Key PointsQuestion
How does family caregivers’ health change when the person they care for begins to receive Medicaid home care services?
Findings
In this longitudinal cohort study, family caregivers’ self-rated mental health improved significantly after their family member began receiving Medicaid home care services. Their self-rated physical health did not change.
Meaning
The findings of this longitudinal cohort study suggest that Medicaid home care programs may have spillover benefits, affecting not only their direct recipients but also those recipients’ family caregivers.
Importance
More than 17 million people in the US provide uncompensated care for adults with physical or cognitive limitations. Such caregiving is associated with worse mental and physical health, yet little research has investigated how publicly funded home care might ameliorate these harms.
Objective
To investigate the association between Medicaid home care services and family caregivers’ health.
Design, Setting, and Participants
This longitudinal cohort study used data from the 1996 to 2017 Medical Expenditures Panel Survey. Data on all household members were collected in 5 interviews over 2 years. Person-level difference-in-difference models were used to isolate within-person changes associated with new onset of Medicaid home care. The Medical Expenditures Panel Survey longitudinal data sets included 331 202 individuals (approximately 10% excluded owing to loss to follow-up). Adult (age ≥21 years) members of households that contained at least 1 person with limited activities of daily living were included in our study. The analysis itself was performed from March to August of 2020.
Exposures
New onset of regular (≥1 time per month) Medicaid home care in the household.
Main Outcomes and Measures
Self-rated mental and physical health (planned prior to beginning the study).
Results
The study population was 14 013 adults; 7232 were “likely caregivers,” or nondisabled adult coresidents of someone with activities of daily living limitations. Overall, 962 likely caregivers were ever exposed to Medicaid home care in the household; for 563, we observed the onset. Of likely caregivers exposed to Medicaid home care, 479 (50%) were women; 296 (31%) were White non-Hispanic, 309 (31%) were Hispanic or Latinx, and 279 (29%) were Black non-Hispanic individuals, respectively; 326 (34%) had less than a high school education; and 300 (31%) were in or near poverty. Median age of participants was 51 (interquartile range, 39-62) years. New-onset Medicaid home care was associated with a 0.08 standard deviation improvement in likely caregivers’ self-rated mental health (95% CI, 0.01-0.14; P = .02) measured 1 to 6 months after onset, equivalent to a 3.39% improvement (95% CI, 0.05%-6.33%) over their average preonset mental health. No association with self-rated physical health was found (<0.001 standard deviations; 95% CI, −0.06 to 0.06; P = .99).
Conclusions and Relevance
In this cohort study, Medicaid home care was associated with improvement in caregiver self-rated mental health, but not with any short-term change in self-rated physical health. When evaluating the social value of home care programs, policy makers should consider spillover benefits to caregivers.
In the US, more than 17 million people care for family or friends who need assistance with activities of daily living.1 Like most domestic and care labor, this labor is often unpaid, invisible, and performed by women.2-4 When paid, caregiving labor is largely performed by immigrant women and women of color, paid below a living wage, and excluded from labor protections.5-7 Epidemiology, economics, and psychology research shows that uncompensated caregiving is associated with worse physical and mental health, including depression and anxiety2,8-17; elevated cardiometabolic risk factors11,18-21; and higher rates of future cardiovascular disease,12,18,22,23 frailty,24 and, according to some studies, death.25-32 These harms are concentrated among women, people with low incomes, people providing more intensive care, and people without choice in becoming caregivers.8,18,19,33 Although caregiving can provide a sense of purpose and connection, in the US, these benefits are typically offset by physical, financial, and emotional strain.1,11,28,34
Public policy plays a role in structuring caregiving and can mitigate or exacerbate its associated health harms. Research has shown that policies supporting caregivers of children, like paid parental leave and publicly funded childcare, benefit these caregivers’ health.35-38 Similarly, respite care, payments to family caregivers,39 and case management have been shown to improve the well-being of caregivers for adults.40-49 Yet although caregiving-associated health harms are structural—shaped by policies, social norms, and economic pressures grounded in sexism, ableism, racism, and classism50-54—many efforts to ameliorate them focus on individual-level interventions like support groups and education.1,3,55-58 Though these interventions may help caregivers cope, they do not change the demands of caregiving or address the underlying social and economic pressures.
