Paying family caregivers: self-direction in medicaid personal care

Abstract

Introduction

As people age, many require help with personal care—often provided by family members or friends or direct care workers. Over the last three decades, driven by the disability rights movement, direct care worker shortages, and caregiver burden, self-direction has emerged as a Medicaid option that allows individuals to hire and pay their own caregivers, including family members.

Methods

We use 2021 TMSIS Medicaid claims data linked to Medicare Master Beneficiary Summary File to identify dually eligible beneficiaries 65+ receiving personal care. We describe the demographic characteristics of self-direction users compared to non-self-direction users and present the percentage of self-direction users across states.

Results

We find that over half of dually eligible beneficiaries 65+ receiving personal care use self-direction. Compared to individuals who use agency-based personal care, self-direction users have higher prevalence of chronic disease, higher home health use, and higher Medicare costs.

Conclusion

Self-direction has become a common model of personal care among older adults enrolled in Medicaid. Examining how funds allocated for self-direction are spent; the effects of self-direction on consumers and their caregivers; and how self-direction may impact Medicare and Medicaid costs is critical to inform the expansion and funding of Medicaid self-direction programs.

Introduction

As one ages, individuals may incur functional or cognitive limitations necessitating the need for help with daily living and personal care. In 2016, 42% of older adults have functional or cognitive disabilities.¹ As the population ages, the percentage of older individuals with disabilities increased to 44% in 2022.^1,2 To meet care needs, individuals often rely on direct care workers, such as personal care aides, who are paid and trained to provide care for their occupation. Medicaid is the primary payer for help with daily living activities for older adults and, as such, plays a critical role in the expansion of home- and community-based services (HCBS) to help low-income elderly individuals receive assistance with daily living activities at home to age-in-place. Direct care workers are often assigned to individuals through Medicaid-certified home care agencies which may result in misalignment with the specific needs and preferences of beneficiaries regarding the type, amount, timing, and delivery methods of care.³ Consequently, self-direction (ie, consumer direction) has emerged as an alternative model that allows care recipients to hire and pay their own caregivers, including family members, instead of relying on agency-assigned direct care workers. However, many states impose restrictions on which family members can be paid to provide care, often excluding spouses or legally responsible relatives from eligibility.

Self-direction grants care-recipients greater control over their care which is associated with improved outcomes for individuals.^3-5 Previous evidence of the effects of self-direction in non-veteran settings in the United States stem from a randomized controlled trial called Cash and Counseling which was implemented by the Centers for Medicare & Medicaid in 2002 and examined Medicaid beneficiaries receiving self-directed personal care in Arkansas, Florida, and New Jersey compared to beneficiaries receiving agency-based personal care in the same three states.^6,7 Evidence from the Cash and Counseling Demonstration suggests that participants using self-direction were more likely to receive paid care, reported higher satisfaction with their care, experienced comparable or superior health outcomes, and had a lower likelihood of nursing home placement compared to those receiving traditional agency-based services.^3-5 Self-directed participants in Cash and Counseling had higher Medicaid personal care costs due to increased use of authorized hours, while these higher costs were largely offset by reductions in nursing home costs and other Medicaid costs. Notably, self-direction reduced nursing home admissions by 40%. The difference in total Medicaid costs between the self-direction and agency-model was not statistically significant by the second year of the demonstration.^3,8,9 More recent analysis of Cash and Counseling further supports these findings, showing that allowing family members to be paid caregivers under self-direction contributed to reductions in emergency room use, Medicaid-financed inpatient days, overall hospital expenditures, and improved health outcomes.¹⁰ Moreover, analysis of the Cash and Counseling found that the availability of self-direction increased uptake of Medicaid personal care among persons with less severe impairments.¹¹ While the use of self-direction to meet care needs has grown to be offered in all 50 states and Washington, D.C. (only 47 states and Washington, D.C. serve older adults in non-veteran settings) since the initial Cash and Counseling Demonstration in 2002,¹² the prevalence of self-direction, who opts to use self-direction vs agency-based care, and the effects of self-direction on consumer outcomes at the national level in the years following Cash and Counseling remain largely unknown.

