Strapped for Time or Stressed Out? Predictors of Work Interruption and Unmet Need for Workplace Support Among Informal Elder Caregivers

Abstract

Objective: We evaluate whether strong associations between unmet need and work interruption observed among informal elder caregivers are explained by caregiver personal characteristics, caregiving situations, or diminished caregiver well-being. Method: We analyze a proprietary survey of informal elder caregivers conducted by a single large U.S. employer (N = 642 caregivers). Results: Unmet need was about twice as common among those experiencing mild (29.79%) or severe interruption (35.00%) relative to those who did not interrupt work (16.87%). Although caregiver characteristics and aspects of caregiving situation show links to either unmet need or work interruption specifically, the association between unmet need and work interruption was not explained by these factors, or by caregiver well-being. Discussion: Needs-related work interruptions may arise by other workplace processes, such as organizational culture, that should be measured and tested explicitly by future research.

Keywords

informal caregiving employment work interruption caregiver unmet need caregiver well-being

About one in every four employed adults in the United States provides informal care for a parent, in-law, or another family member above the age of 65 (Moen & Roehling, 2005; Pavalko, 2011; Robison, Fortinsky, Kleppinger, Shugrue, & Porter, 2009; U.S. Census, 2010). Informal elder caregiving has been linked to increased personal stress, decreased health, work–family conflict, and work interruption (Pavalko, 2011; Pavalko & Henderson, 2006; Pinquart & Sörensen, 2011). For instance, Robison et al. (2009) found that work interruptions occurred among 25% of employed elder caregivers.

In addition to lost productivity, work interruptions may lead to missed training opportunities or diminished job-relevant skills, both of which are principal concerns for human resource professionals seeking to retain and invest in their employees. As for employees themselves, work interruptions can produce unanticipated financial strain or burden, as well as strained coworker relationships and a diminished sense of self or competence due to an inability to exercise one’s identity as a worker (Burke & Stets, 2009; Pearlin, 2010). Overall, then, gaining a clearer understanding of work interruption among informal elder caregivers would help build a more productive workforce.

One of the most well-known, yet least understood, predictors of work interruption among informal elder caregivers is unmet need for support (Baird & Reynolds, 2004; Pavalko & Henderson, 2006; Pearlin, Aneshensel, & Leblanc, 1997). Unmet need refers to when a subjective personal need to use a workplace support for informal elder caregiving is not met, either due to not having access to such support or not using it despite access (e.g., Berg, Kossek, Misra, & Belman, 2014; Budd & Mumford, 2006; Knight, Lutzsky, & Macofsky-Urban, 1993). Although it is well established that unmet need is common among employed caregivers, our knowledge is limited regarding which caregivers experience unmet need and potentially also interrupt their paid work.

In this study, we utilize survey data collected from a single large U.S. employer to study informal elder caregivers involved in a variety of caregiving situations. The employer we study, a large public university, offers universal entitlements and benefits for informal elder caregivers, thus allowing us to advance previous examinations of unmet need in two ways. First, we avoid confounding access and use of caregiver supports, a common shortcoming of multiemployer and population studies. Second, our large sample covers employees with diverse personal backgrounds, providing a unique opportunity to evaluate specific and shared predictors of unmet need and work interruption.

Background

As they face competing demands to remain productive at work and provide complex support to their significant others, informal elder caregivers may opt for reduced labor force participation (Lima, Allen, Goldscheider, & Intrator, 2008; Silverstein & Giarrusso, 2010). Such instances of reduced work participation, termed work interruptions, range from relatively mild changes such as changing or reducing hours each week (Chesley & Moen, 2006; Dentinger & Clarkberg, 2002; J. Kim, Ingersoll-Dayton, & Kwak, 2013; Lee & Tang, 2015; Lilly, Laporte, & Coyte, 2007; Pavalko, 2011; Pavalko & Henderson, 2006) to more severe interruptions such as taking a leave of absence, voluntary unemployment, or early retirement (Kubicek, Korunka, Hoonakker, & Raymo, 2010; Szinovacz & Davey, 2005; Wang, 2012). By facilitating access to workplace support for informal caregivers, employers help to assuage the difficulties of an unanticipated “second career” of caring for family members who face significant physical or mental limitations (Applebaum & Breitbart, 2013; Casado, van Vulpen, & Davis, 2011; Given, Sherwood, & Given, 2008).

