National Estimates of Incremental Work Absenteeism Costs Associated with Adult Children of Parents with Alzheimer’s Disease and Related Dementias

Abstract

Objective:

More than half of primary caregivers for ADRD patients are adult children, yet there is little empirical evidence on how caring for parents with ADRD affects their employment. Using a nationally representative dataset, this study aimed to estimate incremental work absenteeism costs for adult children of parents with ADRD.

Design, setting, and participants:

The study used the data from the 2015-2021 Medical Expenditure Panel Survey (MEPS). Multivariate regressions and two-part models were employed to estimate the incremental work absenteeism costs among adult children aged 40 to 64 who had at least one parent diagnosed with ADRD, compared with those who did not have ADRD parents.

Measurements:

The incremental work absenteeism costs due to caregiving for adult children with ADRD parents was a cumulated estimation of labor productivity cost at three stages: (1) the likelihood of not working due to unemployment, (2) the likelihood of missing any workdays for caregiving, and (3) the number of workdays missed due to caregiving.

Results:

Adult children with ADRD parents were more likely to be unemployed (OR=1.80, p=0.024) and 2.95 times more likely to miss work for caregiving (p=0.002) than those with non-ADRD parents. The difference in the number of workdays missed for caregiving between children with and without ADRD parents was not significant. The incremental effects of having ADRD parents were estimated to be $4,510.29 ($1,702.09 - $6,723.69) per person per year.

Conclusions:

Having ADRD parents significantly increases the chances of unemployment and missing any workdays for caregiving, leading to higher lost labor productivity costs for adult children with ADRD parents.

INTRODUCTION

Informal caregivers serve as the backbone of long-term care and support systems for older adults in the US.^(1,2) About 83% of assistance to seniors is provided by informal caregivers, who are often unpaid family members and friends, and there are approximately 53 million of them.^(2,3) Alzheimer’s disease and related dementias (ADRD) is a progressive disease that worsens with varying levels of symptoms, including memory loss, personality change, and functional and cognitive limitations. In advanced stages of dementia, individuals can no longer communicate, recognize family and friends, swallow, or walk independently. ADRD patients are highly likely to be confined to bed and require full-time caregiving.⁽³⁾

More than half of primary caregivers for ADRD patients are adult children.^(4,5) These adult children often face a significant caregiving burden, which includes daily activities, emotional support, respite support for another primary caregiver, or financial support to their parents with ADRD.^(3–6) Previous research has assessed and compared social, health, and financial consequences of caregiving between ADRD caregivers and non-ADRD caregivers, mainly focusing on middle-aged or older child caregivers.^(1,4,6,7) Studies that estimated direct costs of healthcare utilization and unpaid caregiving of ADRD suggest that children with ADRD parents often face significant challenges, substantially higher health care costs, and worse health status.^(4,6–10) Family caregivers pay about 70% of the $400,000 lifetime care costs of an ADRD person through unpaid caregiving and out-of-pocket spending.^(3,11,12)

Adult children are often caught between generations, a scenario commonly referred to as the sandwich generation. They are not only responsible for caring for their aging parents with ADRD but also taking care of their own children, while being the primary breadwinners in their families. Empirical evidence on how caring for parents with ADRD affects their employment is limited.^(4,7) This gap is primary due to the lack of existing surveys that specifically measure the impact of caregiving on employment.^(4,8) The knowledge of the value of unpaid caregiving is often fragmented.⁽¹³⁾

In this study, we aimed to estimate the work absenteeism costs of adult children with ADRD parents due to caregiving. Using a nationally representative dataset, we were able to estimate the work absenteeism costs while controlling caregivers’ demographic and socioeconomic status. This study is among the first to estimate work-related indirect costs of caring for parents with ADRD. Caregiving for ADRD poses special challenges, requiring substantially longer hours and more intensive caregiving activities than caregivers with non-ADRD care recipients.^(3,11,14,15) Hence, we hypothesize that children with ADRD parents will have a higher probability of unemployment, a higher probability of missing any workdays, and longer workdays missed for caregiving compared to children with non-ADRD parents.

