Skip to main content
NIHPA Author Manuscripts logoLink to NIHPA Author Manuscripts
. Author manuscript; available in PMC: 2024 Mar 1.
Published in final edited form as: J Am Geriatr Soc. 2022 Nov 25;71(3):799–809. doi: 10.1111/jgs.18138

A National Profile of Sandwich Generation Caregivers Providing Care to Both Older Adults and Children

Lianlian Lei a, Amanda N Leggett b, Donovan T Maust a,c,d
PMCID: PMC10023280  NIHMSID: NIHMS1845906  PMID: 36427297

Abstract

Background:

Many U.S. caregivers provide care to the generation above and below simultaneously, described as “sandwich” generation caregivers. We seek to provide the first national estimates characterizing sandwich generation caregivers and the older adults for whom they care.

Methods:

We used 2015 National Study of Caregiving (NSOC) and National Health and Aging Trends Study (NHATS) to compare individual (demographic, socio-economic, health, and caregiving characteristics) and caregiving-related experience (financial and emotional difficulties, caregiver role overload and gains, supportive services, employment and participation restrictions) between sandwich and non-sandwich generation caregivers. The analysis included adult child caregivers with or without any minor child under 18 years (n=194 and 912 NSOC respondents, respectively) providing care to n=436 and 1,217 older adult NHATS respondents.

Results:

Of all adult child caregivers, 24.3% also cared for a minor child (i.e., sandwich generation caregivers), representing 2.5 million individuals. Sandwich generation caregivers provided similar care hours to older care recipients as non-sandwich caregivers (77.4 vs. 71.6 hours a month, p=.60), though more of them worked for pay (69.4% vs. 53.9%, p=.002). Both sandwich generation caregivers (21.0% vs. 11.1%, p=.005) and their care recipients (30.1% vs. 20.9%, p=.006) were more likely to be Medicaid enrollees than their non-sandwich caregiving counterparts. More sandwich generation caregivers reported substantial financial (23.5% vs. 12.2%, p<.001) and emotional difficulties (44.1% vs. 32.2%, p=.02) than non-sandwich caregivers; they also reported higher caregiver role overload (score: 2.9 vs. 2.4, p=.04). Their supportive services use was similarly low as non-sandwich caregivers except for seeking financial help (24.8% vs. 14.7%, p=.008).

Conclusions:

Besides caring for minor child(ren), sandwich generation caregivers provided similarly intense care to care recipients as non-sandwich caregivers and had higher labor force participation; they experienced more caregiving-related financial and emotional difficulties and role overload. Policymakers may consider supportive services that address their unique needs and roles.

Keywords: sandwich generation, intensity of care, financial difficulties, emotional difficulties, caregiver role overload

INTRODUCTION

Due to the demographic shifts such as delayed fertility and longer life expectancy, many caregivers in the U.S. provide care to both the generation above (i.e., older adult care) and below (i.e., child care) simultaneously, described as “sandwich” generation caregivers.15 The proportion of adult child caregivers living with a minor child under 18 years more than doubled from 1999 to 2015, growing from 12.6% to 26.0%.6 The growing size of this group may help account for “support of the sandwich generation” having been an explicit goal of the Biden Administration’s Build Back Better infrastructure proposal.7

A recent survey conducted by the U.S. Centers for Disease Control and Prevention highlights the potential toll of this dual-caregiving role: during the COVID-19 pandemic, 51.5% of sandwich generation caregivers endorsed serious past-month suicidal ideation (compared to 16.0% of caregivers without a minor child), with odds eight times higher than respondents who were neither parents nor caregivers.8 And while these caregivers have received recent policy attention, there is surprisingly little research specifically on the sandwich generation caregiving group. The limited work available is based on interviews with relatively small samples,916 drawn from international cohorts,1721 or uses an alternative definition of sandwich generation caregivers (e.g., including those caring for the same generation such as spouses or siblings).22,23

Given the array of caregivers who work to support older adults, resource-constrained policies and interventions would ideally be tailored to the needs of particular caregivers in order to maximize their impact, in the same way that clinical medicine aspires to use risk stratification to tailor care based on an individual patient’s risk profile. To further this goal, the purpose of this analysis is to explore the construct of sandwich generation caregivers to better understand who they are, to whom they provide care, and their caregiving-related experience relative to other caregivers. This study uniquely provides the first nationally representative analysis of both sandwich generation caregivers and the older adults to whom they provide care by pairing the National Health and Aging Trends Study (NHATS), a nationally representative survey of age-eligible Medicare beneficiaries in the U.S., with the National Study of Caregiving (NSOC), a nationally representative survey of caregivers who support NHATS respondents.

METHODS

Data and study populations

We used the 2015 National Health and Aging Trends Study (NHATS) and National Study of Caregiving (NSOC), ongoing longitudinal, nationally representative surveys of U.S. older adults and their caregivers.24,25 Using both NHATS and NSOC uniquely captures characteristics of both sandwich generation caregivers and the older adults for whom they care. For this analysis, we considered sandwich generation caregivers to be those adult children caregivers who provided care to both a parent aged ≥65 years and at least one minor child under 18 years.6

Participants were eligible for NHATS if they were Medicare enrollees aged ≥65 years who lived in the contiguous U.S.24 The response rate was 73.2%. Of 7,576 NHATS respondents, we limited our analysis to those who received assistance with mobility, self-care, or household activities due to health or functioning reasons from at least one adult child caregiver (n=1,653): 436 (32.9%) received assistance from at least one adult child with any minor child under 18 years (i.e., sandwich generation caregiver)6 and 1,217 (67.1%) from adult children without any minor child (i.e., non-sandwich generation caregiver; Supplementary Figure S1).