With the rise of self-direction in the last decade, 47 states and Washington, D.C. have used multiple mechanisms to provide self-direction for older adults in non-veteran settings,¹² resulting in substantial variation in self-direction programs. Most states adopt one or more of the following approaches: (1) Medicaid State Plan, including 1915(j) and 1915(k) State Plan Option,^13-15 (2) Medicaid 1915(c) waivers, and (3) Medicaid 1115 demonstrations.¹² The choice of mechanism, combined with state government flexibility, influences how states establish enrollment caps, eligibility criteria, and administrative structures, all of which contribute to self-direction prevalence.¹⁶ Despite the expansion of Medicaid self-direction,^13-15,17 little evidence exists about the current prevalence of using self-direction vs agency-based care among older adults. Understanding the differences in self-direction penetration can help reveal the state-level variations in barriers to access, such as lack of program awareness and limited supply of available workers. Thus, our objective is to describe users of Medicaid self-direction in the United States relative to Medicaid beneficiaries who receive personal care services, but do not receive self-directed services. These findings will inform policymakers about the current landscape of self-direction use and help guide future decisions on the expansion and funding of Medicaid self-direction programs.

Methods

Data and sample

We use 2021 TMSIS Medicaid claims data linked to Medicare Master Beneficiary Summary File to identify Medicaid beneficiaries aged 65 and above receiving any personal care who are dually eligible for Medicare. We identify the personal care users through procedure codes.¹⁸ We include procedure codes with a description of assistance with activities of daily living, personal care, attendant care, chore services, homemaker service, companion care, laundry service, and other synonymous descriptions. While self-direction can be used for a broader range of services not fully captured by these codes, our focus is on individuals requiring help with personal care.¹⁹ Service providers of personal care are not identified, though the TMSIS Medicaid claims data include an annual provider file with information about Medicaid service providers.

We include the following six states in our primary sample due to the reliability of their self-direction enrollment data, as detailed in Table SA1: California, Connecticut, Massachusetts, New Jersey, New York, and Wisconsin. These states have 2021 TMSIS Medicaid self-direction numbers that either (1) fall between 2019 and 2023 Applied Self-Direction numbers¹² or (2) differ from both 2019 and 2023 Applied Self-Direction numbers by no more than 50%. We calculate the percentage difference using the formula: (TMSIS 2021—Applied Self-Direction)/Applied Self-Direction. States are included if the difference is ≤ 50% for both 2019 and 2023.

Our larger sample for context (41 states and Washington, D.C.) includes the primary states plus all additional states where 2021 TMSIS self-direction number is lower than 2023 Applied Self-Direction number, regardless of the size of the difference. These states are more likely to undercount self-direction users. States in the primary sample on average have higher percentage of older adults using personal care compared with the states in the larger sample for context, as shown in Table 1 and Table SA2.

Table 1.

Percentage of personal care users among Medicaid beneficiaries aged 65 and above in 2021, primary sample.

	Personal care users (n = 561 588)
Total	19.7%
State (%)
CA	28.2%
CT	7.5%
MA	22.1%
NJ	21.0%
NY	8.1%
WI	13.1%

Authors’ analysis of data from the TMSIS Medicaid claims, 2021.

The total number of Medicaid beneficiaries aged 65 and above in 2021 in the 6 states above is N = 2 855 693.

We exclude Florida, Indiana, Nevada, North Carolina, Ohio, Oklahoma, Pennsylvania, Washington, and West Virginia, because their 2021 TMSIS self-direction numbers exceed the 2023 Applied Self-Direction numbers. This suggests a potential over-identification of self-direction users, which could bias the results by misclassifying non-users as users.