METHODS

Study design

This study used data from the 2015-2021 Medical Expenditure Panel Survey (MEPS), a nationally representative survey of the civilian non-institutionalized US population. The MEPS employs a complex overlapping panel design and gathers extensive information on demographic and socioeconomic characteristics, healthcare utilization and expenditures, health status, sources of payments, health insurance coverage, and health conditions.⁽¹⁶⁾ The study merged seven years of MEPS’ consolidated household and medical condition files using individual unique identifiers. The medical conditions files are based on respondents’ self-reported information about dates of healthcare visits, medical conditions, diagnostic codes, and healthcare payments, supplementing the household files. The respondents’ healthcare providers are contacted to validate self-reported information in medical conditions files.⁽¹⁶⁾

Study sample

About 84% of adult child caregivers of older adults in the US are under 65 and below retirement age.^(4,5) Caregivers for ADRD patients are mostly middle-aged daughters and daughters-in-law.^(1,3,4,17) Thus, our samples include children aged 40 and under 65 years.^(1,18) Parent identifiers were used to identify and match with the biological, adopted, or stepchildren and measure whether parents had been diagnosed with ADRD or not. Particularly, our study compared the work absenteeism costs among children who had at least one parent diagnosed with ADRD and those with non-ADRD parents. ADRD was identified by the International Classification of Diseases, Ninth Revision, Clinical Modification (ICD-9-CM) codes (290, 294, 331, or 797)^(19–21) for 2015 data and ICD-10 codes (F01, F03, G30, and G31)^(21,22) for 2016-2021 data.

Measures

Outcome measures

We evaluated the incremental labor productivity costs due to caring for parents with ADRD by examining three stages of labor participation: (1) the likelihood of not working due to unemployment, (2) the likelihood of missing any workdays for the health problems of another family member, and (3) the number of workdays missed due to another’s health problems. Work status was measured by a binary variable equal to “1” if respondents indicated being unemployed during the entire survey year and “0” otherwise. Among employed individuals, respondents reported taking a half-day or more off from work to care for illness, injury, or healthcare needs of another individual in the family and the number of days missed to care for another’s health during the survey year. The three-stage work absenteeism costs have been used to estimate incremental cost of health conditions, such as prostate cancer⁽²³⁾ and rheumatoid arthritis.⁽²⁴⁾

Independent variables

The following covariates were included to control for differences in sample characteristics between children with ADRD parents and those with non-ADRD parents: age; sex; race/ethnicity (Non-Hispanic White, Non-Hispanic Black, Hispanic, or other); marital status (not married or married); education level (no degree, high school diploma or General Educational Development, or above high school diploma); family income as percentage of federal poverty line; health insurance coverage (uninsured, public insurance only, or any private insurance); area of residence (Northeast, Midwest, South, or West); perceived health status (excellent/good or fair/poor); perceived mental health status (excellent/good or fair/poor). These covariates were selected based on the Andersen behavioral model of healthcare utilization⁽²⁵⁾ and have been widely used in the literature to examine healthcare utilization and expenditures.^(26,27)

For the regressions of work absenteeism, we also controlled for types of occupations. Occupations were identified from the Bureau of Labor Statistics occupation codes available in the MEPS. The models controlled for the children’s occupation as follows: management, professionals, sales, services, office, or construction and manufacturing. Individuals working in management or administrative positions comprised the reference group. Variables were also constructed to categorize the child caregivers into three groups: those who were self-employed, those who were not self-employed but received sick pay, and those who were not self-employed and did not have sick pay (the reference group). The variables on occupation types and employment type with sick pay were included only in the work absenteeism analyses.

Statistical analysis

First, sample characteristics between children with and without ADRD parents were presented, and Chi-squared tests were used to test the significance of the differences. Logistic regressions were used to estimate the likelihood of unemployment and that of missing any workday due to caregiving. Odds ratios were reported.

Among those who missed any workdays, a two-part model was used to account for clustered zeros and right skewed distribution in the workdays missed due to caregiving data. First, a logistic regression model was used to estimate the probability of adult children with ADRD parents missing any workday to care for the health problems of another individual in the family. Then, an ordinary least squares estimation was used to ascertain days lost from work due to caring for family member(s) among those who had missed at least a half workday. To estimate marginal effects and predict the number of workdays missed, we used Duan’s smearing estimator and bootstrapped standard errors 1,000 times to adjust for potential log retransformation bias in the second part of the two-part model.^(28,29) Duan’s smearing estimate is calculated by the twopm command for use in post-estimation commands in Stata.^(28,29) Caregiving-related job absenteeism costs were calculated by multiplying the predicted probability of being unemployed, the probability of missing work because of caregiving, and the number of workdays missed for caregiving. The job absenteeism costs were weighted by the average income of six occupations. All the costs were adjusted to 2022 US dollars using the Consumer Price Index.⁽³⁰⁾

Lastly, using the estimated incremental lost productivity costs per person, we provided estimates for the aggregate national annual lost productivity costs for caregiving. We calculated the national estimates by multiplying the estimated job absenteeism costs with the prevalence rates ranging from 53% to 75% for children caregivers who care for their ADRD parents based on the previous literature.^(2–5,8,13) Survey weights were applied to all the estimations. Hence, the results are nationally representative. All analyses were conducted using Stata version 16.