NSOC is a supplement of NHATS that surveys family and unpaid caregivers to NHATS participants who received assistance with mobility, self-care, or household activities due to health or functioning reasons.25 Up to five caregivers per care recipient could be included. The response rate was 61.9%. Of 1,653 NHATS respondents receiving assistance from adult child caregivers, 1,024 had at least one family and unpaid caregiver participating in NSOC; in 2015, there was a total of 1,676 NSOC respondent caregivers. We limited our analysis to adult child caregivers (n=1,106 for 735 NHATS respondents): 194 (24.3%) with any minor child under 18 years (i.e., sandwich generation caregiver) and 912 (75.7%) without any minor child (i.e., non-sandwich generation caregiver; Supplementary Figure S1).

Caregiver characteristics

Information about caregivers were obtained from NSOC. We examined caregivers’ socio-demographics (sex, age, race/ethnicity), socioeconomic status (marital status, education, family income, Medicaid enrollment), general health status, and caregiving characteristics (co-residence with care recipient, whether care recipient was living in the community, duration of caregiving, hours of care provided in the last month). Hours of care were imputed for those with missing values.26 In addition, using the NHATS Other Person file to determine the total amount of help provided to a care recipient, we derived the proportion of total help hours the NSOC respondent caregiver provided, whether they were the sole caregiver (i.e., the only caregiver) or primary caregiver (i.e., providing the highest proportion of total help hours among all caregivers in their network), and number of functional disability-related activities (mobility, self-care, and household activities) for which they provided assistance.

Care recipient characteristics

Information about care recipients’ characteristics was drawn from NHATS. We examined care recipients’ socio-demographics (sex, age, race/ethnicity), socio-economic characteristics (marital status, education, family income, Medicaid enrollment, living alone, living in the community versus residential care settings), general health and functional status, and self-reported comorbidities (heart disease, diabetes, lung disease, stroke, cancer, arthritis, hypertension, dementia, depression, anxiety). We measured functional impairment by count of self or proxy report of difficulties in performing or needing help with the following tasks: mobility (getting out of bed, getting around inside and outside [0–3]), self-care (bathing, dressing, eating, and using the toilet [0–4]), and household activities (doing laundry, going shopping, preparing meals, handling banking, and managing medications [0–5]).27 Information about the nature and intensity of care received by each care recipient was obtained from the NHATS Other Person file, which includes all unique persons that the NHATS respondent identified as providing any functional assistance, along with the hours of care each helper provided. For each care recipient (i.e., NHATS respondent), we calculated the total number of family and unpaid caregivers and the total hours of care received from them in the last month. Hours of care were imputed for those with missing values.26 Family and unpaid caregivers include individuals related to the care recipient such as spouse, adult children, grandchildren or siblings, as well as unrelated, unpaid helpers such as friends or neighbors.25

Caregiving-related experience and employment

Finally, after characterizing the profile of caregivers and their care recipients, we used NSOC to describe the experience of these caregivers, including caregiving-related financial and emotional difficulties, caregiver role overload and gains, use of supportive services, and caregivers’ employment and participation restrictions.

Caregivers reported whether providing care was financially or emotionally difficult and, if yes, the extent of difficulties, from 1 (a little difficult) to 5 (very difficult); we categorized a response of 1–2 as “some” difficulty and 3–5 as “substantial”.28

Caregiver role overload included four items measuring feelings of exhaustion and fatigue related to caregiving on a scale of 0 (not so much) to 2 (very much): (1) feeling exhausted, (2) having more things than they can handle, (3) having no time for themselves, and (4) care recipients’ needs changing frequently. Caregiving-related gains included four items: (1) feeling more confident about their abilities, (2) learning how to deal with difficult situations, (3) bringing caregiver closer to care recipient, and (4) giving caregiver satisfaction. The total overload and gain scores ranged from 0–8, with higher scores indicating higher levels of caregiver role overload or gain (Cronbach’s α = .73 and .71, respectively).29 We also examined whether the caregiver reported “very much” for the separate role overload and gain items.

We examined caregivers’ use of supportive services, including support group participation, respite care, caregiving training, and financial support including helping care recipient apply for Medicaid.30

Finally, we examined whether the caregiver worked for pay (i.e., from non-caregiving employment) and, among those who did, productivity loss due to caregiving including any missed hours of work because of caregiving in the last month, absenteeism (proportion of missed hours among total hours worked typically), and presenteeism (degree to which the caregiver reported caregiving affected productivity when at work; 0–100%).31 Participation restrictions refer to activities limited because of caregiving, including visiting friends or family, attending religious or group activities, and going out for enjoyment.28

Statistical analysis

In this descriptive analysis, we first examined characteristics of sandwich generation caregivers compared to non-sandwich generation caregivers (i.e., NSOC respondents), and older adults (i.e., NHATS respondents) who received assistance from a sandwich generation caregiver (i.e., “sandwich care recipients” hereafter) compared to those receiving care from non-sandwich adult child caregivers. We then examined caregiving-related experience of sandwich generation caregivers compared to non-sandwich generation caregivers. We compared differences between each group using adjusted Wald tests and Rao-Scott chi-square tests for continuous and categorical characteristics, respectively.