Over recent decades, US long-term care policy has undergone a major shift toward home- and community-based services.59,60 Still, access to these services remains limited and varies widely from state to state.61-63 To investigate how these structural factors shape caregivers’ health, we used an ecosocial theoretical framework,64 longitudinal data from the Medical Expenditures Panel Survey (MEPS), and individual-level difference-in-difference models to assess the relationship between onset of Medicaid home care and family caregivers’ health. Only Medicaid—the means-tested public health insurance program—provides long-term home care services in the US; Medicare, the universal health insurance program for people aged 65 years or older or with disabilities, provides just limited, short-term services.59,62,65-67 We focused on publicly funded long-term care, rather than private-paid care, because such programs can reduce caregivers’ physical and emotional burden without increasing their financial burden.65,68-71 Little research has investigated how publicly funded home care affects caregivers, and existing studies do not assess caregiver health outcomes, are from outside the US, or compare home care to institutionalization rather than to continued stand-alone family caregiving.72-77
We hypothesized that family caregivers’ health would improve when their care recipient begins receiving Medicaid home care, and that these benefits would vary by gender, race or ethnicity, socioeconomic status, and age of the caregiver, and by care recipient cognitive impairment.8,19,71,78-81 If such spillover exists, this would have important policy implications: investment in Medicaid home care might produce greater population health gains than previously described, providing benefits not traditionally counted in cost-benefit analyses.
Setting, Data Sources, Participants, and Study Size
The MEPS is a nationally representative, longitudinal survey with data on households’ health care utilization and spending. Recruitment and data collection methods are described elsewhere.82 All data were deidentified and publicly available, and our study was deemed not human participants research by the Harvard University institutional review board. We followed the Strengthening the Reporting of Observational Studies in Epidemiology (STROBE) reporting guidelines.83
We used data from all 21 complete MEPS panels available at the time of analysis (March-August 2020), which span the years 1996 to 2017. Each household was interviewed 5 times: once every 6 months for 2 years. We used MEPS longitudinal data files, which exclude individuals lost to follow-up (approximately 10%; exact share varies by panel). We limited our sample to households that included an adult with activities of daily living (ADL) limitations, assessed by a single question about “help or supervision with personal care such as bathing, dressing, or getting around the house because of an impairment or a physical or mental health problem.” We excluded children younger than 21 years and individuals who lived in 2 or more households during the survey, lived alone, or were missing covariates or outcomes. Further details are available in eMethods: Population in the Supplement.
Study Design and Approach to Bias
Only people with low income and assets and medical need are eligible for Medicaid home care; these factors are associated with worse health among other household members. Thus, rather than comparing households that did and did not receive Medicaid home care—which would be biased by unobserved confounding—we used person-level difference-in-difference models to estimate within-person change in outcomes associated with within-person onset of the exposure.84,85 Models were conditional on person and round, controlling for all person-level, time-invariant confounding and overall time trends in the outcomes. To control time-varying confounding, we included person- and household-level covariates.
Variables and Measurement
The primary exposure was new onset of regular Medicaid home care. At each interview, a household respondent reported all health services received by household members since the last interview; the payer was documented in this interview and by contacting the relevant physicians, hospitals, home health agencies, and/or pharmacies. We defined regular Medicaid home care as any home health services paid by Medicaid and received more than once per month for at least 1 month during the reference period. If anyone received Medicaid home care in a given round, everyone in the household was considered exposed (eMethods: Variables in the Supplement).
