Patterns of Home and Community Based Service Use by Beneficiaries Enrolled in the Pennsylvania Medicaid Aging Waiver

Abstract

Background:

This study examines multiple services are used across a population and the association between type and amount of services use with level of disability and living arrangement.

Methods:

This is a descriptive cross-sectional analysis examining HCBS use among older Pennsylvanians from 2014 to 2016 enrolled in Pennsylvania’s 1915(c) waiver program. Data were derived from Medicaid claims. Logistic regression and OLS regression were used to examine the association between service use and level of disability, controlling for age, gender, race, and other covariates.

Results:

People with Alzheimer’s or a related dementia were more likely to use adult day care. People with higher ADL and IADL limitations were more likely to use higher amounts of PAS and less likely to have delivered meals.

Conclusions:

These findings demonstrate HCBS is a complex package of services that are allocated regarding the level of need and resources available to individual program participants.

Keywords

long term services and supports home and community based services Medicaid

What this paper adds

Home and Community Based Services (HCBS) do not necessarily function in a linear capacity. As a person’s level of need increases, the necessary types and amounts of services best suited to meet that person’s needs may change.

Attendant care, both service directed and consumer directed, may be a complement for some forms of HCBS like home delivered meals and in other instances may be a substitute for some forms of HCBS.

The Expanded Andersen Model can be used to better understand how older adults use HCBS.

Applications of study findings

Attendant care trends with level of disability, and a diagnosis of Alzheimer’s or a related Dementia greatly increases average attendant care a person receives.

Increases in a person’s limitations with regards to activities of daily living is associated with a lower likelihood that a person will go to an adult day care center.

Older adults living with children or other family members were less likely to use adult day care, suggesting that the family member might be providing some additional care.

Background

Creating long-term care programs that support aged and disabled people living independently requires attention to the unique needs of the disabled individual (Kane, 2001). Older adults prefer to live independently, and as the population ages the demand for home and community based services (HCBS) will continue to grow (Ng, et al., 2014; Ng, et al., 2015). The primary goal of HCBS is to enable people with physical and cognitive disabilities and limitations to age in their own homes rather than a nursing home (Duckett & Guy, 2000). HCBS can assist people with accomplishing activities of daily living, managing of medications, emotional and social support, and environmental modifications in the individual’s home. In order to support a person living independently, HCBS must compensate for the loss of functioning experienced with the onset of disability (Verbrugge & Jette, 1994). This support must be tailored to the unique health needs, home setting, and personal preferences of the individual (Vertrees, et al., 1989; H. Degenholtz, et al. 1997; Kane, et al., 1998).

There is a growing body of literature that examines outcomes associated with specific types of HCBS. Personal care, also known as attendant care or Personal Assistive Services (PAS) has demonstrated the ability to preserve an older adults’ autonomy and reduce the risk hospitalization (Cohen, 1988; Townsend, et al., 1988; Doty, et al., 2010). Recent studies have found that providing meals to HCBS beneficiaries can delay placement in a nursing home (Thomas & Mor, 2013; Sahyoun & Vaudin, 2014; Campbell, et al., 2015). Research on adult day care is mixed. Although studies have found high levels of user satisfaction (Ellen, et al., 2017) the evidence that adult day care can prevent or delay nursing home placement is less clear (Gaugler, Jarrott et al. 2003; Gaugler, Kane et al. 2003; Gaugler, Kane et al. 2005).

Although HCBS waiver programs operate in all US states, there is limited research on the type and amount of services actually delivered under these programs. Each state Medicaid program determines what services are covered, and there is no consistent national data source despite significant effort (Peebles & Bohl, 2014). In addition, administrative claims data do not include measures of physical or cognitive function, thus without additional data, it is impossible to determine if the distribution of services is appropriately allocated (Davidson, Moscovice, and McCaffrey, 1989).

This paper addresses the gap in the HCBS literature by examining HCBS claims data from a large state Medicaid program, linked with comprehensive assessment data. In particular, we examine the association between use of PAS, adult day care, and home delivered meals and living arrangement, physical function, and cognitive impairment.