This previous research indicates that work interruptions are more likely when caregivers do not have access to, or do not use, workplace programs designed to address their needs. Such programs include extended leave under the Family Medical Leave Act (FMLA), flexible work arrangements in terms of the allocation of hours or changing the site or nature of work, or eldercare information and counseling. Although having access to these supports at work mitigates the odds of a work interruption (Pavalko, 2011; Pearlin, 2010; Pearlin et al., 1997; Silva, Teixeira, Teixeira, & Freitas, 2013), not using workplace supports even when they are available also has been tied to work interruptions among caregivers (e.g., Applebaum & Breitbart, 2013; Gaugler et al., 2004; Jeon, Brodaty, & Chesterson, 2005; Mitra, Bogen, Long-Bellil, & Heaphy, 2011; Robison et al., 2009; Silva et al., 2013). Elder caregivers who forgo an opportunity to address an unmet need for employer support may lack the time or personal resources to participate in available workplace programs.

Caregiver unmet need may be predictable according to caregiver characteristics or basic aspects of caregiving situations (Lilly et al., 2007; Pavalko, 2011; Pinquart & Sörensen, 2011). For instance, higher levels of education may enhance one’s problem-solving skills, sense of control, or informal social networks (Cutler & Lleras-Muney, 2010). Meanwhile, caregiver age is related to personal health problems or other personal limitations, potentially influencing one’s ability to recruit workplace support or provide informal care in the first place. Gender is associated with tendencies to seek and give social support, with women typically showing more robust support formation tendencies in stressful life situations (Taylor & Stanton, 2007; Thoits, 2011). Racial and ethnic differences in caregiving norms, resources, and probability of caregiver burnout have been reported (Pavalko, 2011; Pinquart & Sörensen, 2005). Finally, being married predisposes one to spousal caregiving, which is a particularly difficult form of care provision that may make workplace coordination more difficult (Pinquart & Sörensen, 2011). Altogether, caregiver characteristics serve as prime determinants of the nature of the caregiving role and, thus, may also carry consequences for unmet need, work interruption, or both among informal elder caregivers.

Unmet needs among caregivers also are predicted by challenges inherent to the caregiving situation itself (Black et al., 2013; Casado et al., 2011). For instance, number of care recipients, amount of time spent caregiving, and specific health conditions affecting care recipients all matter to predicting caregiver unmet need (Bainbridge, Cregan, & Kulik, 2006; J. Kim et al., 2013; Pinquart & Sörensen, 2011). Finally, a caregiver’s own diminished well-being may contribute to his or her unmet needs. Numerous studies have documented links between informal care provision and diminished caregiver psychological or physical well-being (for reviews, see Pinquart & Sörensen, 2007; Schulz & Sherwood, 2008). Caregiving may diminish the sense of control over one’s life, and thereby increase psychological distress (Skaff, Pearlin, & Mullan, 1996). Diminished well-being may inhibit a caregiver’s ability to achieve needed workplace support.

Objective of the Present Study and Research Questions

Although caregiver characteristics, caregiving situations, and caregiver well-being all may be related to unmet need for workplace support, work interruption, or both, what remains unclear is which factors may be responsible for explaining the association between unmet need and work interruption. Chesley and Moen (2006) suggested that when employed caregivers experience time conflicts involving their caregiving situations, they may forgo a need to seek out and participate in a workplace program, resolving the conflict by reducing their hours at work (see also Pavalko, 2011). However, potential pathways linking caregiver factors to unmet need and work interruption remain untested.

In this study, we test whether caregiver and caregiving factors associate with unmet need or work interruption. Specifically, we test whether these factors are associated with unmet need and work interruption individually, and then to what extent they explain the strong association between these phenomena. We expect that, if needs-based work interruptions are triggered by certain caregiver backgrounds or caregiving arrangements, then measured factors should either partially or fully explain observed associations between unmet need and work interruption. We focused on four questions:

Research Question 1: To what extent are unmet needs for workplace support associated with work interruption among informal elder caregivers?

Research Question 2: How do caregivers reporting unmet needs differ from those who do not?

Research Question 3: How do caregivers who interrupt work differ from those who do not?

Research Question 4: To what extent is the association between unmet needs and work interruption explained by caregiver factors (characteristics, situation, and well-being)?

Method

In 2012, we worked with a large Midwestern public university to field an online, point-in-time survey, for the purpose of examining issues pertaining to aging employees. All paid university employees (N = 14,799) received an email invitation requesting participation in the survey. Those who did not respond received two more follow-up invitations.

The survey included a root question asking employees whether they were currently or had recently been a caregiver for any individual above the age of 65. Informal caregiving was defined as currently or recently providing regular or intermittent assistance to someone without being paid to do so (see Pavalko, 2011; Pearlin et al., 1997), and included assisting with medications or medical equipment, personal hygiene, preparing meals, providing rides to the store or doctor’s office, and other activities that entail assistance with finances, living arrangements, and other life-sustaining issues (Pearlin et al., 1997; Savla, Almeida, Davey, & Zarit, 2008).¹ The survey collected information about caregiving situations, and inquired about awareness, need, and use of campus and community supports designed for those in an elder caregiving role. The study was approved by our university institutional review board.

