

Bartley

, Manggaard

, Fischer

, et al. Dementia care in the last year of life: Experiences in a community practice and in skilled nursing facilities. J Palliat Care, 2023; 38(2):135–142; doi: 10.1177/08258597221125607

27.

Buchanan

, Choi

, Wang

, Ju

. End-of-Life care in nursing homes: Residents in hospice compared to other end-stage residents. J Palliat Med, 2004; 7(2):221–232; doi: 10.1089/109662104773709341

28.

Gozalo

, Plotzke

, Mor

, et al. Changes in medicare costs with the growth of hospice care in nursing homes. N Engl J Med, 2015; 372(19):1823–1831; doi: 10.1056/NEJMsa1408705

29.

Gozalo

, Miller

, Intrator

, et al. Hospice effect on government expenditures among nursing home residents. Health Serv Res, 2008; 43(1 Pt 1):134–153; doi: 10.1111/j.1475-6773.2007.00746.x

30.

, Zheng

, Temkin-Greener

. Quality of end-of-life care of long-term nursing home residents with and without dementia. J Am Geriatr Soc, 2013; 61(7):1066–1073; doi: 10.1111/jgs.12330

31.

Miller

, Lima

, Looze

, Mitchell

. Dying in U.S. Nursing homes with advanced dementia: How does health care use differ for residents with, versus without, end-of-life medicare skilled nursing facility care? J Palliat Med, 2012; 15(1):43–50; doi: 10.1089/jpm.2011.0210

32.

Miller

, Lima

, Mitchell

. Hospice care for persons with dementia: The growth of access in US nursing homes. Am J Alzheimers Dis Other Demen, 2010; 25(8):666–673; doi: 10.1177/1533317510385809

33.

Monroe

, Carter

, Feldt

, et al. Pain and hospice care in nursing home residents with dementia and terminal cancer. Geriatr Gerontol Int, 2013; 13(4):1018–1025; doi: 10.1111/ggi.12049

34.

Reinhardt

. What predicts hospice use in the nursing home? J Soc Work End Life Palliat Care, 2020; 16(3):286–293; doi: 10.1080/15524256.2020.1800553

35.

Unroe

, Sachs

, Dennis

, et al. Hospice use among nursing home and non-nursing home patients. J Gen Intern Med, 2015; 30(2):193–198; doi: 10.1007/s11606-014-3080-x

36.

Kiely

, Givens

, Shaffer

, et al. Hospice use and outcomes in nursing home residents with advanced dementia. J Am Geriatr Soc, 2010; 58(12):2284–2291; doi: 10.1111/j.1532-5415.2010.03185.x

37.

Monroe

, Carter

. Hospice care in US nursing homes: Benefits and barriers. Eur J Palliat Care, 2010; 17(3):144–149.

38.

Hamel

, Wu

, Brodie

. Views and Experiences with End-of-Life Medical Care in the U.S. the Kaiser Family Foundation; 2017. Available from: https://files.kff.org/attachment/Report-Views-and-Experiences-with-End-of-Life-Medical-Care-in-the-US

39.

Tsui

, Hirsch

, Bayer

, et al. Patterns in geographic access to health care facilities across neighborhoods in the united states based on data from the national establishment time-series between 2000 and 2014. JAMA Netw Open, 2020; 3(5):e205105; doi: 10.1001/jamanetworkopen.2020.5105

40.

Burke

, Heller

. Disparities in unmet service needs among adults with intellectual and other developmental disabilities. J Appl Res Intellect Disabil, 2017; 30(5):898–910; doi: 10.1111/jar.12282

41.

Harrington

, Kang

. Disparities in service use and expenditures for people with intellectual and developmental disabilities in California in 2005 and 2013. Intellect Dev Disabil, 2016; 54(1):1–18; doi: 10.1352/1934-9556-54.1.1

42.

Havercamp

, Scandlin

, Roth

. Health disparities among adults with developmental disabilities, adults with other disabilities, and adults not reporting disability in North Carolina. Public Health Rep, 2004; 119(4):418–426; doi: 10.1016/j.phr.2004.05.006

43.

Horner-Johnson

, Dobbertin

, Lee

, Andresen

, Expert Panel on Disability and Health Disparities. Disparities in chronic conditions and health status by type of disability. Disabil Health J, 2013; 6(4):280–286; doi: 10.1016/j.dhjo.2013.04.006

44.

Reichard

, Stolzle

, Fox

. Health disparities among adults with physical disabilities or cognitive limitations compared to individuals with no disabilities in the United States. Disabil Health J, 2011; 4(2):59–67; doi: 10.1016/j.dhjo.2010.05.003

45.

Harpur

, Stein

. The Convention on the Rights of Persons With Disabilities as a Global Tipping Point for the Participation of Persons With Disabilities. In: Oxford Research Encyclopedia of Politics. Oxford University Press; 2017; doi: 10.1093/acrefore/9780190228637.013.245

Supplementary Material

Please find the following supplemental material available below.

For Open Access articles published under a Creative Commons License, all supplemental material carries the same license as the article it is associated with.

For non-Open Access articles published, all supplemental material carries a non-exclusive license, and permission requests for re-use of supplemental material or any part of supplemental material shall be sent directly to the copyright owner as specified in the copyright notice associated with the article.

0.12 MB

0.00 MB