Conceptual Framework

Disability is the permanent loss of physical function or onset of a cognitive limitation that interferes with individuals’ ability to independently meet their daily needs (Verbrugge & Jette, 1994). Bradley et al. proposed a comprehensive model for understanding use of long-term services and supports based on the behavioral model of health services use (Bradley, et al., 2002). That model links need (disability), enabling factors (availability of support and financial resources), with service use. In the context of Medicaid HCBS waiver programs, this process is mediated by care managers who allocate publicly financed LTSS to address participant needs. Specifically, Medicaid HCBS waiver programs are an alternative to nursing home placement for individuals who require assistance to compensate for loss of physical or cognitive function. Since Medicaid programs have strict financial eligibility requirements, the emphasis is in non-financial resources that can meet individual needs. In order to identify participant needs, care managers use comprehensive assessments to document physical and cognitive function and availability non-financial resources (e.g., family or other informal supports). Care managers rely on these comprehensive assessments to allocate services to individuals based on need with the goal of preventing or delaying nursing home placement, but also avoid supplanting the role of informal supports (Doty, 1986). In addition, care managers must incorporate participant preferences into the care planning process (Degenholtz 2021). For example, participants may vary in terms of their willingness to use adult day care or delivered meals. Thus, the patterns of service use can be understood as the result of the assessment and care management process, filtered through participant preferences.

This conceptual model suggests that service use should be positively correlated with disability, but negatively correlated with non-financial resources. However, it is not clear that different types of HCBS will have the same directional relationships. This study provides empirical data on the pattern of use that can be used to further refine the theory.

Methods

This study was a pooled cross-sectional analysis examining HCBS use among community-dwelling disabled older Pennsylvanians from 2014 to 2016 who were enrolled in the 1915(c) Medicaid aging waiver program.

Data

The data for this analysis came from Pennsylvania Medicaid enrollment and administrative claims files. Comprehensive assessments of functional status were obtained from the Department of Aging. The enrollment data were used to determine when a person began and ended his or her participation in the waiver program. The enrollment data also included information regarding a person’s race, ethnicity, gender, age, and the zip code and county where the person lived. The unit of analysis was the person-month. The span of our observation went from July 2014 to December 2016, so one person could have a maximum of 30 person-months and a minimum of one person-month. In order to have a person-month included in the analysis, that person-month had to have a claim for at least one of the following HCBS categories.

Claims data were used to measure the three most common service categories: Personal Attendant Service (PAS), Home Delivered Meals, and Adult Day Care. Each service category is billed using different units which were aggregated to the person-month level. PAS is billed in 15-minute increments; these bills were aggregated to calculate the average number of minutes per person per day of PAS used by each person each month. Home delivered meals were used by a small proportion of participants, and meal providers were inconsistent in the way they documented the number of “units” billed. Thus, we constructed an indicator of whether the participant received any delivered meals in a given month, rather than the number of meals. Similarly, the proportion of participants using adult day care services was relatively low, we constructed a binary indicator for any adult day care center use during a given month.

Drawing on Bradley et al. {Bradley}, measures of need and non-financial resources (availability of family support) were extracted from comprehensive assessments conducted by the Area Agencies on Aging (AAA). The comprehensive assessment is used to determine if an individual is eligible for Medicaid HCBS Waiver services and is repeated annually or if there is a change in the participant’s health or functional status has changed. From these data we extracted measures of need (continence, physical function, and cognitive function). To construct a measure of continence, we examined indicators of bladder control, bowel control, and, if applicable, the ability to manage an ostomy bag. When examining activities of daily living (ADL), we counted limitations regard eating, bathing, toileting, transfer, walking indoors, and dressing. The instrumental activities of daily living (IADL) measured were housework, walking outside, managing money, using a telephone, preparing meals, and shopping. The measurements of continence, ADLs, and IADLs were weighted based on a magnitude estimation score and then converted into a zero to ten scale (Finch, et al., 1995; Barnato, et al., 2011). A zero indicates no limitation in a particular domain, and a 10 indicates complete assistance required for all aspects of the domain. We also included indicators of Parkinson’s disease, stroke, both conditions which are associated with significant levels of dependence that may not be captured by IADL and ADL measures. Finally, the measurement of cognitive impairment was limited to a diagnosis of Alzheimer’s Disease or other form of Dementia (ADRD)