Our study implemented a sensitivity analysis to ensure the robustness of our results (See the appendix). For example, instead of using unemployment status as the dependent variable, we grouped children matched with any parents in the MEPS based on their working status (not working vs. working) and estimated the likelihood of missing any workdays for caregiving among those working. We also tested the different model specifications and tested different age groups of adult children. The results are similar and available upon request.

RESULTS

Our total sample consists of 3,161 individuals. Among them, 207 (6.6%) were children who had at least one parent diagnosed with ADRD. Table 1 showed that children with ADRD parents were not significantly more likely to be unemployed (33.6% vs. 26.2%, p=0.099), but were more likely to have missed work for caregiving (54.6% vs. 27.4%, p<0.001). Among those who had missed at least a half workday, the average number of working days missed did not differ depending on the parental ADRD status (4.0 days vs. 4.5 days, p>=0.541). Compared to children with non-ADRD parents, children who had ADRD parents were older (53.9 years vs. 49.7 years, p<0.001), more likely to be female (64.9% vs. 48.8%, p=0.004), married (43.3% vs. 27.4%, p=0.005), self-employed (22.3% vs. 13.4%) and less likely to have jobs without sick pay (16.5% vs. 29.5%, p=0.043).

Table 1.

Weighted sample characteristics for adult children with and without ADRD parents (Ages 40-64)

	Adult Children with ADRD parents	Adult Children with Non-ADRD parents
	n=207 (6.6%)	n=2,954 (93.4%)
	Weighted % / Mean(SD)	Weighted % / Mean(SD)	p
Dependent variables
Employment status
Unemployed	33.6	26.2	0.099
Missed work for caregiving, if employed
Yes	54.6	27.4	<0.001
No. of workdays missed, if any (unit:days)
	4.0 (4.2)	4.5 (4.8)	0.541
Explanatory variables
Age (unit: years)	53.9 (6.2)	49.7 (6.9)	<0.001
Sex			0.004
Male	35.1	51.2
Female	64.9	48.8
Race			0.471
NH White	54.5	52.2
NH Black	20.1	16.5
Hispanic	17.8	18.9
the rest	7.6	12.4
Marital status			0.005
Not married	56.7	72.6
Married	43.3	27.4
Education			0.486
No degree	9.2	13.2
HS diploma/GED	67.2	61.6
Above HS diploma	23.6	25.2
Family income as % of FPL	496.4 (322.9)	460.6 (343.3)	0.460
Insurance coverage			0.811
Uninsured	13.0	14.9
Public only	26.9	28.0
Any private	60.1	57.1
US census region			0.067
Northeast	18.9	15.8
Midwest	20.1	17.7
South	47.9	41.2
West	13.2	25.3
Perceived health status			0.659
Excellent/Good	73.1	71.0
Fair/Poor	26.9	29.0
Perceived mental health status			0.185
Excellent/Good	84.2	79.5
Fair/Poor	15.9	20.6
Occupation if employed			0.275
Management	19.8	13.3
Professionals	13.7	19.7
Sales	11.8	8.6
Service	19.1	19.6
Office	19.1	12.0
Construction & Manufacturing	16.6	26.8
Job if employed			0.043
Not self-employed without sick pay	16.5	29.5
Not self-employed with sick pay	61.2	57.1
Self-employed	22.3	13.4

Notes: Values are weighted proportions of children with and without ADRD, and weighted means of workdays missed for caregiving, age, family income as percentage of FPL (standard deviation). The weighted population size is 4,072,170 (children with ADRD parents=266,729 and children with non-ADRD parents=3,805,441). p-values are based on Pearson’s chi-square test and adjusted Wald test comparing children with and without ADRD parent groups. Proportions may not sum to 100% because of rounding. Survey years (2015-2021) were included in the analysis but omitted due to limited space in the table.