We completed sensitivity analyses with a broader definition of sandwich generation caregivers, including in-laws (e.g., daughters-in-law) and grandchildren of the care recipients who also had a minor child.14,16 We also revisited caregiver experience using multivariate regression analysis to account for caregivers’ demographics, socio-economic status, and caregivers and care recipients’ health status.

For all analyses, we used NSOC and NHATS analytic weights to take into account differential probabilities of selection and nonresponse and provide nationally representative estimates for caregivers and the older adults receiving assistance. We used STATA svy commands to incorporate strata and clustering elements of the sample design when calculating standard errors of the estimates. All analyses were performed using STATA, version 15.1.

RESULTS

In 2015, 194 NSOC respondents cared for both an adult ≥65 years as well as a minor child, representing 2.5 million individuals and 24.3% of all adult child caregivers (Table 1). These sandwich generation caregivers were younger (46.3 vs. 56.5 years, p<.001) and more likely to be Medicaid recipients (21.0% vs. 11.1%, p=.005) compared to non-sandwich generation caregivers, but did not vary from non-sandwich generation caregivers across other demographic, socioeconomic, and health characteristics. Sandwich generation caregivers were more likely to assist care recipients that were living in the community than their non-sandwich counterparts (91.3% vs. 78.0%, p=.004) and had been a caregiver for fewer years (5.5 vs. 7.9 years, p<.001). Sandwich and non-sandwich generation caregivers provided similar hours of care in the last month (77.4 vs. 71.6 hours, p=.60), accounted for similar proportions of total caregiving hours that care recipients received, and were similarly likely to be the sole or primary caregiver.

Table 1.

Characteristics of Adult Child Caregivers With and Without Minor Children in a Nationally Representative Sample, 2015

Adult Child Caregivers (N=1,106)
Sandwich generation Non-sandwich generation P-value
Respondents, N (weighted %) National estimate, millions Respondents, N (weighted %) National estimate, millions
Overall 194 (24.3) 2.5 912 (75.7) 7.7
Demographics
 Female, No. (%) 122 (57.7) 1.4 639 (64.5) 4.9 0.18
 Age, years (SD) 46.3 (6.3) 56.5 (9.8) <0.001
  <45, No. (%) 58 (35.2) 0.9 41 (8.8) 0.7 <0.001
  45–64, No. (%) 133 (63.9) 1.6 647 (73.0) 5.6
  65+, No. (%) 224 (18.1) 1.4
 Race/ethnicity, No. (%)
  Non-Hispanic white 93 (62.1) 1.5 554 (70.0) 5.4 0.37
  Non-Hispanic black 70 (15.8) 0.4 262 (12.3) 0.9
  Hispanic 20 (14.2) 0.3 49 (10.6) 0.8
  Other 11 (7.9) 0.2 47 (7.1) 0.5
Socioeconomic status
 Married, No. (%) 111 (64.0) 1.6 480 (58.1) 4.5 0.22
 Some college or above, No. (%) 146 (73.6) 1.8 612 (68.2) 5.3 0.33
 Family income, $, mean (SD) $73,454 (53,317) $66,000 (79,951) 0.19
 Medicaid enrollment, No. (%) 39 (21.0) 0.5 98 (11.1) 0.8 0.005
General health status, No. (%)
  Very good or excellent 107 (60.5) 1.5 484 (52.1) 4.0 0.08
  Good 52 (24.1) 0.6 228 (22.9) 1.7
  Poor or fair 35 (15.4) 0.4 200 (25.0) 1.9
Caregiving characteristics
 Co-residence with care recipient, No. (%) 67 (32.6) 0.8 309 (27.4) 2.1 0.31
 Care recipients living in the community, No. (%) 181 (91.3) 2.3 741 (78.0) 6.0 0.004
 Duration of caregiving, years (SD) 5.5 (5.2) 7.9 (9.5) <0.001
 Hours of care provided in the last month, n (SD) 77.4 (106.5) 71.6 (125.9) 0.60
 Proportion of total help hours provided, % (SD) 40.3 (31.3) 45.2 (38.1) 0.19
 Sole caregiver, No. (%) 22 (11.1) 0.3 137 (13.2) 1.0 0.57
 Primary caregiver, No. (%) 83 (39.1) 1.0 516 (48.3) 3.7 0.08
 Functional disability-related activities helped:
  Mobility (0–3), n (SD) 0.6 (0.7) 0.5 (0.9) 0.64
  Self-care activities (0–4), n (SD) 0.3 (0.8) 0.4 (1.0) 0.50
  Household activities (0–5), n (SD) 1.7 (1.3) 1.7 (1.7) 0.79

Notes: Authors’ analysis of data from the 2015 National Study of Caregiving (NSOC) for a cohort of adult child caregivers. Adjusted Wald tests were performed to compare continuous characteristics and Rao-Scott chi-square tests were performed to compare categorical characteristics among adult child caregivers with or without any minor child under 18 years of age (i.e., sandwich vs. non-sandwich generation caregivers). Data were weighted using the NSOC survey analytic weights.

Estimates based on too few cases (< 11) may not be reported, per NHATS.