As a measure of non-financial support in the form of informal, unpaid caregiving, we include information about the individual participant’s living arrangement. Specifically, we include indicators for living alone, with an adult child, another adult, spouse, or with another family member. Nearly all people who were married were living with their spouse, but to be counted as living with a spouse a person had to indicate that he or she was living with his or her spouse. If a person was married but classified him or herself as living with a child, other family member, or other adult, we classified that person as living with that other adult, rather than with their spouse.

Finally, we include indicators for the region of the state and whether individuals lived in a rural or urban community. Rurality was defined by the National Center for Health Statistics Urban-Rural Continuum Codes (RUCC) that classifies counties based on their population and their proximity to an economic center (Hall, et al., 2006). To capture regional differences within Pennsylvania, we also included indicators for living in the Southwest Pennsylvania (Pittsburgh and surrounding counties), Southeast Pennsylvania (Philadelphia and surrounding Pennsylvania counties), or in the primarily rural central and northern counties.

Analysis

Univariate analysis was performed to gain a picture of the demographic variation, differences in disability levels, and the differences in living arrangements of the enrollees in the aging waiver. We also examined the distribution of service use by unique individuals and at the person-month level.

Multivariate analyses were performed to examine the association between demographic characteristics and disability and use of PAS, adult day care services, and home delivered meals. Use of PAS was measured as a continuous variable so the level of association between PAS use and individual characteristics is evaluated using ordinary least squares. Use of adult day care and meals were analyzed using logistic regression. All regressions used sandwich estimators to control for multiple observations on the same individuals.

Since assessments are conducted annually, we also included a measurement of the number of months enrolled in the waiver program to control for other time related factors that might be affecting a person’s service use.

Results

The total number of unique individuals was 36,193 (see Table 1). The total number of person-months observed was 578,318. The population observed was 72.8% female, 48.8% were non-Hispanic Whites, and 89.1% lived in an urban area. The largest group (56%) lived in the Central or Northern Pennsylvania. More than one fifth (22%) of participants had a diagnosis of Alzheimer’s Disease or a related Dementia, and 20% had experienced a stroke. Only 3% had been diagnosed with Parkinson’s Disease. The majority of the population either lived alone (43%) or lived with a child (22%), 15% of people lived with another family member, 13% lived with a spouse, and 8% lived with another person.

Table 1.

Characteristics of Older Medicaid Waiver Participants (2014–2016).

Variable	Unique Individuals (n=36,205)		Person-Months (n=578,318)
Variable	n	% or Mean (SD)	n	% or Mean (SD)
Age	36,193	78 (9.2)	578,318	78 (9.2)
Gender
Female	26,353	72.8%	427,075	73.8%
Male	9840	27.2%	151,243	26.0%
Race/Ethnicity
Non-Hispanic white	17,687	48.8%	273,745	48.6%
Non-Hispanic black	11,783	32.6%	196,799	34.0%
Hispanic	566	1.6%	8965	1.5%
Asian	2895	8.0%	48,118	8.3%
Other	3262	9.0%	50,691	8.8%
Disability level
Continence	3.7	3.3	3.7	3.3
Instrumental activities of daily living	3.5	1.6	3.4	1.6
Activities of daily living	6.3	2.4	6.3	2.4
Cognitive function
Alzheimer’s disease or related dementia	8210	22.6%	127,075	22%
Stroke	8210	19.8%	113,491	20%
Parkinson’s Disease	1182	3.3%	18,792	3%
Living situation
Lives alone	14,947	41.1%	246,293	43%
Lives with child	8092	22.4%	128,261	22%
Lives with another person	3094	8.6%	45,842	8%
Lives with spouse	4525	12.5%	72,804	13%
Lives with other family	5436	15.1%	85,118	15%
Geographic region
Southwest	5879	16.2%	90,704	15.7%
Central and northern	20,412	56.4%	341,989	59.1%
Southeast	9902	27.4%	145,625	25.2%
Location
Urban	32,255	89.1%	518,922	89.7%
Rural	32,255	10.9%	59,396	10.3%