Abbreviations: ADRD, Alzheimer’s disease and related dementias; NH, non-Hispanic; HS, high school; GED, General Education Development; FPL, Federal Poverty Line

Table 2 presents the results of the association between having parent(s) with ADRD and labor force participation, controlling for demographic characteristics and socioeconomic status. Having parents with ADRD was associated with higher likelihoods of being unemployed (OR=1.80, p=0.024) and 2.95 times more likely to miss work to care for the health problems of another individual in the family (OR=2.95, p=0.002). The difference in the number of workdays missed for caregiving between children with and without ADRD parents was not significant.

Table 2.

Estimation of lost productivity costs due to caring for parents with ADRD in three stages of labor participation

	Model 1: Unemployment		Model 2: Missing workday for caregiving		Model 3: log(workdays missed)
	n=3,161		n=1,606		n=1,606
	OR	p	OR	p	Coef.	p
Children (ref: Without ADRD parent)
With ADRD parent	1.80	0.024	2.95	0.002	−0.20	0.346
Age	1.04	<0.001	1.01	0.544	−0.01	0.454
Sex (ref: Male)
Female	1.09	0.569	2.50	<0.001	0.21	0.072
Race (ref: NH White)
NH Black	0.71	0.094	0.57	0.013	−0.02	0.929
Hispanic	0.48	<0.001	0.47	0.002	−0.10	0.535
The Rest	0.67	0.203	0.67	0.155	−0.12	0.519
Marital status (ref: Not Married)
Married	0.78	0.166	2.06	<0.001	0.21	0.111
Education level (ref: No degree)
HS diploma/GED	0.52	0.001	1.90	0.054	−0.05	0.807
Above HS diploma	0.48	0.009	2.54	0.016	−0.23	0.329
Family income as % of FPL	1.00	<0.001	1.00	0.045	0.00	0.337
Health insurance (ref: Uninsured)
Public only	4.10	<0.001	1.31	0.429	−0.09	0.700
Any private	0.43	<0.001	0.90	0.676	0.03	0.873
US census region (ref: Northeast)
Midwest	1.50	0.171	1.36	0.341	−0.22	0.267
South	2.03	0.007	1.05	0.870	0.00	0.983
West	1.33	0.304	1.35	0.319	0.03	0.876
Perceived Health (ref: Excellent/Good)
Fair/Poor	2.04	<0.001	1.60	0.037	−0.08	0.625
Perceived Mental Health (ref: Excellent/Good)
Fair/Poor	2.04	<0.001	1.64	0.086	0.15	0.453
Job characteristics (ref: Not self-employed without sick pay)
Not self-employed with sick pay	-		1.54	0.084	−0.26	0.057
Self-employed	-		1.05	0.883	0.24	0.257
Occupation (ref: Management)
Professionals	-		0.71	0.236	0.04	0.823
Sales	-		0.75	0.491	−0.01	0.984
Service	-		0.83	0.560	−0.24	0.327
Office	-		0.76	0.388	−0.12	0.619
Construction & Manufacturing	-		1.21	0.551	−0.06	0.783

Notes: Model 1 results are odds ratios from logistic regression on unemployment (n=3,161, df=361) and Model 2 and Model 3 results are odds ratios and coefficients from two-part model with logistic regression on missing any workdays for caregiving in the first part and OLS with logged workdays missed for caregiving in the second part (n=1,606, df=312). Survey years (2015-2021) were added in the analysis but omitted due to limited space in the table. Results are fully adjusted for survey weights. Results of F-adjusted mean residual test and Pregibon’s link test suggest no evidence of model misspecifications or lack of fit. See appendix for the full regression results.

Abbreviations: OR, Odds Ratios; df, degree of freedom; ADRD, Alzheimer’s disease and related dementias; NH, non-Hispanic; HS, high school; GED, General Education Development; FPL, Federal Poverty Line

The odds of unemployment were higher with older age (OR=1.04, p<0.001), those who only had public insurance (OR=4.10, p<0.001) and those living in the South (OR=2.03, p=0.007). Having fair or poor perceived physical and mental health were significantly associated with higher odds of unemployment (OR=2.04, p<0.001). Hispanic individuals (OR=0.48, p<0.001) and those with high school diploma/GED (OR=0.52, p=0.001) and above (OR=0.48, p=0.009) were less likely to be unemployed. Being female (OR=2.50, p<0.001), married (OR=2.06, p<0.001), having education above a high school degree (OR=2.54, p=0.016), and having fair or poor perceived health (OR=1.60, p=0.037) were significantly associated with higher odds of missing work for caregiving. Non-Hispanic Black (OR=0.57, p=0.013) and Hispanic (OR=0.47, p=0.002) individuals were less likely to miss any workdays for caregiving.