In 2015 among NHATS respondents, 32.9% received care from at least one sandwich generation caregiver (Table 2). Sandwich care recipients were younger (76.3 vs. 82.9 years; p<.001), more likely to be married (42.0% vs. 29.8%; p<.001), Medicaid enrollees (30.1% vs. 20.9%; p=.006), and living in the community (86.6% vs. 76.3%; p<.001). The functional and health status of older adults who did and did not receive help from a sandwich generation caregiver were generally similar. Sandwich care recipients had more family and unpaid caregivers (≥3: 49.1% vs. 34.9%; p<.001) and received more total hours of care in the last month (173 vs. 133 hours, p=.005).

Table 2.

Characteristics of Adults ≥65 Years Receiving Assistance from Adult Child Caregivers With and Without Minor Children in a Nationally Representative Sample, 2015

Older Adults Receiving help from Adult Child Caregivers (N=1,653)
≥1 sandwich generation Non-sandwich generation P-value
Respondents, N (weighted %) National estimate, millions Respondents, N (weighted %) National estimate, millions
Overall 436 (32.9) 2.2 1,217 (67.1) 4.4
Demographics
 Female, No. (%) 310 (70.7) 1.5 913 (73.5) 3.2 0.21
 Age, years (SD) 76.3 (7.4) 82.9 (9.6) <0.001
  65–74, No. (%) 147 (48.0) 1.1 109 (17.0) 0.8 <0.001
  75–84, No. (%) 179 (36.2) 0.8 422 (37.1) 1.6
  85+, No. (%) 110 (15.8) 0.3 686 (45.9) 2.0
 Race/ethnicity, No. (%)
  Non-Hispanic white 238 (66.2) 1.4 773 (74.1) 3.3 0.10
  Non-Hispanic black 127 (13.8) 0.3 310 (11.0) 0.5
  Hispanic 47 (12.2) 0.3 81 (8.8) 0.4
  Other 24 (7.8) 0.2 53 (6.0) 0.3
Socioeconomic status
 Married, No. (%) 154 (42.0) 0.9 299 (29.8) 1.3 <0.001
 Some college or above, No. (%) 166 (40.9) 0.9 439 (37.3) 1.7 0.19
 Family income, $, mean (SD) $34,825 (42,354) $35,014 (84,562) 0.95
 Medicaid enrollment, No. (%) 138 (30.1) 0.7 288 (20.9) 0.9 0.006
 Living alone, No. (%) 143 (29.8) 0.6 501 (42.2) 1.9 <0.001
 Living in the community, No. (%) 378 (86.6) 1.9 962 (76.3) 3.4 <0.001
Health conditions
 General health status, No. (%)
  Very good or excellent 73 (17.4) 0.4 255 (21.0) 0.9 0.11
  Good 133 (29.2) 0.6 400 (32.0) 1.4
  Poor or fair 230 (53.3) 1.2 562 (47.0) 2.1
 Functional limitations
  Mobility (0–3), n (SD) 1.8 (1.2) 1.6 (1.4) 0.053
  Self-care activ ities (0–4), n (SD) 1.8 (1.4) 1.6 (1.7) 0.09
  Household activities (0–5), n (SD) 2.9 (1.7) 3.0 (2.0) 0.51
 Comorbidities
  Number of conditions (0–10), n (SD) 3.8 (1.8) 3.8 (2.1) 0.94
  Heart disease, No. (%) 179 (40.8) 0.9 485 (38.6) 1.7 0.48
  Diabetes, No. (%) 187 (43.2) 0.9 385 (32.5) 1.4 0.005
  Lung disease, No. (%) 108 (24.6) 0.5 299 (25.6) 1.1 0.68
  Stroke, No. (%) 87 (18.4) 0.4 287 (22.3) 1.0 0.13
  Cancer, No. (%) 105 (25.1) 0.5 369 (31.2) 1.4 0.04
  Arthritis, No. (%) 330 (72.9) 1.6 906 (73.1) 3.2 0.96
  Hypertension, No. (%) 348 (77.7) 1.7 963 (76.6) 3.4 0.70
  Dementia, No. (%) 135 (28.2) 0.6 447 (32.5) 1.4 0.14
  Depression, No. (%) 124 (27.8) 0.6 310 (26.4) 1.2 0.67
  Anxiety, No. (%) 89 (23.3) 0.5 251 (22.3) 1.0 0.76
Intensity of care received
Number of family and unpaid caregivers, No. (%)
  1 44 (10.7) 0.2 257 (22.2) 1.0 <0.001
  2 162 (40.1) 0.9 501 (43.0) 1.9
  3+ 230 (49.1) 1.0 459 (34.9) 1.5
 Total hours of care received from family and unpaid caregivers in the last month, n (SD) 173.3 (254.6) 133.1 (251.8) 0.005

Notes: Authors’ analysis of data from the 2015 National Health and Aging Trends Study (NHATS) for a cohort of older adults who received assistance with mobility, self-care, or household activities due to health or functioning reasons from at least one adult child caregiver. Adjusted Wald tests were performed to compare continuous characteristics and Rao-Scott chi-square tests were performed to compare categorical characteristics among older adults who received assistance from at least one adult child with any minor child under 18 years of age (i.e., sandwich generation caregiver) and those receiving assistance from adult children without any minor child under 18 years of age (i.e., non-sandwich generation caregiver). Data were weighted using the NHATS survey analytic weights.

Among the 436 older adults receiving help from at least 1 sandwich generation caregiver, 234 (58.8%) received help from only sandwich generation caregivers and 202 (41.2%) received help from both sandwich and non-sandwich adult child caregivers.