To distinguish direct effects of Medicaid home care on recipients from potential spillover effects on caregivers, we categorized household members as either Medicaid home care recipients, disabled nonrecipients (adults with ADL limitations who never received Medicaid home care), or likely caregivers (adult nonrecipients without ADL limitations). People changed exposure status frequently (eFigures 1 and 2 in the Supplement), so our primary exposure was initial onset of Medicaid home care because this was straightforward to operationalize and substantively important. We defined an exposure variable that was 0 in rounds with no Medicaid home care and 1 in the first round with Medicaid home care in the household; after that round, individuals dropped out of the model. Thus, our estimates reflect within-person changes in the outcome associated with the initial onset of Medicaid home care, above and beyond any changes in health experienced by unexposed individuals over time.
The primary outcomes were self-rated physical and mental health. Household respondents reported both measures on a scale from 1 (excellent) to 5 (poor). We standardized these scales and reversed the coding so that positive associations indicate improvement in self-rated mental or physical health. Outcomes were measured at each interview, ensuring that exposures temporally preceded outcomes.
Person-level covariates included hospitalizations and emergency department visits between the previous and current interviews (1, 2, 3, 4, or ≥5), number of nights spent in hospital between the previous and current interviews, and employment status at the interview date (employed vs unemployed). Household-level covariates included the number of other household members with fair/poor self-rated physical and mental health (1, 2, 3, 4, or ≥5) at the interview date, whether any other household members were hospitalized or had emergency department visits since the last interview, and whether anyone in the household was employed at the interview date. We selected these covariates because they were available in every round and because socioeconomic status, prior health status, and family members’ health status are known determinants of both Medicaid eligibility and current health status (eMethods: Variables in the Supplement).
We assessed 6 potential association measure modifiers: gender (women vs men), baseline poverty status (in or near poverty vs low/middle/high income), whether someone in the household had a cognitive limitation/dementia (yes vs no), employment status (employed vs unemployed), age (younger than 65 years vs older than 65 years), and race and ethnicity (individuals identifying as non-Hispanic White, non-Hispanic Black, Hispanic/Latinx, non-Hispanic Asian, or multiple race/other). These social categories have all been found, in prior research, to be associated with caregiving, degree of caregiver burden, or severity of caregiving-associated health harms.8,19,71,78,79,86 All variables were reported by the household respondent using the categories available on the survey instrument.
For descriptive analyses, we used χ2 tests. For our primary difference-in-difference analysis, we ran unadjusted models (conditional on person and round), models adjusted for individual-level covariates, and models adjusted for individual- and household-level covariates (fully adjusted). In all models, we included an interaction term between the difference-in-difference variable and household role (likely caregiver, disabled nonrecipient, and Medicaid home care recipient); our primary results are those for the likely caregiver group.
As exploratory secondary analyses, we ran fully adjusted event study models to assess changes in self-rated health in each round before and after the onset of Medicaid home care.87 We applied Wald tests to the parameters from these models to test for parallel pretrends, checking for any trend in self-rated health prior to Medicaid home care onset that might render our primary results invalid. In addition, to explore the possibility of association measure modification, we ran separate difference-in-difference-in-difference models for the 6 potential modifiers. In each model, we included a 3-way interaction between the difference-in-difference variable, household role, and demographic variable of interest. Although these models were not fully stratified, we refer to them as “stratified analyses” in the Results section for simplicity. More details, including estimating equations, are in eMethods: Model in the Supplement.
All models used conditional likelihood multivariable linear regression with robust standard errors clustered at the individual level. We did not incorporate the complex survey design because we found that standard errors clustered at the individual level were more conservative (eMethods: Model in the Supplement). We did not apply survey weights because we sought to assess the association between Medicaid home care and self-rated health in households that actually received Medicaid home care in our sample; we have no basis for claiming these results are nationally generalizable. We did perform sensitivity analyses using the survey weights (eMethods: Sensitivity Analyses in the Supplement). We also performed sensitivity analyses comparing self-rated mental health to more clinically interpretable measures of mental health only available in certain rounds (eMethods: Sensitivity Analyses in the Supplement).