The majority of person-months (97%) had some PAS use (Table 2). There were 8.5% of person-months that had a claim for adult day care and there were 28% of person-months that had a claim for a home delivered meal. To get an understanding of service use among people who do not use PAS, we examined the 15,570 person months that did not use any PAS. In this group, the majority of person-months (10,779) had home delivered meals and 5268 of the used adult day care.

Table 2.

Distribution of HCBS Service Use.

Services Use	Unique Individuals		Person- Months
Services Use	n	%	n	%
Minutes of PAS per day
None	1666	4.6	15,570	2.7
0 to 239	19,802	55.2	289,838	50.1
240 to 479	11,542	32.2	214,784	37.1
480 +	2876	8.0	58,126	10.1
Home delivered meals per week
None	25,568	71.3	8766	71.8
At least one meal	10,318	28.7	163,343	28.2
Adult day care per week
None	33,292	92.8	529,224	91.5
At least 1 day	2594	7.2	49,094	8.5

Exactly half (50%) of the population used up to 4 hours of PAS per day, while 37% used between four and 8 hours of PAS per day and the remaining 10% of the population used over 8 hours of PAS per day.

PAS Use. Compared to men, women used on average 11 more minutes more of PAS per day (p <0.001) (Table 3). Compared to non-Hispanic Whites, non-Hispanic Blacks, and Asian Americans used more minutes of PAS per day (17.5; p <0.001 and 16.9; p < .001). Compared to people living in Southwestern Pennsylvania, people in Central and Northern Pennsylvania on average used 45 minutes more PAS per day (p <0.001) while people living in the Southeast on average used almost 16 fewer minutes of PAS per day (p <0.001). People with a diagnosis of Alzheimer’s Disease or a related Dementia used on average 41 more minutes of PAS per day (p < 001). People with a diagnosis of Parkinson’s Disease used on average 45 more minutes of PAS per day (p<.001). People who were married and lived with their spouse used, on average 50 fewer minutes per person day of PAS (p<.001) compared to people who lived alone. When compared to people who lived alone, people who lived with an adult child used on average about 29 fewer minutes of PAS per person day (p<.001).

Table 3.

Participant Characteristics Associated with use of HCBS.