Sandwich generation caregivers were more likely to experience financial difficulties (substantial: 23.5% vs. 12.2%, p<.001), emotional difficulties (substantial: 44.1% vs. 32.2%, p=.02), and report higher caregiver role overload (composite overload score: 2.9 vs. 2.4, p=.04); caregiving-related gains did not differ (Table 3). Their use of services to support caregiving were similar, though sandwich generation caregivers were more likely to seek financial help for the care recipient, such as helping them apply for Medicaid (24.8% vs. 14.7%, p=.008). Sandwich generation caregivers were more likely to work for pay than non-sandwich generation caregivers (69.4% vs. 53.9%, p=.002), but among those who worked, productivity loss due to caregiving did not differ.

Table 3.

Characteristics of Caregiving-related Experience and Employment Participation among Adult Child Caregivers With and Without Minor Children

Adult Child Caregivers (N=1,106)
Sandwich generation Non-sandwich generation p-value
Respondents, N (weighted %) National estimate, millions Respondents, N (weighted %) National estimate, millions
Overall 194 (24.3) 2.5 912 (75.7) 7.7
Caregiving-related difficulties
 Reported financial difficulties. No. (%)
  None 127 (64.2) 1.6 748 (83.2) 6.4 <0.001
  Some 21 (12.3) 0.3 46 (4.6) 0.3
  Substantial 46 (23.5) 0.6 118 (12.2) 0.9
 Reported emotional difficulties, No. (%)
  None 90 (45.4) 1.1 482 (50.5) 3.9 0.02
  Some 27 (10.5) 0.3 163 (17.4) 1.3
  Substantial 77 (44.1) 1.1 267 (32.2) 2.5
Caregiver role overload
Composite overload score (0–8), n (SD) 2.9 (2.0) 2.4 (2.4) 0.04
Exhausted, No. (%) 58 (27.4) 0.7 187 (19.6) 1.5 0.11
More than they can handle, No. (%) 45 (25.1) 0.6 158 (16.7) 1.3 0.07
No time for self, No. (%) 45 (19.0) 0.5 170 (16.5) 1.3 0.48
Care recipient’s needs change frequently, No. (%) 17 (7.7) 0.2 91 (10.5) 0.8 0.34
Caregiving-related gains
Composite gains score (0–8), n (SD) 6.2 (1.7) 6.0 (2.1) 0.49
More confident, No. (%) 100 (45.1) 1.1 445 (43.7) 3.4 0.79
Deal with difficult situations, No. (%) 121 (58.4) 1.5 542 (52.6) 4.0 0.29
Closer to care recipient, No. (%) 142 (70.3) 1.7 641 (67.3) 5.2 0.57
Satisfaction, No. (%) 175 (86.7) 2.1 812 (87.7) 6.7 0.79
Use of supportive services, No. (%)
Support group participation 7 (4.5) 0.1 37 (3.7) 0.3 0.68
Respite care 44 (18.8) 0.5 203 (19.9) 1.5 0.77
Caregiver training 19 (10.6) 0.3 75 (6.6) 0.5 0.15
Identify potential financial support for care recipient 52 (24.8) 0.6 132 (14.7) 1.1 0.008
Use of ≥1 supportive service 87 (41.3) 1.0 340 (33.6) 2.6 0.11
Employment and participation in other activities
Work for pay, No. (%) 130 (69.4) 1.7 433 (53.9) 4.2 0.002
Caregiving prevents from working, No. (%) 36 (17.2) 0.4 108 (11.4) 0.9 0.07
Among those who worked:
Missed work due to caregiving, No. % 36 (17.4) 0.3 173 (16.6) 0.7 0.86
Absenteeism, % (SD) 2.2 (6.7) 1.5 (5.0) 0.47
Presenteeism, % (SD) 6.6 (15.6) 7.0 (20.4) 0.83
Caregiving preventing from participation in valued activities, No. (%)§ 44 (21.4) 0.5 236 (21.1) 1.6 0.94

Notes: Authors’ analysis of data from the 2015 National Study of Caregiving (NSOC) for a cohort of adult child caregivers. Adjusted Wald tests were performed to compare continuous characteristics and Rao-Scott chi-square tests were performed to compare categorical characteristics among adult child caregivers with and without any minor child under 18 years of age (i.e., sandwich vs. non-sandwich generation caregivers). Data were weighted using the NSOC survey analytic weights.

Absenteeism referred to the proportion of hours of work missed because of caregiving in the last month among total hours worked typically.

Presenteeism referred to the degree to which the caregiver reported caregiving affected productivity when at work.

§

Valued activities included visiting with friends or family, attending religious services, participating in group activities, and going out for enjoyment.

Sensitivity analyses showed consistent results with the main findings when using a more broadly defined group of sandwich generation caregivers that included in-laws (e.g., daughter-in-law) and grandchildren of the care recipient (who also had a minor child; Supplementary Tables S1S2). The adjusted results of caregiving-related experience of sandwich generation caregivers were also consistent with those in the unadjusted main analysis (Supplementary Table S3).