The incremental effects of having ADRD parents were calculated in Table 3. Having parents with ADRD increased the probability of unemployment by 8 percentage points and that of missing work by 22 percentage points. Parental ADRD status raised work absenteeism from 1 to 2 days, though statistically insignificant. Using the weighted average of incomes by the occupation types reported in the MEPS, this absenteeism translated into an annual per-person cost increase of $4,510.29 for children with parents who were diagnosed with ADRD. Using the 95% confidence intervals for the predicted estimates, the incremental job absenteeism costs range from $1,702.09 to $6,723.69.

Table 3.

Per capita annual incremental unemployment and missing work costs for children with ADRD parents: 2015-2021

Point Estimate	Children with ADRD Parents	Children with Non-ADRD Parents	Difference
Predicted probability of being unemployed	0.34	0.26	0.08
Predicted probability of missing work due to caregiving	0.50	0.28	0.22
Predicted number of workdays missed due to caregiving	2	1	1
Job absenteeism costs*	$18,900.40	$14,390.11	$4,510.29

Range of Estimate	Lower bound			Higher bound
	Children with ADRD parents	Children with Non-ADRD parents	Difference	Children with ADRD parents	Children with Non-ADRD parents	Difference
Predicted probability of being unemployed	0.27	0.24	0.03	0.41	0.29	0.12
Predicted probability of missing work due to caregiving	0.36	0.24	0.12	0.65	0.31	0.34
Predicted number of workdays missed due to caregiving	1	1	0	3	2	1
Job absenteeism costs*	$14,953.69	$13,251.60	$1,702.09	$22,866.92	$16,143.23	$6,723.69

Notes: The weighted average wage income based on occupations of working individuals aged 40 and below 65 years had $55,177.46; i.e., $212.22 per day (=$55,177.46/260days). The loss of money for children with ADRD parents: 34% x $55,177.46 + (100%-34%) x 50% x $212.22 x 2 days = $18,900.40. The loss of money for children with non-ADRD parents: 26% x $55,177.46 + (100%-26%) x 28% x $212.22 x 1 day = $14,390.11. The difference = $4,510.29. The predicted number of workdays missed due to care for others was not significantly different between children with and without ADRD parents (Table 2, Model 3). If we assume the same number of workdays missed due to caregiving for both groups, the incremental job absenteeism costs would be $4,440.26. The lower and higher bound of estimates represent the 95% confidence interval of the point estimate. The job absenteeism costs are calculated the same as in the point estimate.

Source: Authors’ estimates based on the 2015-2021 Medical Expenditure Panel Survey (MEPS); the average workdays per year (https://hr.uiowa.edu/pay/payroll-services/payroll-calendars/working-day-payroll-calendar-2023)

Aggregate estimates of national absenteeism costs related to having parental ADRD were calculated based on an estimate of 11.5 million informal caregivers for ADRD individuals in the US,⁽³⁾ with the prevalence of children caregivers ranging from 53% to 75% provided by the literature.^(2–5,13) Multiplying the prevalence rates of children with ADRD parents with the incremental job absenteeism costs obtained from Table 3, we estimated that having parents with ADRD increased aggregate annual absenteeism costs by $27.4 billion to $38.8 billion (Table 4).

Table 4.

Estimation of Aggregate Absenteeism Costs

Prevalence of Children with ADRD parents	Job absenteeism costs per person	Estimated total job absenteeism costs
6,083,870(1)	$4,510.29	$27,440,018,022
8,609,250(2)		$38,830,214,183

Notes: Total job absenteeism costs were calculated by multiplying prevalence of children with ADRD parents and job absenteeism costs per person ($4,510.29 from Table 3). (1) 6,083,870 = 11.479 million unpaid caregivers for ADRD patients^(a) x 53% children caregivers for ADRD parents^(b); (2) 8,609,250 = 11.479 million unpaid caregivers for ADRD patients^(a) x 75% working-age children caregivers for older adults^(c).

Source: (a): Alzheimer’s Association, 2022, (b): Alzheimer’s Association, 2022; National Alliance for Caregiving in Partnership with Alzheimer’s Association, 2017; Stokes et al., 2023; Wolff et al., 2018, (c): Mudrazija, 2019.