	Personal Attendant Care (Hours per Person per Day)		Adult Day Care(% Any Use)			Home Delivered Meals(% Any Use)
	Coefficient	P>t	Odds Ratio	95% Conf. Int.		Odds Ratio	95% Conf. Int.
Gender (male)
Female	11.31	<0.001	0.93	0.90	0.95	0.82	0.81	0.83
Race (non-Hispanic white)
Non-Hispanic bBack	17.51	<0.001	0.47	0.46	0.48	1.42	1.39	1.44
Hispanic	−5.23	0.006	0.56	0.50	0.62	1.18	1.12	1.24
Asian	16.88	<0.001	2.00	1.94	2.06	0.17	0.16	0.18
Other race	18.12	<0.001	0.40	0.38	0.42	1.05	1.03	1.08
Age category (60–-65)
65 to 69	−9.10	<0.001	1.51	1.41	1.62	0.90	0.88	0.93
70 to 74	−14.07	<0.001	2.11	1.97	2.25	0.83	0.81	0.86
75 to 79	−15.38	<0.001	3.07	2.88	3.28	0.68	0.66	0.70
80 to 84	−11.99	<0.001	3.10	2.90	3.30	0.70	0.68	0.72
85+	1.82	0.082	2.93	2.74	3.12	0.63	0.62	0.65
Disability level
Continence*	4.41	<0.001	1.01	1.01	1.01	1.01	1.01	1.01
Activities of daily living*	17.54	<0.001	0.91	0.91	0.92	0.97	0.97	0.97
Instrumental activities of daily living*	23.50	<0.001	1.15	1.14	1.16	0.90	0.89	0.90
Alzheimer’’s disease or related dementia	40.84	<0.001	2.04	1.99	2.09	0.98	0.96	1.00
Stroke	−1.35	0.049	0.99	0.97	1.02	1.09	1.07	1.11
Parkinson’s Disease	44.93	<0.001	0.96	0.90		1.02	1.13	1.09
Living situation (lives alone)
Lives with a spouse	−50.06	<0.001	1.30	1.26		1.35	0.58	0.57
Lives with child	−28.49	<0.001	0.97	0.94		1.00	0.56	0.55
Lives with another family member	−21.73	<0.001	0.94	0.90		0.97	0.55	0.53
Lives with another person	−13.80	<0.001	1.17	1.12		1.22	0.64	0.62
Geography (urban)
Rural	−8.95	<0.001	0.45	0.42		0.49	1.35	1.32
Region (southwestern PA)
Southeastern PA	−16.48	<0.001	6.26	5.99		6.55	0.45	0.44
Northern and central PA	45.23	<0.001	0.82	0.78		0.87	0.72	0.70
Months observed	1.13	<0.001	1.01	1.01		1.01	1.00	0.99
Constant	57.13	<0.001	0.01	0.01		0.01	2.21	2.14

Score ranges from 0 (no limitation) to 10 (complete assistance required.

Adult Day Care. When examining use of adult day care, women were 7% less likely to use adult day care (OR 0.93, p< 0.001, 95% CI [0.90, 0.95]) compared to men. People who lived in rural areas were 55% less likely to use adult day care when compared to people living in urban areas (OR 0.45, p< 0.001, 95% CI [0.42, 0.49]). Non-Hispanic Blacks and Hispanic people were 53% (OR 0.47, p<0.001, 95% CI [0.46, 0.48]) and 44% (OR 0.56, p<0.001, 95% CI [0.50, 0.62]), respectively, less likely to use adult day care compared to white people. Asian Americans were twice (OR 2.0, p<0.001, 95% CI [1.94, 2.06]) as likely to use adult day care compared to non-Hispanic Whites.

A person’s ADL, IADL, and continence levels were significantly associated with use of adult daycare (Table 3). A 10% increase in continence was associated with a 1% (OR 1.01, p<0.001, 95% CI [1.007, 1.014]) increase in the likelihood of using adult daycare. A 10% increase in IADL limitation was associated with a 15% increase in likelihood of using adult daycare (OR 1.15, p<0.001, 95% CI [1.14, 1.16]). However, a 10% increase in a person’s ADL level was associated with a 9% decrease in a person’s likelihood of using adult daycare (OR 0.91, p<0.001, 95% CI [0.909, 0.919]). People diagnosed with Alzheimer’s disease or a related Dementia were twice as likely to use adult day care (OR 2.04, p<0.001, 95% CI [1.99, 2.09]).

Home Delivered Meals. When examining use of home delivered meals, women were 18% (OR 0.82, p<0.001, 95% CI [0.81, 0.83]) less likely to receive home delivered meals compared to men. Asian Americans were much less likely to receive home delivered meals when compared to non-Hispanic Whites (OR 0.17, p<0.001, 95% CI [0.16, 0.18]). People of Hispanic ethnicity and non-Hispanic Blacks were 41% (OR 1.41, p<0.001, 95% CI [1.39, 1.44]) and 18% (OR 1.18, p<0.001, 95% CI [1.12, 1.24]), respectively, more likely to use home delivered meals compared to Non-Hispanic White people. People who lived in a rural area were 35% (OR 1.35, p<0.001, 95% CI [1.32, 1.38]) more likely to use home delivered meals compared to people living in urban areas.