DISCUSSION

Although there is no universally accepted definition of sandwich generation caregivers in the existing literature,32 this study provides a nationally representative analysis of adult caregivers caring for both a parent and a minor (i.e., <18 years) child. In 2015, nearly one quarter of adult child caregivers or an estimated number of 2.5 million individuals provided care to both the generation above and below. Nearly 70% worked for pay, and nearly one quarter endorsed substantial financial difficulty and nearly one half endorsed substantial emotional difficulty. Using NHATS and NSOC together uniquely captures characteristics of both sandwich generation caregivers and the older adults for whom they care, whereas other national surveys (e.g., Behavioral Risk Factor Surveillance System and American Time Use Survey) have only captured the caregivers.5,22

Our analysis highlights the extraordinary level of demand faced by these individuals. By a variety of measures, sandwich generation caregivers provided as much care to the generation “above” as did non-sandwich generation caregivers —whether measured as total care hours, proportion of care hours, sole caregiver, or types of functional assistance provided. However, by definition, sandwich generation caregivers were also providing care “below” to a generation of minor children. On average, child care is more time consuming than care provided to older adults.5 So, while not assessed as part of NSOC, this additional unmeasured caregiving contributes to a significant total amount of caregiving time and effort from sandwich generation caregivers. Furthermore, we found that sandwich generation caregivers were also more likely to be in the paid workforce. The trilemma of older adult care, child care, and employment may impose unique challenges on the economic and health wellbeing of sandwich generation caregivers.3

Older adults receiving assistance from sandwich generation caregivers received more intense care—measured both in terms of the total number of care hours received as well as the total number of caregivers in their network. This additional caregiving may help allow more of these care recipients to remain in a community setting by virtue of the additional caregiving hours provided by their family and other unpaid caregivers.33 This is particularly notable given a higher proportion of care recipients of sandwich generation caregivers had Medicaid, which would be the primary payor of their nursing home care. While our analysis raises this potential hypothesis, a longitudinal analysis is needed to test whether this potential delay of institutionalization is, in fact, occurring.

However, this additional care is associated with potential costs for these caregivers, who reported more caregiving-related financial and emotional difficulties and higher caregiver role overload: nearly one quarter reported substantial financial difficulties and 44% reported substantial emotional difficulties. These findings were consistent when adjusted for caregiver and care recipient characteristics. The unique financial strains experienced by sandwich generation caregivers may be related to multiple factors: raising a minor child to 18 years, an estimated cost of $230,000 on average in the U.S.;34 out-of-pocket spending they contribute towards the older adults’ care; and employment-related costs, such as loss of income and career opportunities if caregivers cut back on paid work hours or leave the workforce.35 Caregivers with lower income are more likely to provide financial support to their parents,36 which can only further exacerbate their financial strain.

While caregiving related gains were similar for sandwich generation caregivers and their non-sandwich counterparts, the former endorsed more financial and emotional difficulties, as well as higher overload overall. This may not be surprising, given that the amount of care provided is similar, yet the sandwich generation caregivers also are providing child care and a higher proportion of them work for pay. Our findings echo an analysis of caregivers in Canada that found sandwich generation caregivers were more likely to experience stress than caregivers providing only either child or older adult care.21 Our findings also echo an analysis of caregivers in the U.S. that found sandwich caregiving were associated with poorer general health.22 Given this pre-existing level of burden, the COVID-related exacerbation in mental health symptoms including suicidal ideation of sandwich generation caregivers is not surprising.8

Finally, this work suggests that sandwich generation caregivers do not use more supportive services relative to the non-sandwich generation caregivers, with both groups using relatively few such supports. The only type of service which the sandwich generation caregivers reported seeking more often was financial help, which may reflect the relatively greater disadvantage demonstrated by their higher proportion of enrollment in Medicaid. Given the limited use of supportive services but the relatively high proportion of employed sandwich generation caregivers, they may benefit from specific employment-related supports such as paid time off for caregiving responsibilities.37

This study has several limitations. First, this study lacked detailed information about the minor child and the corresponding child care, such as whether the minor child lived in the caregiver’s household was a teenager or toddler. Second, NSOC does not capture whether the caregiver has grandchildren for whom they are providing care, which is likely for those in their 50s.38 Finally, we used the most recent years of these surveys that can be used to generate nationally representative cross-sectional estimates, which was in 2015; our analysis does not account for the impact of COVID-19 on these caregivers.

In conclusion, our analysis of paired nationally representative surveys provides a profile of sandwich generation caregivers, who are providing “upwards” care equivalent to their non-sandwich counterparts. However, this is in addition to care provided to minor child(ren) in their household and higher rates of labor force participation. While few elements of the Biden Administration Build Back Better plan have made it into law, the critical societal role of caregivers continues to receive attention.39 Research and policy to support caregivers may have greater impact if tailored to the needs of specific populations of caregivers such as sandwich generation caregivers, given their unique combinations of roles across the workforce, raising children, and helping their older parents.

Supplementary Material

Supplement

Figure S1: Study Cohorts

Table S1: Characteristics of Adult Child, Child-in-law, and Grandchild Caregivers With and Without Minor Children in a Nationally Representative Sample, 2015

Table S2: Characteristics of Caregiving-related Experience and Employment Participation among Adult Child, Child-in-law, and Grandchild Caregivers With and Without Minor Children in a Nationally Representative Sample, 2015

Table S3: Characteristics of Caregiving-related Experience and Employment Participation among Adult Child Caregivers With and Without Minor Children: Adjusted Analysis

Impact Statement:

We certify that this work is novel. It is the first nationally representative analysis to examine the characteristics and caregiving-related experience of both sandwich generation caregivers (i.e., adult caregivers caring for both their parents as well as minor children) and the older adults to whom they provide care.

Key Points.