Key outcomes

Among those receiving personal care, we identify users of self-direction as those with: (1) Medicaid procedure codes with a description of self-direction or consumer-direction, (2) HCBS service codes indicating the HCBS service was provided under 1915(j) or 1915(k), (3) Medicaid program type code 16 indicating the personal care was self-directed under 1915(j) state plan option or 1915(c) waivers, or (4) Medicaid program type code 11 indicating the service is provided under Community First Choice 1915(k). The specific variables and codes to identify self-direction are summarized in Table SA3. Since program type code 11 (Community First Choice 1915(k)) may be used for services beyond personal care, we require both the procedure code/HCBS service code and the 1915(k)-program type code to be present (instead of using either one).

Explanatory Variables include age, sex, race and ethnicity, rurality, number of days enrolled in Medicaid, chronic conditions, Medicare healthcare use and costs, and submitting state of individuals. The variables are drawn from the Medicare Master Beneficiary Summary File's base segment, the 30 CCW (Chronic Conditions Data Warehouse) Chronic Conditions segment, and the Cost and Use file segment. Medicare use and costs include services in hospital outpatient, acute inpatient, skilled nursing facility, home health, outpatient emergency room, and inpatient emergency room. Metropolitan, micropolitan, and rural areas (including small towns) are identified by linking beneficiary ZIP codes to the primary rural-urban commuting area codes, which delineate sub-county components of rural and urban areas.²⁰

Approach

Among all dually eligible beneficiaries aged 65 and above, we first report the percentage of personal care users across states. Second, among those receiving personal care, we describe the demographic characteristics of self-directed users compared to those without self-direction by sociodemographic characteristics, health status, and health care costs. We report standardized differences between self-direction users and non-self-direction users where the absolute standardized difference values >0.1 indicate meaningful group differences.²¹ To validate our identification of self-directed users, we then compare our state-estimates of number of self-directed users with the 2019 and 2023 Self-Direction National Inventory conducted by Applied Self-Direction.¹² Finally, we present the percentage of self-direction users among all dually eligible beneficiaries aged 65 and above receiving personal care in each state.

Results

Primary sample

Prevalence

Among 2 855 693 dually eligible beneficiaries aged 65 and older in the primary sample, approximately 19.7% received personal care services in 2021. Table 1 shows the percentage of personal care users aged 65 + in each state. In three states-California, Massachusetts, New Jersey-more than 20% of older adults use Medicaid-funded personal care.

Characteristics of self-direction users

Table 2 summarizes the characteristics of dually eligible beneficiaries receiving any personal care, personal care through self-direction and personal care through agency-based care. Overall, personal care users aged 65 and older are on average 79 years old, and 68.7% are female. About 40% of personal care users are White individuals, followed by Asian and then Black individuals. Most personal care users live in metropolitan areas, and they are enrolled in Medicaid for almost the entire year. Compared to individuals receiving agency-based care, self-direction users are more likely to be Asian or Hispanic individuals and less likely to live in rural areas or small towns, with significant standardized differences (>0.1). About 27.9% of those using self-direction are aged 85 and above, with 68.6% female. White beneficiaries represent the largest proportion of self-direction users, followed by Asian and then Black beneficiaries. Only 1.4% of self-directed beneficiaries reside in small towns or rural areas.

Table 2.

Characteristics of dual-eligible beneficiaries aged 65 and above receiving personal care in 2021 by self-direction use, primary sample.