All hypothesis tests were 2-sided. For exploratory analyses, we used an a priori statistical significance cut-off of P = .05. For our primary analyses, we used a Bonferroni-adjusted cutoff of P = .025 to account for testing two primary outcomes (self-rated physical and mental health); because these outcomes are correlated, this approach is conservative. All statistical analysis was conducted using Stata statistical software (version 16, Stata Corp), using the xt series of commands for conditional likelihood models. The analysis itself was performed from March to August of 2020.
The MEPS longitudinal data sets included 331 202 individuals. After excluding people who lived in 2 or more households (4.2% of the population) and restricting to households where at least 1 adult had ADL limitations, our sample contained 21 184 individuals. We excluded 3713 people who lived alone, 3408 children, and 50 adults who were missing data (demographics available in eTable 1 in the Supplement), for a final sample of 14 013. All these adults contributed to covariate parameter estimates, but because we used conditional likelihood models, only the 1201 living in a household first exposed to Medicaid home care after round 1 contributed to main association estimates. Of these, 560 were Medicaid home care recipients, 563 were likely caregivers, and 78 were disabled nonrecipients. Overall, 30 individuals, mostly Medicaid home care recipients, did not contribute to association estimates because they died or stopped responding after the exposure was measured but before outcomes were measured.
Table 1 and eTables 2 and 3 in the Supplement show descriptive statistics. People living in households where someone received Medicaid home care were more likely than others in the population to be women, unemployed, and Black non-Hispanic or Hispanic or Latinx; they had lower education levels, lower income, and worse self-rated physical and mental health, though similar rates of urgent health care utilization and a similar age distribution were reported. In Medicaid home care-receiving households, the Medicaid home care recipients themselves were more likely to be older and female and have lower education levels and household incomes. They were less likely to be employed and had worse self-rated health and greater urgent health care utilization.
Results of the primary difference-in-difference analyses are shown in Table 2 and Table 3. Among likely caregivers, onset of Medicaid home care in the household was associated with an improvement in self-rated mental health of 0.08 standard deviations (95% CI, 0.01-0.14; P = .02) in the fully adjusted model. This was equivalent to a 3.39% improvement (95% CI, 0.05%-6.33%) over their average preonset self-rated mental health. Equivalent results on the original (unstandardized) scale are in the eResults and eTable 4 in the Supplement. There was no association between onset of Medicaid home care and self-rated physical health (<0.001 standard deviations; 95% CI, −0.06 to 0.06; P = .99).
Results of the event study model are in Figure 1; eTables 5 and 6, and eFigure 3 in the Supplement. This model showed no trend in likely caregivers’ self-rated mental health (F statistic = 1.10; P = .35) or self-rated physical health (F statistic = 0.65; P = .58) prior to the onset of Medicaid home care, indicating the parallel pretrends assumption was upheld. Qualitatively, there was sustained self-rated mental health improvement among likely caregivers after the onset of Medicaid home care, with no apparent change in self-rated physical health.
Results of the exploratory stratified difference-in-difference analyses are in Figure 2 and eTable 7 in the Supplement. The analysis was underpowered to assess whether there were statistically significant differences in associations by demographic category, but we observed notable trends suggesting association measure modification. Among likely caregivers who were women, Medicaid home care was associated with statistically significant self-rated mental health improvement (0.09 standard deviations; 95% CI, 0.001-0.18; P = .05), an average improvement over baseline of 4.2% (95% CI, 0.04%-8.24%); among men, the association was smaller and not statistically significant (0.06 standard deviations; 95% CI, −0.03 to 0.15; P = .23). Similarly, the association was larger and statistically significant among caregivers for people with cognitive impairments (0.08 standard deviations; 95% CI, 0.01-0.15; P = .03), caregivers living in or near poverty (0.15 standard deviations; 95% CI, 0.02-0.27; P = .02), unemployed caregivers (0.10 standard deviations; 95% CI, 0.005-0.19; P = .04), and caregivers younger than 65 years (0.09 standard deviations; 95% CI, 0.02-0.17; P = .01). The association was smaller and not statistically significant among caregivers for people without cognitive impairments, caregivers living above poverty, employed caregivers, and caregivers older than 65 years. There was statistically significant association measure modification by race or ethnicity. Among Hispanic/Latinx caregivers, onset of Medicaid home care was associated with a 0.19 standard deviation improvement in self-rated mental health (95% CI, 0.07-0.31; P = .002). Among Black non-Hispanic caregivers, this association was only 0.09 standard deviations and not significant (95% CI, −0.03 to 0.204; P = .14); among White non-Hispanic caregivers, the association was even smaller (0.05 standard deviations; 95% CI, −0.07 to 0.16; P = .43).