DISCUSSION

Our study presents the annual average lost productivity costs per person among 40- to 64-year-old children with parents who had ADRD diagnoses and its aggregate national estimate. We found that having at least one parent with ADRD was significantly associated with a higher likelihood of being unemployed and missing any workday to care for the health problems of another individual in the family. Having a parent with ADRD was also associated with a higher number of missed workdays, but the association was not significant. This study extends the limited literature by providing nationally representative estimates of work absenteeism of adult children who have ADRD parents directly related to caregiving after controlling for a wide range of individual characteristics.

Understanding the absenteeism-related productivity costs for children with ADRD parents is critical for policymakers, employers, and family caregivers. Caregiving for ADRD persons is extremely challenging and incurs greater caregiving and financial burdens compared to non-ADRD caregivers. Their caregiving tasks are extensive, ranging from self-care assistance to medical care and making decisions on behalf of ADRD patients.⁽³⁾ Caregivers of ADRD persons also provide 27 hours more care per month on average than non-ADRD caregivers,⁽¹⁵⁾ and the hours for caregiving increases along the ADRD continuum.^(3,11) The time required for caregiving at the onset of dementia increased from 151 hours per month to 283 hours per month eight years later.⁽¹¹⁾ Moreover, ADRD caregivers are more likely to experience emotional stress and depression, have health problems, and report financial strains.^(3,31,32) ADRD caregivers also incur nearly twice the average out-of-pocket costs per year than non-ADRD caregivers.⁽³⁾ Although our study focused on the indirect cost of caregivers, evidence of ADRD caregivers’ direct cost was also substantial.^(9,10) The financial burden of care was substantially greater for individuals at or below the 200% of the FPL, which may impact their decisions on job absenteeism more challenging.^(9,10)

Furthermore, working-age adult children with ADRD parents may also face challenges of their own, such as taking care of their children. Reducing work hours, switching to a part-time job, or temporarily exiting the labor market to take care of their parent with dementia may also have longer impacts on child caregivers, such as losing employer-sponsored health insurance, having difficulties returning to the job market in the future, and reducing savings for their future retirement.^(8,32) As we expect the number of ADRD patients to grow while that of potential family caregivers per ADRD patient to decline, an increasing number of child caregivers will likely bear even higher caregiving burdens and work-related opportunity costs of caregiving in the coming years.^(3,13,33)

The opportunity costs of caregiving by foregone earnings alone are estimated to be $67 billion now and to reach $132-$147 billion in 2050.⁽¹³⁾ Our study suggests that the aggregate lost productivity costs borne by children with ADRD parents would account for 41% to 58% of the current total opportunity costs of caregiving. The work-related opportunity costs of caregiving should be considered in addition to direct healthcare costs when measuring the costs of ADRD to develop policies designed to alleviate both the societal and financial burden of family caregivers. High levels of stress in informal caregivers have been associated with adverse outcomes on care recipients and their use of healthcare services.^(3,31)

Overall, our findings are consistent with previous findings that having parents with ADRD adversely impacts the ability of working-age adult children to do paid work and miss any workday for caregiving.^{(3,4,8,13,33)} Reporting fair or poor health had consistent adverse impacts on adult children’s probability of unemployment and missing any workday.^(3,33) In line with the findings that daughters make up over one-third of dementia caregivers,^(3,14,15) we also found that women were two and half times more likely than men to miss work for caregiving when employed, but no significant difference was found in the number of workdays missed.

The previous literature has found that Black and Hispanic caregivers work longer hours and provide more intensive care than White caregivers regardless of the older adults’ dementia status, which will likely result in more frequent missing workdays for caregiving among racial and ethnic minorities.^(2,7,34,35) We found that Hispanic individuals had significantly lower probability of unemployment after adjusting for parental ADRD status and other covariates. However, employed non-Hispanic Black and Hispanic individuals were less likely to miss any workdays for family caregiving. Stokes et al. (2023) also found no racial and ethnic differences in employment and work hour changes among adult children following parental ADRD onset, highlighting the interplay between structural disadvantages and the norms of family support among racial and ethnic minorities.^(4,36) Structural inequities, such as fewer financial resources and limited access to healthcare, borne by racial and ethnic minorities may incentivize the adult children of racial and ethnic minorities to stay in employment or even increase work hours to meet their needs, including ADRD care costs for their parents.^(4,36) For example, Black male dementia caregivers were three times more likely to experience financial burdens than White female dementia caregivers.^(3,37) Our study sample also indicates that racial and ethnic minorities may have jobs that are more restrictive in time and require in-person work, making it difficult to miss workdays for caregiving. While about 50% of the non-Hispanic White individuals had management or professional jobs, about 46% of non-Hispanic Black and 56% of Hispanic individuals worked in services or construction and manufacturing jobs. Moreover, about 30% of Hispanic individuals in the study sample had jobs without sick pay, compared to 19.5% of non-Hispanic White individuals.