	All personal care users (n = 561 588)	Non-self-direction users (n = 63 656)	Self-direction users (n = 497 932)	Standardized difference
Demographics characteristics
Age in years, mean (SD)	78.8 (8.7)	78.7 (8.7)	78.8 (8.7)	0.01
Age group (%)				0.02
65-74	36.2%	36.8%	36.1%
75-84	36.0%	36.3%	35.9%
85 and above	27.8%	26.8%	27.9%
Female (%)	68.7%	69.8%	68.6%	0.03
Race and ethnicity (%)				0.44
White	41.6%	56.5%	39.7%
Black	12.8%	15.1%	12.5%
Asian	14.2%	11.2%	14.6%
Hispanic	10.3%	7.7%	10.7%
Other	21.1%	9.5%	22.5%
Rurality (%)				0.15
Metropolitan	95.7%	93.5%	96.0%
Micropolitan	2.6%	2.7%	2.6%
Small town and rural areas	1.7%	3.7%	1.4%
Number of days of Medicaid enrollment, mean (SD)	349.0 (59.5)	343.9 (67.2)	349.6 (58.5)	0.09
Chronic conditions (%)
Alzheimer's disease	5.0%	5.1%	4.9%	0.01
Anemia	26.4%	24.9%	26.6%	0.04
Asthma	8.5%	8.4%	8.5%	0.00
Breast cancer	2.5%	2.6%	2.4%	0.01
Chronic kidney disease	20.8%	18.2%	21.1%	0.07
Chronic obstructive pulmonary disease	14.8%	14.3%	14.8%	0.02
Depression, bipolar, or other depressive mood disorders	19.4%	20.3%	19.3%	0.02
Diabetes	34.1%	29.5%	34.6%	0.11
Heart failure and non-ischemic heart disease	17.3%	16.1%	17.4%	0.03
Hyperlipidemia	47.0%	41.1%	47.8%	0.14
Hypertension	55.2%	48.4%	56.0%	0.15
Rheumatoid arthritis/osteoarthritis	32.6%	27.0%	33.3%	0.14
Parkinson's disease and secondary Parkinsonism	2.4%	2.6%	2.4%	0.01
Stroke/transient ischemic attack	7.4%	7.8%	7.4%	0.02
Healthcare use and costs
Hospital outpatient visits, mean (SD)	14.1 (31.1)	13.1 (26.9)	14.2 (31.6)	0.04
Hospital outpatient Medicare payments, mean (SD)	4476.4 (11 880.8)	3806.2 (9880.7)	4564.45 (12 116.4)	0.07
Acute inpatient stays, mean (SD)	1.7 (1.3)	1.8 (1.4)	1.7 (1.3)	0.08
Acute inpatient covered days, mean (SD)	10.0 (12.1)	11.1 (13.2)	9.8 (11.9)	0.11
Acute inpatient Medicare payments, mean (SD)	31 495.0 (41 230.0)	27 337.3 (29 469.4)	32 102.5 (42 644.3)	0.13
Skilled nursing facility stays, mean (SD)	1.4 (0.8)	1.5 (0.9)	1.4 (0.8)	0.08
Skilled nursing facility covered days, mean (SD)	41.0 (33.1)	38.5 (30.3)	41.6 (33.8)	0.10
Skilled nursing facility Medicare payments, mean (SD)	27 583.2 (22 355.9)	23 208.6 (18 598.7)	28 668.8 (23 066.5)	0.26
Home health visits, mean (SD)	35.4 (47.5)	26.1 (36.0)	36.3 (48.4)	0.24
Home health Medicare payments, mean (SD)	9503.4 (8050.6)	6117.8 (5713.4)	9825.1 (8165.6)	0.53
Hospital outpatient emergency room visits, mean (SD)	2.0 (2.4)	2.2 (2.3)	2.0 (2.4)	0.09
Inpatient emergency room visits, mean (SD)	1.7 (1.2)	1.8 (1.3)	1.7 (1.2)	0.09

Authors’ analysis of data from the Medicare Master Beneficiary Summary File, 2021.

Metropolitan areas have primary flows to or within urbanized areas of 50 000 or more residents. Micropolitan areas have primary flows to or within urban clusters (UCs) of 10 000 to 49 999 residents. Small towns have primary flows to or within small UCs of 2500 to 9999 residents. Rural areas have primary flow to a tract outside of any UA or UC.