Sensitivity analysis results are in eResults, eTables 8-11, and eFigure 4 in the Supplement.
In this cohort study, we found that Medicaid home care not only served its direct recipients, but also indirectly benefited their caregivers. Caregivers’ self-rated mental health improved after the onset of home care, gains that were observed 1 to 6 months after onset and appeared to be sustained over time. The initial improvement was approximately equivalent to an 8% decrease in the odds of screening positive for depression on the PHQ-2 or a 9% decrease in the odds of screening positive for severe psychological distress on the Kessler-6 scale (eResults in the Supplement). It was similar in magnitude to the decrease in mental health associated with unemployment or recent hospitalization (though our models were not designed to estimate causal effects for these covariates). Moreover, the association with improved mental health was nearly twice as large among certain socially disadvantaged populations. These results were consistent with prior research showing that women provide more and higher-intensity care, that Black and Latinx people are more likely to bear multiple caregiving responsibilities in paid and unpaid roles, that caregiver burden is greater for socioeconomically disadvantaged people and caregivers for people with cognitive impairments, and that younger caregivers are often members of the “sandwich generation,” with simultaneous obligations to provide childcare and eldercare.8,19,71,78,79
We found no spillover association for self-rated physical health. Prior research shows that while mental health88-90 and some biomarkers21 respond quickly to changes in caregiver burden or circumstances, many physical health changes occur over years, not months.22-24 Since our study only assessed short-term changes, subsequent research should examine whether Medicaid home care affects longer-term physical health.
This study had several strengths, particularly the use of longitudinal data and person-level difference-in-difference models. Although no observational study can provide definitive causal evidence, person-level difference-in-difference models are a rigorous method of reducing confounding. This method also eliminates selection bias due to missing data or loss to follow-up; although such issues might reduce estimates’ generalizability, they will not bias within-person estimates. Because outcomes were always measured after exposures, reverse causation was unlikely. Thus, the primary threat to internal validity is time-varying confounding, which we addressed by applying detailed health care utilization and socioeconomic information to control for likely confounders. Because confounding by socioeconomic status and prior health would create a negative association between Medicaid home care and mental health, it is exceedingly unlikely that the positive association we observe is due to confounding.
Although our study had strong internal validity, it was not generalizable to all caregivers. People who receive Medicaid home care have lower socioeconomic status and worse health than the general population; MEPS also oversamples lower-income populations and people of color.82 Our secondary analyses suggested that these populations benefited more from Medicaid home care than others, and our sensitivity analyses showed that when we weighted our study population to reflect national demographics, our primary association attenuated. Our results also may not generalize to caregivers who live in other households. Thus, our results should not be interpreted as the national population-average effect of Medicaid home care, but as evidence that Medicaid home care spills over to at least some caregivers, particularly those who bear the highest burden of care.
Because self-identified caregiver status was unavailable, this study examined “likely caregivers,” diluting the effect of Medicaid home care on true caregivers and excluding caregivers who themselves have disabilities. Data on all household members were reported by a single household respondent, not by each individual, potentially inducing further measurement error and bias toward the null. We defined Medicaid home care as a single, binary exposure; we did not investigate the role of heterogeneity in the volume, type, and duration of services, nor did we investigate long-term associations. Future longitudinal studies should identify caregivers; interview caregiver-care recipient dyads, including non-coresident dyads; explore different types and amounts of Medicaid home care; and use intersectional approaches in larger sample sizes to assess whether particular overlapping racial or ethnic, gender, and socioeconomic groups may benefit most from publicly funded home care.