On the other hand, family obligations and social stigmatization stemming from the fear of what others may think if they do not take care of their ill parents may work as barriers to seek support from outside and lead children to provide care to ADRD parents themselves.^(38–40) These findings suggest that the need to make ends meet may have been a stronger force than feeling guilt in work-related decisions, at least among our sample of employed adult children with ADRD parents. Our results also underscore that what used to be determinative factors for unemployment/employment may no longer be important factors at different stages of caregiving decisions.

Though our study is among the first to estimate the annual incremental lost productivity costs of adult children with ADRD parents to care for the health problems of another individual in the family, it has important limitations. First, the MEPS data excludes institutionalized population; thus, these findings may not be generalizable to facility dwellers. Our adult children sample in the MEPS dataset was limited to those identified through unique parent identifiers, leaving out those who could have parents with ADRD due to having no parent identifiers. The limited sample size of adult children with ADRD parents also constrained further examination of job absenteeism among subgroups, such as low-income families. Also, we could not account for the severity of parents’ ADRD symptoms, nor adult children’s work hours, intensity, and regularity of care, which may impact the work-related outcomes of adult children caregivers, due to data unavailability in the MEPS. Moreover, the pooled cross-sectional study design and unavailability of information on the parental onset of ADRD limited us to establish a causal relationship between having parental ADRD and work-related outcomes. Finally, though our measure of missing any workdays and the number of workdays missed were directly related to family caregiving activities, the MEPS did not specify to whom the adult children are providing care. Our study included all adult children regardless of whether they provide direct care to their ADRD parents; thus, our findings better reflect the potential impact of having parents with ADRD on the work absenteeism of adult children.⁽⁴⁾

Future studies on the impact of parental ADRD on adult children’s work-related outcomes should consider how the onset of ADRD in parents may impact the work-related outcomes of their children based on the caregivers’ care intensity and regularity. Given different levels of perceptions on caregiving and structural disadvantages among racial and ethnic minorities, more research is needed to examine how caregiving for ADRD parents results in different consequences for caregivers across various racial and ethnic groups.

CONCLUSION

Informal caregiving for ADRD requires considerable time and commitment, often putting caregivers under high financial strain due to caregiving activities, including work absenteeism.^(2,3,32) Our study provides robust estimates of the lost labor productivity costs among adult children with parents diagnosed with ADRD compared with children with non-ADRD parents. Having ADRD parents significantly increased the chances of unemployment and missing any workdays for caregiving if employed, leading to higher lost labor productivity costs for adult children with ADRD parents. Understanding these costs is vital as financial strain may significantly increase the amount of stress on informal caregivers, who are primarily adult children among ADRD caregivers,^(3–5,31) and likely harm care recipients.^(3,31) Thus, policymakers and employers must acknowledge career disadvantages among ADRD caregivers and provide tangible support, such as tax credits and flexible work schedules for employed caregivers, to promote a better quality of life for both ADRD patients and their families.

Highlights.

• What is the primary question addressed by this study?

  Using nationally representative data, we aimed to estimate the adjusted work absenteeism costs of adult children with at least one parent diagnosed with Alzheimer’s disease and related dementias (ADRD) due to caregiving.

• What is the main finding of this study?

  Having at least one parent with ADRD significantly increased the chances of unemployment and missing any workdays for caregiving if employed, leading to higher lost labor productivity costs for adult children with ADRD parents.

• What is the meaning of the finding?

  Policymakers must consider the work-related opportunity costs of caregiving in addition to direct healthcare costs when measuring the costs of ADRD to develop policies designed to alleviate both the societal and financial burden of family caregivers.

Data sharing statement:

The dataset used in our study is publicly available at Agency for Healthcare Research and Quality (AHRQ)’s Medical Expenditure Panel Survey (MEPS) website. The data has not been previously presented orally or by poster at scientific meetings.