The data suggest that self-direction users have more acute conditions. Compared to personal care users who access services through agencies, people who self-direct have significantly more home health visits, but fewer inpatient Medicare covered days. Self-directed beneficiaries incur significantly higher Medicare payments for acute inpatient, skilled nursing facility, and home health services—ranging from $3707.3 higher for home health to $5460.2 higher for skilled nursing facility care.

Compared to beneficiaries using agency-based care, beneficiaries who self-direct are more likely to have 17 out of 30 chronic conditions. The full list of chronic conditions is provided in Table SA4. The difference is statistically significant for 6 conditions, including benign prostatic hyperplasia, diabetes, hyperlipidemia, hypertension, osteoporosis, and rheumatoid arthritis/osteoarthritis. Figure 1 shows the share of the five most prevalent chronic conditions among personal care users, comparing self-direction users and beneficiaries receiving agency-based care. For self-direction users, the probability of each condition is higher—ranging from 1.7% for anemia to 7.6% for hypertension.

Figure 1.

Percent of older adults receiving any personal care with chronic conditions by self-direction use, 2021 (primary sample). Source: Authors’ analysis of data from the Medicare Master Beneficiary Summary File, 2021. Notes: The sample includes dually eligible beneficiaries aged 65 and above who receive personal care services in California, Connecticut, Massachusetts, New Jersey, New York, and Wisconsin. Arthritis is the abbreviation of Rheumatoid Arthritis/Osteoarthritis.

Geographic variation in use of self-direction

Among 561 588 dually eligible beneficiaries aged 65 and older receiving Medicaid-funded personal care in the six states of our primary sample, about 88.7% use self-direction. Figure 2 illustrates the percentage of self-direction users among personal care recipients aged 65 and above in each state. The share of self-direction varies widely across states, ranging from 11.8% to 99.6%. Nearly all personal care users in California and New York self-direct their care, and more than half use self-direction in Massachusetts.

Figure 2.

Percent of older adults receiving any personal care enrolled in self-direction by state, 2021 (primary sample). Source: Authors’ analysis of data from the TMSIS Medicaid claims, 2021. Notes: The sample includes dually eligible beneficiaries aged 65 and above who receive personal care services in California, Connecticut, Massachusetts, New Jersey, New York, and Wisconsin.

Analysis of larger sample for context

Prevalence

Among 6 432 637 dually eligible beneficiaries aged 65 and older in 41 states and Washington, D.C., approximately 15.3% received personal care services in 2021. Table SA2 shows the percentage of personal care users aged 65 + in each state. In addition to California, Massachusetts, and New Jersey from the primary sample, Minnesota, Missouri, and Texas also have more than 20% of older adults using personal care.

Characteristics of self-direction users

Table SA5 summarizes the characteristics of dually eligible beneficiaries receiving any personal care, personal care through self-direction, and personal care through agency-based care. Most results are consistent with the primary sample, but a few differences emerge.

Compared to individuals receiving agency-based care, self-direction users are older, more likely to have cataract and glaucoma, and have longer Medicare-covered skilled nursing facility stays, all with statistically significant differences. While the primary sample shows the same trends, these differences are not statistically significant.

In the primary sample, self-direction users have significantly fewer acute inpatient covered days and more home health visits compared to those using agency-based care. However, these differences are not statistically significant in the larger sample for context.

Geographic variation in use of self-direction

Among 982 378 dually eligible beneficiaries aged 65 and older receiving personal care, about 51.2% use self-direction. Figure SA1 illustrates the percentage of self-direction users among personal care recipients aged 65 and above in each state. The share of self-direction varies widely across states, ranging from 0.1% to 99.6%. Seven states have more than 20% of self-direction users. Twelve states have <1% self-direction users.