A 10% increase in ADL and IADL levels were associated with a 3% (OR 0.97, p<0.001, 95% CI [0.967, 0.973]) and 10% (OR 0.90, p<0.001, 95% CI [0.895, 0.904]) lower likelihood of using home delivered meals. However, people who were diagnosed with Alzheimer’s Disease or a related dementia no more or less likely to use home delivered meals (OR 0.98, p<0.001, 95% CI [0.96, 1.00]) compared to people who did not have a diagnosis of Alzheimer’s Disease or a related dementia. People living with a spouse were 42% (OR 0.58, p<0.001, 95% CI [0.57, 0.59]) less likely to use home delivered meals than people living alone. People living with an adult child were 44% (OR 0.56, p<0.001, 95% CI [0.55, 0.57]) less likely to use home delivered meals than people living alone. People living with another family member were 45% (OR 0.55, p<0.001, 95% CI [0.53, 0.56]) less likely to use home delivered meals than people living alone (see Table 3).

Discussion

During the years 2014–2016, Pennsylvania Medicaid served an average of 35,737 older adults per at a total cost that nearly doubled from $667 hundred million to over $1 billion in 2016 {Department of Human Services, 2022 https://www.dhs.pa.gov/contact/DHS-Offices/Pages/OLTL-Workbook.aspx}. The program does not have a waiting list. This study focused on a sample of 36,205 unduplicated individuals. The objective was to examine how services were being used across the population and investigate to see if certain services were used more or less in specific subgroups. Our findings are generally consistent with our conceptual model that HCBS is associated with level of disability, but does not generally supplant family caregiving. Although we were not able to measure unpaid caregiving directly, the presence of another person (either a spouse or adult child) was also highly associated with service use. On average people living with a spouse used 50 fewer minutes of PAS every day compared to people living alone. Similarly, the association between ADL and IADL limitation and PAS use is evidence that the HCBS program is adjusting the level of services based on participant need. Future research should examine whether the amount of PAS is associated with important outcomes such as hospitalization or nursing home placement.

Although higher levels of disability were associated with lower use of home delivered meals, we should note that this effect was stronger with regard to IADLs (which includes meal preparation). It is possible that this represents a tradeoff between delivered meals and PAS. That is, PAS use increases with level of disability, potentially obviating the need for a prepared meal.

Older adults who lived with either their children or other family members, were less likely to go to an adult day center, adjusting for physical and cognitive function. This category of caregiver is likely to be working age, and could likely benefit from the respite provided by using adult day service (Gaugler, Jarrott, et al., 2003). The explanation for this difference is not something we can discern from our data. Future research should examine whether this pattern is common in other HCBS programs, as it suggests differential use of adult day service with particular types of family members, for example, by considering the heterogeneity of contributions of paid and unpaid caregivers to daily living.

Overall, these findings reveal a complex set of associations between disability and HCBS use. Presence of cognitive disability or higher levels of physical disability were not associated with higher likelihood of using all types of HCBS. This was particularly the case for adult day care and home delivered meals. These findings clearly demonstrate that the process by which services are allocated to individuals is not simple but takes into account multiple factors. The conceptual framework proposed by Bradley et al. may need to be extended to take into account the characteristics of different types of HCBS.

Policy makers should consider whether adult day services and home delivered meals are being used to the fullest extent possible to meet the needs of program participants. There is evidence of some tradeoff between these services and PAS; however, adult day can provide benefits of socialization for people that might otherwise be socially isolated. This may be particularly acute for people who live alone. Care managers should consider how to use these services in concert with one another. This study did not examine wages or other factors that affect the workforce. However, an important avenue for future research is to consider whether the availability of PAS workers limits the amount of care hours that people receive. Finally, future research should examine the role of individual and cultural preferences as well as possible barriers to specific in the use HCBS.