  • In this national survey study, sandwich generation caregivers provided similar hours of care to care recipients as non-sandwich generation caregivers, though more of them were in the paid workforce and, by definition, were caring for at least one minor child.

  • Sandwich generation caregivers reported higher caregiving-related financial and emotional difficulties and higher caregiver role overload.

What does this matter:

This study provides the first national estimates of characteristics and caregiving-related experience of sandwich generation caregivers and the older adults for whom they are caring.

Acknowledgments

National Health and Aging Trends Study (NHATS) is sponsored by National Institute on Aging Grant NIA U01AG032947 through a cooperative agreement with the Johns Hopkins Bloomberg School of Public Health.

Sponsor’s Role: No sponsor had any 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.

Declaration of Sources of Funding:

The work of Lianlian Lei were supported by 1K99AG075145 and 5R01AG056407 from the National Institute of Aging (NIA). The work of Donovan T. Maust were supported by 5R01AG056407 from NIA. Amanda N. Leggett received support through research grants (K01AG056557; 3K01AG056557–04S1; P30AG053760) from NIA.

Footnotes

Declaration of interests: All the authors declare no conflict of interest

REFERENCES

  • 1.Pierret CR. The sandwich generation: Women caring for parents and children. Monthly Lab Rev. 2006;129(9):3–9. [Google Scholar]
  • 2.Friedman EM, Park SS, Wiemers EE. New estimates of the sandwich generation in the 2013 panel study of income dynamics. Gerontologist. 2017;57(2):191–196. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 3.Spillman BC, Pezzin LE. Potential and active family caregivers: Changing networks and the ‘sandwich generation’. Milbank Q. 2000;78(3):347–374. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 4.Wiemers E, Bianchi S. Sandwiched between aging parents and boomerang kids in two cohorts of American women. Boston (MA): University of Massachusetts Boston Department of Economics; 2014. Accessed June 8, 2022. Working paper 2014–16. http://repec.umb.edu/RePEc/files/2014_06.pdf [Google Scholar]
  • 5.Suh J Measuring the “sandwich”: Care for children and adults in the American Time Use Survey 2003–2012. J Fam Econ Issues. 2016;37(2):197–211. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 6.Wolff JL, Mulcahy J, Huang J, Roth DL, Covinsky K, Kasper JD. Family caregivers of older adults, 1999–2015: Trends in characteristics, circumstances, and role-related appraisal. Gerontologist. 2018;58(6):1021–1032. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 7.The White House. FACT SHEET: How the Build Back Better framework will support the sandwich generation. Published September 21, 2021. Accessed May 30, 2022. https://www.whitehouse.gov/briefing-room/statements-releases/2021/09/21/fact-sheet-how-the-build-back-better-framework-will-support-the-sandwich-generation/.
  • 8.Czeisler MÉ, Rohan EA, Melillo S, et al. Mental health among parents of children aged< 18 years and unpaid caregivers of adults during the COVID-19 pandemic—United States, December 2020 and February–⁠ March 2021. Morb Mortal Wkly Rep. 2021;70(24):879–887. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 9.Boyczuk AM, Fletcher PC. The ebbs and flows: Stresses of sandwich generation caregivers. J Adult Dev. 2016;23(1):51–61. [Google Scholar]
  • 10.Steiner A The lived experiences of sandwich generation women and their health behaviours [Master Thesis]. Waterloo (ON), Canada: Department of Kinesiology and Physical Education, Wilfrid Laurier University; 2015. [Google Scholar]
  • 11.Steiner AM, Fletcher PC. Sandwich generation caregiving: A complex and dynamic role. J Adult Dev. 2017;24(2):133–143. [Google Scholar]
  • 12.Marrs L The sandwich generation: The challenges of caring for an elderly parent and raising children [Master Thesis]. Long Beach (CA): School of Social Work, California State University; 2011. [Google Scholar]
  • 13.Hammer LB, Neal MB. Working sandwiched-generation caregivers: Prevalence, characteristics, and outcomes. Psychol Manag J. 2008;11(1):93–112. [Google Scholar]
  • 14.Chassin L, Macy JT, Seo D-C, Presson CC, Sherman SJ. The association between membership in the sandwich generation and health behaviors: A longitudinal study. J Appl Dev Psychol. 2010;31(1):38–46. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 15.Ganapathy V, Graham GD, DiBonaventura MD, Gillard PJ, Goren A, Zorowitz RD. Caregiver burden, productivity loss, and indirect costs associated with caring for patients with poststroke spasticity. Clin Interv Aging. 2015;10:1793. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 16.Rubin RM, White-Means SI. Informal caregiving: Dilemmas of sandwiched caregivers. J Fam Econ Issues. 2009;30(3):252–267. [Google Scholar]
  • 17.Brenna E Should I care for my mum or for my kid? Sandwich generation and depression burden in Italy. Health policy (Amsterdam, Netherlands). 2021;125(3):415–423. [DOI] [PubMed] [Google Scholar]
  • 18.Kenny P, King MT, Hall J. The physical functioning and mental health of informal carers: evidence of care-giving impacts from an Australian population-based cohort. Health Soc Care Community. 2014;22(6):646–659. [DOI] [PubMed] [Google Scholar]