Discussion

The study highlights the widespread use of self-direction among older adults enrolled in Medicare and Medicaid who receive personal care, with over half of the population self-directing the care—a finding largely driven by the large self-directing populations in a few states. We find large geographic variation, with California and New York accounting for the majority of self-direction users, as aligned with findings from the AARP National Inventory of Self-Directed Long-Term Services and Supports Programs.¹² We also observe wide geographic variation in the overall use of personal care services among Medicaid beneficiaries aged 65 and older. States with higher self-direction penetration, such as California and Massachusetts, generally also have higher rates of personal care use. However, New York stands out as an exception, with a lower-than-average rate of personal care use but a relatively high level of self-direction penetration. In addition, self-directed beneficiaries have a higher prevalence of chronic conditions, use home health more frequently, and incur higher Medicare payments for acute inpatient, skilled nursing facility, and home health services. Our findings provide valuable insights for policymakers on the characteristics of self-directed users and the role of self-direction in supporting older adults with complex health needs.

Self-direction has become a common model of personal care among older adults enrolled in Medicaid. In addition to the policy efforts that expanded access to self-direction, this phenomenon demonstrates two possible mechanisms. First, the high uptake may reflect a preference for greater autonomy in care, as self-direction could better align with the needs of older adults compared to traditional agency-based services. This finding also challenges the assumption that older adults might not want to or be able to manage their own care. At the same time, it may reflect the important role that family caregivers often play as surrogate decision-makers or employers on behalf of the older adults. Second, the enrollment may indicate the limited access to direct care workers from home care agencies. Self-direction serves as a viable and adequate model to empower older adults to secure care. Examining these mechanisms would help identify the key factors driving self-direction enrollment. Given the large number of beneficiaries using self-direction, understanding how funds are allocated and spent is vital for policymakers to consider further investment in self-direction programs, such as caregiver support and training. Despite the overall high prevalence, we find that self-direction remains rare in rural areas, despite its potential to address workforce shortages in these communities by enabling care delivery through family members. The low usage of self-direction in rural areas may therefore represent a missed opportunity for improving access to care and underscores the need for targeted policy efforts to expand self-direction models in underserved regions.

Despite higher rates of chronic conditions, self-direction users experience significantly fewer Medicare-covered inpatient days. This finding aligns with prior research showing that allowing family members to be paid caregivers under self-direction reduces Medicaid-financed inpatient days.¹⁰ In addition, self-direction users have fewer hospital stays, emergency department visits, and skilled nursing facility stays than non-self-direction users, though not statistically significant. This suggests that self-direction may help beneficiaries better manage their conditions at home, potentially preventing avoidable hospitalizations and reducing overall inpatient days. However, when self-direction users are hospitalized, their care may be more intensive or involve more complex conditions, leading to higher per-stay costs. This could reflect delayed care-seeking behavior, higher acuity at admission, or the presence of multiple chronic conditions that require more resource-intensive treatment. Investigating the causal impact of self-direction on consumers’ health outcomes and Medicare and Medicaid costs is crucial, since the evidence on these aspects remains scarce since Cash and Counseling two decades ago. As federal and state governments implemented self-direction policies, such as Medicaid 1915(j) and (k), evaluating these newer programs is essential to understand how they differ from earlier models and what unique challenges or benefits the newer approaches present.

The concentration of self-direction in California and New York reflects differences in state policies, program generosity, and administrative structures. Both states have implemented Medicaid 1915(j) or (k) state plans, allowing beneficiaries to self-direct without enrollment caps. California, which has the largest self-direction enrollment, has long-established self-direction program named In-Home Supportive Services dating back to the last century. In contrast, many states with lower self-direction enrollment may rely on Medicaid waivers with enrollment caps or have restrictive eligibility criteria (eg, limitations on selecting caregivers or recipient age requirements) that deter participation. Additionally, the variation in self-direction penetration may also reflect differences in how states promote self-direction. In some states, beneficiaries may be unaware of self-direction unless they are connected to peers or advocates who know about it, suggesting that program promotion plays a key role in the enrollment. Future research should examine how state-level policy differences influence self-direction use and whether expanding self-direction programs in lower-utilization states could improve access to personal care.