Limitations

There are several limitations that should be acknowledged. First, these data are from one state, and may not generalize to other Medicaid HCBS programs. Pennsylvania is a large state with a mature waiver program, with both rural and urban communities, that ranks near the center on the AARP rebalancing scorecard (Peebles & Bohl, 2014). Nevertheless, other states that differ in the generosity of their HCBS programs might have different associations between need and service use. Second, we did not have a detailed picture of the social support system of people in our sample. We used living arrangement as a proxy for informal and family care giving, which represents the potential availability of unpaid caregiving. This is imprecise, because participants may have unpaid caregivers who do not live with them. The pattern of findings is consistent, however, with the assumption that people who live alone have less access to or use of unpaid caregiving. Second, we restricted the analysis to three categories of HCBS based on the prevalence of these services. There are other services that HCBS users may value, but are used infrequently, leading to sparse data. Finally, our data do not include individual preferences, or peer, community, and cultural values that may affect an individuals’ decisions around use of HCBS. Future research should investigate methods to include important constructs such as preferences, satisfaction, and unmet need in order to produce a more complete picture of HCBS use.

Conclusion

This paper highlights some important takeaways about HCBS use across a state Medicaid population. The first is that different factors are associated with use of different categories of HCBS. As people become more disabled, they tend to use more PAS, but higher levels of disability are not necessarily associated with using more adult day care or home delivered meals. Similarly, living with other people is associated with using less PAS, but higher likelihood of adult day care, a service that should be a relief to caregivers. Demographic characteristics like race, ethnicity, and gender are known to play important roles in how people interact with the healthcare system and in this analysis, we did see that these characteristics were associated with the type and amount of service a person received but overall, there was no consistent pattern of service use. This paper lays the foundation for further research on how people use HCBS and how these services can potentially allow a person to remain independent in their communities.

Footnotes

Declaration of conflicting interests

The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The author(s) disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: Health Policy Institute at the University of Pittsburgh.

ORCID iDs

Raymond Van Cleve

Howard Degenholtz

References

Barnato

A. E.

Albert

S. M.

Angus

D. C.

Lave

J. R.

Degenholtz

H. B.

(2011). Disability among elderly survivors of mechanical ventilation. American Journal of Respiratory and Critical Care Medicine, 183(8), 1037–1042. https://doi.org/10.1164/rccm.201002-0301OC.

Bradley

E. H.

McGraw

S. A.

Curry

Buckser

King

K. L.

Kasl

S. V.

Andersen

(2002). Expanding the Andersen model: The role of psychosocial factors in long-term care use. Health Services Research, 37(5), 1221–1242. https://doi.org/10.1111/1475-6773.01053.

Campbell

A. D.

Godfryd

Buys

D. R.

Locher

J. L.

(2015). Does participation in home-delivered meals programs improve outcomes for older adults? Results of a systematic review. Journal of Nutrition in Gerontology and Geriatrics, 34(2), 124–167. https://doi.org/10.1080/21551197.2015.1038463.

Cohen

E. S.

(1988). The elderly mystique: Constraints on the autonomy of the elderly with disabilities. The Gerontologist, 28(Suppl), 24–31. https://doi.org/10.1093/geront/28.suppl.24.

Davidson

Moscovice

McCaffrey

(1989). Allocative efficiency of case managers for the elderly. Health Services Research, 24(4), 539.

Degenholtz

Kane

R. A.

Kivnick

H. Q.

(1997). Care-related preferences and values of elderly community-based LTC consumers: Can case managers learn what's important to clients?. The Gerontologist, 37(6), 767–776. https://doi.org/10.1093/geront/37.6.767.

Degenholtz

H. B.

(2021 Jan-Feb). The Importance of Ascertaining Participant Preferences: The Importance of What Is Important. Journal of Gerontological Social Work, 64(1), 8–20. https://doi.org/10.1080/01634372.2020.1864799.

Doty

(1986). Family care of the elderly: the role of public policy. Milbank Q, 64(1), 34–75. https://doi.org/10.2307/3350005.

Doty

Mahoney

K. J.

Sciegaj

(2010 Jan-Feb). New state strategies to meet long-term care needs. Health Affairs, 29(1), 49–56. https://doi.org/10.1377/hlthaff.2009.0521.

10.