  • 19.McGarrigle CA, Cronin H, Kenny RA. The impact of being the intermediate caring generation and intergenerational transfers on self-reported health of women in Ireland. Int J Public Health. 2014;59(2):301–308. [DOI] [PubMed] [Google Scholar]
  • 20.Daatland SO, Veenstra M, Lima IA. Norwegian sandwiches. Eur J Ageing. 2010;7(4):271–281. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 21.Halinski M, Duxbury L, Higgins C. Working while caring for mom, dad, and junior too: Exploring the impact of employees’ caregiving situation on demands, control, and perceived stress. J Fam Issues. 2018;39(12):3248–3275. [Google Scholar]
  • 22.Do EK, Cohen SA, Brown MJ. Socioeconomic and demographic factors modify the association between informal caregiving and health in the Sandwich Generation. BMC public health. 2014;14(1):1–8. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 23.National Alliance for Caregiving. Burning the candle at both ends: sandwich generation caregiving in the U.S. November 2019. Accessed November 19, 2021. https://caringacross.org/wp-content/uploads/2019/11/NAC_SandwichCaregiving_Report_digital112019.pdf [Google Scholar]
  • 24.Kasper JD, Freedman VA. National Health and Aging Trends Study User Guide: Rounds 1–10 Final Release. Baltimore: Johns Hopkins University School of Public Health; 2021. Accessed September 21, 2021. www.nhats.org [Google Scholar]
  • 25.Freedman VA, Skehan ME, Wolff J, Kasper JD. National Study of Caregiving I-III User Guide. Baltimore: Johns Hopkins Bloomberg School of Public Health; 2020. Accessed March 15, 2022. www.nhats.org [Google Scholar]
  • 26.Freedman VA, Spillman BC, Kasper J. Hours of Care in Rounds 1 and 2 of the National Health and Aging Trends Study. NHATS Technical Paper #7. Baltimore: Johns Hopkins University School of Public Health; 2014. www.nhats.org [Google Scholar]
  • 27.Freedman VA, Spillman BC. Disability and care needs among older Americans. Milbank Q. 2014;92(3):509–541. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 28.Wolff JL, Spillman BC, Freedman VA, Kasper JD. A national profile of family and unpaid caregivers who assist older adults with health care activities. JAMA Intern Med. 2016;176(3):372–379. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 29.Spillman BC, Wolff J, Freedman VA, Kasper JD. Informal caregiving for older Americans: An analysis of the 2011 National Study of Caregiving. Washington, DC: US Department of Health and Human Services; 2014. Accessed June 8, 2022. https://aspe.hhs.gov/sites/default/files/migrated_legacy_files//44496/NHATS-IC.pdf [Google Scholar]
  • 30.Guets W Does the formal home care provided to old-adults persons affect utilization of support services by informal carers? An analysis of the French CARE and the US NHATS/NSOC surveys. GATE; January 2021. Accessed June 8, 2022. Working paper WP 2105. https://halshs.archives-ouvertes.fr/halshs-03115306/document [Google Scholar]
  • 31.Wolff J Calculating Work Productivity Loss in the National Study of Caregiving. NHATS Technical Paper #13. Baltimore: Johns Hopkins Bloomberg School of Public Health; 2016. Accessed June 8, 2022. www.nhats.org [Google Scholar]
  • 32.Dukhovnov D, Zagheni E. Who takes care of whom in the United States? Time transfers by age and sex. Population and development review. 2015;41(2):183–206. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 33.Luppa M, Luck T, Weyerer S, König H-H, Brähler E, Riedel-Heller SG. Prediction of institutionalization in the elderly. A systematic review. Age and ageing. 2010;39(1):31–38. [DOI] [PubMed] [Google Scholar]
  • 34.Merrill. The financial journey of modern parenting: Joy, complexity and sacrifice. 2018. Accessed June 8, 2022. https://mlaem.fs.ml.com/content/dam/ml/registration/ml_parentstudybrochure.pdf
  • 35.National Academies of Sciences, Engineering, Medicine. Families caring for an aging America. Washington, DC: National Academies Press; 2016. [PubMed] [Google Scholar]
  • 36.Pew Research Center. The sandwich generation rising: Financial burdens for middle-aged Americans. January 2013. Accessed November 19, 2021. https://www.pewresearch.org/social-trends/2013/01/30/the-sandwich-generation/
  • 37.Feinberg L Keeping up with the times: supporting family caregivers with workplace leave policies. 2013. https://www.aarp.org/content/dam/aarp/research/public_policy_institute/ltc/2013/fmla-insight-keeping-up-with-time-AARP-ppi-ltc.pdf
  • 38.Boaz RF, Hu J, Ye Y. The transfer of resources from middle-aged children to functionally limited elderly parents: Providing time, giving money, sharing space. Gerontologist. 1999;39(6):648–657. [DOI] [PubMed] [Google Scholar]
  • 39.Span P The Quiet Cost of Family Caregiving. Published September 4, 2022. Accessed October 2, 2022. https://www.nytimes.com/2022/09/04/science/elderly-work-caregiving.html.

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Supplementary Materials

Supplement

Figure S1: Study Cohorts

Table S1: Characteristics of Adult Child, Child-in-law, and Grandchild Caregivers With and Without Minor Children in a Nationally Representative Sample, 2015

Table S2: Characteristics of Caregiving-related Experience and Employment Participation among Adult Child, Child-in-law, and Grandchild Caregivers With and Without Minor Children in a Nationally Representative Sample, 2015

Table S3: Characteristics of Caregiving-related Experience and Employment Participation among Adult Child Caregivers With and Without Minor Children: Adjusted Analysis

RESOURCES