In this cohort study, family caregivers’ self-reported mental health improved after the onset of Medicaid home care. As the US population ages and the number of available family caregivers shrinks, policy makers must address a growing caregiving crisis.1 Our results suggest publicly funded home care is part of the solution. Although past research has explored Medicaid home care’s benefits for its direct recipients, to our knowledge, ours is the first to show that these programs are associated with better mental health for family caregivers—particularly those who are most disadvantaged. When assessing policies like President Biden’s proposed expansion of Medicaid home- and community-based services,91 policy makers should consider not only health benefits for direct recipients, but also spillover health benefits for caregivers.
Accepted for Publication: July 23, 2021.
Published: September 17, 2021. doi:10.1001/jamahealthforum.2021.2671
Open Access: This is an open access article distributed under the terms of the CC-BY License. © 2021 Unger ES et al. JAMA Health Forum.
Corresponding Author: Emily S. Unger, MD/PhD candidate, 12 Agassiz Park #2, Jamaica Plain, MA 02130 (emily_unger2@hms.harvard.edu).
Author Contributions: Ms Unger had full access to all of the data in the study and takes responsibility for the integrity of the data and the accuracy of the data analysis.
Concept and design: Unger, Grabowski, Chen, Berkman.
Acquisition, analysis, or interpretation of data: Unger.
Drafting of the manuscript: Unger.
Critical revision of the manuscript for important intellectual content: Grabowski, Chen, Berkman.
Statistical analysis: Unger.
Supervision: Grabowski, Chen, Berkman.
Conflict of Interest Disclosures: None reported.
Funding/Support: The project was supported by award Number T32AG51108 from the National Institute of Aging.
Role of the Funder/Sponsor: The National Institute of Aging had no role in the design and conduct of the study; collection, management, analysis, and interpretation of the data; preparation, review, or approval of the manuscript; and decision to submit the manuscript for publication.
Disclaimer: The content is solely the responsibility of the authors and does not necessarily represent the official views of the National Institute of General Medical Sciences or the National Institutes of Health.
Data Sharing Statement: All data used in this analysis are publicly available at https://www.meps.ahrq.gov/. Restrictions on use of this data are described by the data owner on this website. Code used in this analysis is available from the corresponding author upon request and may be used for any non-commercial purposes.
1.National Academies of Science, Engineering, and Medicine.
Families Caring for an Aging America. The National Academies Press; Washington, DC. 2016:1-367. doi:
10.17226/23606 8.Bom
J, Bakx
P, Schut
F, van Doorslaer
E. The impact of informal caregiving for older adults on the health of various types of caregivers: a systematic review.
Gerontologist. 2019;59(5):e629-e642. doi:
10.1093/geront/gny137PubMedGoogle Scholar 13.de Zwart
PL, Bakx
P, van Doorslaer
EKA. Will you still need me, will you still feed me when I’m 64? The health impact of caregiving to one’s spouse.
Health Econ. 2017;26(S2)(suppl 2):127-138. doi:
10.1002/hec.3542
PubMedGoogle Scholar 21.Allen
AP, Curran
EA, Duggan
Á,
et al. A systematic review of the psychobiological burden of informal caregiving for patients with dementia: focus on cognitive and biological markers of chronic stress.
Neurosci Biobehav Rev. 2017;73:123-164. doi:
10.1016/j.neubiorev.2016.12.006
PubMedGoogle ScholarCrossref 26.O’Reilly
D, Rosato
M, Ferry
F, Moriarty
J, Leavy
G. Caregiving, volunteering or both? comparing effects on health and mortality using census-based records from almost 250,000 people aged 65 and over.