This study has several limitations. First, while we describe differences in healthcare use and costs between self-direction users and non-self-direction users, we cannot establish causal relationships because more acute individuals may select into self-direction. Beneficiaries who self-direct may differ systematically from those who do not in ways we cannot fully observe. For example, it is ambiguous whether differences between self-direction users and non-users are pre-existing or shaped, in part, by participation in self-direction. Future research could apply matching or other quasi-experimental methods to better account for observed and unobserved differences and identify causal effects. Second, state-level variation in Medicaid data reporting may affect the comparability of self-direction enrollment across states, and thus the results from the primary sample may not be nationally representative. While the primary sample includes states with more reliable reporting, it also captures some of the largest self-direction programs in the country, such as California, New York, Massachusetts, and Wisconsin. It is possible that the quality of Medicaid data reporting is correlated with program characteristics. Since these states have high rates of self-direction among older adults using personal care, results from the primary sample may reflect outcomes in large, well-established programs and should be interpreted with caution when generalizing to the national context. Third, the self-direction users identified in TMSIS Medicaid claims include beneficiaries participating in the “Agency with Choice” model. This model allows individuals to retain control over the selection and dismissal of their caregivers, even though the caregivers are employed by an agency rather than directly by the individual.¹³ While these consumers may not hire relatives or friends, the “Agency with Choice” remains an important variant within the spectrum of self-direction, and we include these users in our definition of self-direction. Fourth, the racial composition of the self-direction population may be disproportionately affected by the overrepresentation of Asian beneficiaries in California and New York and therefore may not reflect the racial distribution of self-direction users in other states. Finally, our findings focus on Medicaid beneficiaries aged 65 and older, and results may not generalize to younger populations.

Our study shows that as of 2021 self-direction has become a common model of personal care among older adults enrolled in Medicaid across the 41 states and the Washington, D.C. included in our sample. However, significant heterogeneity exists across states, with overall self-direction use largely driven by a few states with large programs, most notably California and New York. Moreover, given that self-direction users have higher prevalence of chronic disease, higher home health use, and higher Medicare costs compared to non-users, future research is necessary to disentangle the mechanisms by which self-direction impacts health, health care use and costs. Specifically, future research should examine (1) how funds allocated for self-direction are spent, (2) the effects of self-direction on consumers and their caregivers, and (3) how self-direction contributes to or mitigates Medicare and Medicaid costs. Examining program expenditures is key to identify the areas most in need of funding and provide guidance for future investments. Evidence of how self-direction impacts consumer health outcomes and costs are critical to evaluate the cost-effectiveness of self-direction. In addition, research on consumers autonomy and quality of life, such as those captured by the National Core Indicators and Personal Outcome Measures surveys in the Centers for Medicare & Medicaid Services HCBS Quality Measure Set,²² can shed light on whether self-direction helps meet the care needs in a consumer-centered way. While self-direction is expected to reduce caregiver burden, limited research has explored its impact on caregivers’ economic security and well-being. Existing research shows that expanding access to formal personal care services increases labor supply among daughters of Medicaid-eligible older adults, suggesting improved economic outcomes for caregivers.²³ Given the high prevalence of self-direction among older adults using personal care, it is crucial to examine the impact of self-direction on caregivers’ economics outcomes and quality of life. Addressing these gaps will help inform state-level policies as self-direction continues to play a growing role in meeting personal care needs of an aging population.

Supplementary material

Supplementary material is available at Health Affairs Scholar online.

Funding

This manuscript is the result of funding in whole or in part by the National Institutes of Health (NIH). It is subject to the NIH Public Access Policy. Through acceptance of this federal funding, NIH has been given a right to make this manuscript publicly available in PubMed Central upon the Official Date of Publication, as defined by NIH. Dr. Miller’s time was supported, in part, by the National Institute of Aging (1R21AG089381-01A1) and an internal grant from the Johns Hopkins Bloomberg School of Public Health. The content is solely the responsibility of the authors and does not necessarily represent the official views of the National Institutes of Health or any institution with which the authors are affiliated.