Duckett

M. J.

Guy

M. R.

(2000). Home and community-based services waivers. Health Care Financing Review, 22(1), 123–125. https://www.ncbi.nlm.nih.gov/pubmed/25372343

11.

Ellen

M. E.

Demaio

Lange

Wilson

M. G.

(2017). Adult day center programs and their associated outcomes on clients, caregivers, and the health system: A scoping review. The Gerontologist, 57(6), e85–e94. https://doi.org/10.1093/geront/gnw165.

12.

Finch

Kane

R. L.

Philp

(1995). Developing a new metric for ADLs. Journal of the American Geriatrics Society, 43(8), 877–884. https://doi.org/10.1111/j.1532-5415.1995.tb05530.x.

13.

Gaugler

J. E.

Jarrott

S. E.

Zarit

S. H.

Stephens

M. A.

Townsend

Greene

(2003). Respite for dementia caregivers: the effects of adult day service use on caregiving hours and care demands. International Psychogeriatrics, 15(1), 37–58. https://doi.org/10.1017/s1041610203008743.

14.

Gaugler

J. E.

Kane

R. L.

Kane

R. A.

Clay

Newcomer

(2003). Caregiving and institutionalization of cognitively impaired older people: utilizing dynamic predictors of change. The Gerontologist, 43(2), 219–229. https://doi.org/10.1093/geront/43.2.219.

15.

Gaugler

J. E.

Kane

R. L.

Kane

R. A.

Newcomer

(2005). Early community-based service utilization and its effects on institutionalization in dementia caregiving. The Gerontologist, 45(2), 177–185. https://doi.org/10.1093/geront/45.2.177.

16.

Hall

S. A.

Kaufman

J. S.

Ricketts

T. C.

(2006). Defining urban and rural areas in US epidemiologic studies. Journal of Urban Health, 83(2), 162–175. https://doi.org/10.1007/s11524-005-9016-3.

17.

Kane

R. A.

(2001). Long-term care and a good quality of life: Bringing them closer together. The Gerontologist, 41(3), 293–304. https://doi.org/10.1093/geront/41.3.293.

18.

Kane

R. A.

, et al (1998). The heart of long term care. Oxford University Press, USA.

19.

Harrington

Musumeci

(2014). Medicaid Home and Community-Based Services Programs: 2010 Data Update. The Kaiser Commission on Medicaid and the Uninsured.

20.

Stone

Harrington

(2015). Medicaid home and community-based services: How consumer access is restricted by state policies. Journal of aging & social policy, 27(1), 21–46. https://doi.org/10.1080/08959420.2015.969078.

21.

Peebles

Bohl

(2014). The HCBS taxonomy: a new language for classifying home-and community-based services. Medicare & medicaid research review, 4(3). https://doi.org/10.5600/mmrr.004.03.b01.

22.

Sahyoun

N. R.

Vaudin

(2014). Home-Delivered Meals and Nutrition Status Among Older Adults. Nutrition in Clinical Practice, 29(4), 459–465. https://doi.org/10.1177/0884533614536446.

23.

Thomas

K. S.

Mor

(2013). Providing more home-delivered meals is one way to keep older adults with low care needs out of nursing homes. Health Affairs, 32(10), 1796–1802. https://doi.org/10.1377/hlthaff.2013.0390.

24.

Townsend

Piper

Frank

A. O.

Dyer

North

W. R.

Meade

T. W.

(1988 Aug 20-27). Reduction in hospital readmission stay of elderly patients by a community based hospital discharge scheme: a randomised controlled trial. Bmj: British Medical Journal, 297(6647), 544–547. https://doi.org/10.1136/bmj.297.6647.544.

25.

Verbrugge

L. M.

Jette

A. M.

(1994). The disablement process. Social Science & Medicine, 38(1), 1–14. https://doi.org/10.1016/0277-9536(94)90294-1.

26.

Vertrees

J. C.

Manton

K. G.

Adler

G. S.

(1989). Cost effectiveness of home and community-based care. Health Care Financing Review, 10(4), 65.