Age Ageing. 2017;46(5):821-826. doi:
10.1093/ageing/afx017
PubMedGoogle ScholarCrossref 27.Maguire
A, Rosato
M, O’Reilly
D. Mental health and morbidity of caregivers and co-residents of individuals with dementia: a quasi-experimental design.
Int J Geriatr Psychiatry. 2017;32(10):1104-1113. doi:
10.1002/gps.4573
PubMedGoogle ScholarCrossref 29.Roth
DL, Haley
WE, Hovater
M, Perkins
M, Wadley
VG, Judd
S. Family caregiving and all-cause mortality: findings from a population-based propensity-matched analysis.
Am J Epidemiol. 2013;178(10):1571-1578. doi:
10.1093/aje/kwt225
PubMedGoogle ScholarCrossref 30.Ramsay
S, Grundy
E, O’Reilly
D. The relationship between informal caregiving and mortality: an analysis using the ONS Longitudinal Study of England and Wales.
J Epidemiol Community Health. 2013;67(8):655-660. doi:
10.1136/jech-2012-202237
PubMedGoogle ScholarCrossref 31.Fredman
L, Cauley
JA, Hochberg
M, Ensrud
KE, Doros
G; Study of Osteoporotic Fractures. Mortality associated with caregiving, general stress, and caregiving-related stress in elderly women: results of caregiver-study of osteoporotic fractures.
J Am Geriatr Soc. 2010;58(5):937-943. doi:
10.1111/j.1532-5415.2010.02808.x
PubMedGoogle ScholarCrossref 36.Berkman
LF, Avendano
M. Labor markets, employment policies, and health. In: Berkman
LF, Kawachi
I, Glymour
M, eds. Social Epidemiology. 2nd ed. Oxford University Press; 2014.
40.Smith
VA, Lindquist
J, Miller
KEM,
et al. Comprehensive family caregiver support and caregiver well-being: preliminary evidence from a pre-post-survey study with a non-equivalent control group.
Front Public Health. 2019;7:122. doi:
10.3389/fpubh.2019.00122
PubMedGoogle ScholarCrossref 41.Corvol
A, Dreier
A, Prudhomm
J, Thyrian
JR, Hoffmann
W, Somme
D. Consequences of clinical case management for caregivers: a systematic review.
Int J Geriatr Psychiatry. 2017;32(5):473-483. doi:
10.1002/gps.4679
PubMedGoogle ScholarCrossref 42.Ellen
ME, Demaio
P, Lange
A, Wilson
MG. Adult day center programs and their associated outcomes on clients, caregivers, and the health system: a scoping review.
Gerontologist. 2017;57(6):e85-e94. doi:
10.1093/geront/gnw165
PubMedGoogle ScholarCrossref 44.Vandepitte
S, Van Den Noortgate
N, Putman
K, Verhaeghe
S, Verdonck
C, Annemans
L. Effectiveness of respite care in supporting informal caregivers of persons with dementia: a systematic review.
Int J Geriatr Psychiatry. 2016;31(12):1277-1288. doi:
10.1002/gps.4504
PubMedGoogle ScholarCrossref 45.Bass
DM, Judge
KS, Snow
AL,
et al. Caregiver outcomes of partners in dementia care: effect of a care coordination program for veterans with dementia and their family members and friends.
J Am Geriatr Soc. 2013;61(8):1377-1386. doi:
10.1111/jgs.12362
PubMedGoogle ScholarCrossref 50.Federici
S. Notes on Elder Care Work and the Limits of Marxism. In: van der Linden
M, Roth
KH, eds. Beyond Marx: Theorising the Global Labour Relations of the Twenty-First Century. Vol 56. Historical Materialism Book Series. Brill; 2013.
51.Piepzna-Samarasinha
LL. Care Work. Arsenal Pulp Press; 2018.
52.Abel
EK. Love Is Not Enough: Family Care of the Frail Elderly. American Public Health Association; 1987.
53.Tronto
JC, Fisher
B. Toward a Feminist Theory of Caring. In: Abel
EK, Nelson
MK, eds. Circles of Care: Work and Identity in Women’s Lives. SUNY series on Women and Work. State University of New York Press; 1990.
54.Hooyman
NR. Social and health disparities in aging: Gender inequities in long-term care.
Gener J Am Soc Aging. 2014;38(4):25-32.
Google Scholar 55.Murfield
J, Moyle
W, O’Donovan
A. Mindfulness- and compassion-based interventions for family carers of older adults: A scoping review.
Int J Nurs Stud. 2021;116:103495. Published online December 2, 2019. doi:
10.1016/j.ijnurstu.2019.103495PubMedGoogle Scholar 62.Willink
A, Kasper
J, Skehan
M, Wolff
J, Mulcahy
J, Davis
K. Are older Americans getting the long-term services and supports they need?
Issue Brief (The Commonwealth Fund). 2019:1-9.
PubMedGoogle Scholar 67.Willink
A, Davis
K, Mulcahy
J, Wolff
JL, Kasper
J. The financial hardship faced by older Americans needing long-term services and supports.
Issue Brief (Commonw Fund). 2019;2019:1-12.
PubMedGoogle Scholar 70.Willink
A, Davis
K, Mulcahy
J, Wolff
JL. Use of paid and unpaid personal help by Medicare beneficiaries needing long-term services and supports.
Issue Brief (Commonw Fund). 2017;2017:1-9.
PubMedGoogle Scholar 71.Riffin
C, Van Ness
PH, Wolff
JL, Fried
T. Multifactorial examination of caregiver burden in a national sample of family and unpaid caregivers.
J Am Geriatr Soc. 2019;67(2):277-283. doi:
10.1111/jgs.15664
PubMedGoogle ScholarCrossref 76.Eom
K, Penkunas
MJ, Chan
AWM. Effect of caregiving relationship and formal long-term care service use on caregiver well-being.
Geriatr Gerontol Int. 2017;17(10):1714-1721. doi:
10.1111/ggi.12958PubMedGoogle Scholar 82.Chowdhury
SR, Machlin
SR, Gwet
KL. Sample Designs of the Medical Expenditure Panel Survey Household Component, 1996–2006 and 2007–2016. Agency for Healthcare Research and Quality, Rockville, MD; 2019:1-44.
83.von Elm
E, Altman
DG, Egger
M, Pocock
SJ, Gøtzsche
PC, Vandenbroucke
JP; STROBE Initiative. The Strengthening the Reporting of Observational Studies in Epidemiology (STROBE) statement: guidelines for reporting observational studies.
Ann Intern Med. 2007;147(8):573-577. doi:
10.7326/0003-4819-147-8-200710160-00010
PubMedGoogle ScholarCrossref 84.Angrist
JD, Pischke
J-S. Ch. 5: Parallel Worlds: Fixed Effects, Differences-in-Differences, and Panel Data. In:
Mostly Harmless Econometrics. Princeton University Press; 2008:221-247. doi:
10.2307/j.ctvcm4j72.12
85.Wooldridge
JM. Ch. 10: Basic Linear Unobserved Effects Panel Data Models. In: Econometric Analysis of Cross Section and Panel Data. MIT Press; 2002:752.
86.van den Kieboom
R, Snaphaan
L, Mark
R, Bongers
I. The trajectory of caregiver burden and risk factors in dementia progression: a systematic review.
J Alzheimers Dis. 2020;77(3):1107-1115. doi:
10.3233/JAD-200647
PubMedGoogle ScholarCrossref 87.Venkataramani
AS, Cook
E, O’Brien
RL, Kawachi
I, Jena
AB, Tsai
AC. College affirmative action bans and smoking and alcohol use among underrepresented minority adolescents in the United States: a difference-in-differences study.
PLoS Med. 2019;16(6):e1002821. doi:
10.1371/journal.pmed.1002821
PubMedGoogle Scholar