We obtained 3,246 responding employees. Our response rate (22%) resembles those obtained using similar designs (i.e., web-based surveys or surveys conducted in educational organizations; Baruch & Holtom, 2008; Cook, Heath, & Thompson, 2000; Kaplowitz, Hadlock, & Levine, 2004). As in other survey studies of academic institutions, most employees work full time and women and professional or scientific staff were slightly overrepresented in our obtained sample; 24.8% of this sample self-identified as elder caregivers (N = 805), a rate similar to those reported in comparable firm-specific research (see Kossek, Pichler, Bodner, & Hammer, 2011; Wagner & Hunt, 1994).

Measures

Caregiver work interruption

We measured work interruption related to caregiving by asking respondents to indicate whether, during their time spent as a caregiver, they have ever interrupted their job or their career. Examples of interruption ranged from none (e.g., never having to change work schedule due to caregiving) to mild (e.g., adjusting starting and ending hours of work, working partial rather than full days) to severe (moving from full- to part-time, taking a demotion or turning down a promotion, entering into phased or early retirement, taking a leave of absence).

Unmet caregiving need

We measured unmet need by asking respondents about their participation in a variety of university and community programs geared toward helping caregivers. To reduce survey response burden, respondents were queried about six randomly selected programs from a total set of 14 possible programs. Six of these possible programs—eldercare resource and referral, employee assistance program, financial counseling, flexible work arrangements, unpaid leave, or FMLA leave—were university based. The other eight were based in the local community and included adult day care services, caregiver support groups, information and education meetings, in-home skilled or unskilled services, respite care services, and transportation services.² If respondents indicated that they need, but do not use, one or more of the six programs about which they were randomly asked, they were designated as having unmet need (1 = 1 or more unmet needs, 0 = no unmet needs).³

Individual caregiver characteristics

To assess basic personal differences or resources relevant to caregiving, we considered caregiver age, gender, race (White/non-White), years of education, and job classification within the university (e.g., Fredriksen & Scharlach, 1997; Lee & Tang, 2015). We used job classification as it is relevant to broad differences in work flexibility and to work–family conflict experienced by caregivers (e.g., faculty have more inherent work flexibility on average than staff).

Caregiving situation

We performed a basic assessment of caregiving situation (e.g., Chesley & Moen, 2006; Y. Kim & Schulz, 2008; Lilly et al., 2007). First, we took into account extensiveness of caregiving, in terms of reported number of hours per week spent giving care (0 = 0-10 hr per week, 1 = more than 10 hr per week) and number of care recipients (one, two, three, or more). We also considered the geographic distance the recipient lives from the caregiver (in household, 15-min drive or less, 15-30 min, 30-min to 1-hr drive, farther than 1-hr drive). To account for the situational aspects of caregiving, we examined any specific health condition(s) affecting the care recipient(s) (i.e., independence issues or physical limitations, chronic conditions, and/or mental illness or cognitive impairment). For respondents providing care to more than one recipient, maximum values of responses to the time commitment and geographic distance questions were used and health conditions were pooled across recipients.

Caregiver psychological and physical well-being

We used a 10-item psychological well-being index capturing self-esteem, mastery, and optimism. Self-esteem items included “I feel useless at times” (reverse scored), “I am able to do things as well as most other people,” and “On the whole, I am satisfied with myself”; mastery items included “I can solve many of my own problems,” “I sometimes feel overwhelmed by my problems” (reverse), “I can change most of the important things in my life,” and “I have control over the things that happen to me” (similar to Pearlin, Menaghan, Lieberman, & Mullan, 1981, p. 353). Optimism items included “I am more optimistic than pessimistic,” “I avoid disappointment by expecting the worst” (reverse), and “I feel haunted by bad luck” (reverse; similar to Scheier, Carver, & Bridges, 1994, p. 1073). All items had a Likert-type response format (1 = strongly disagree, 2 = disagree, 3 = somewhat disagree, 4 = somewhat agree, 5 = agree, 6 = strongly agree), and respondents randomly received five of these 10 items.

Due to their similar effects on outcomes, psychological resources are sometimes treated as an index or single factor within stress-process research (Pearlin, 2010; Pearlin et al., 1981). For the present study, a confirmatory factor analysis conducted in Stata 12 using a maximum likelihood with missing values estimator demonstrated that these 10 items reflect a single latent well-being factor (root mean square error of approximation [RMSEA] = .039, comparative fit index [CFI] = .955, Tucker–Lewis index [TLI] = .935, χ²(31) = 185.3, p < .001). Factor scores from this measurement model were calculated and correlated almost perfectly with short administration scores based on five items (r = .96). Therefore, we present and use factor scores rather than observed scores.

We also measured caregiver well-being using respondents’ reports of self-rated health. Respondents were asked whether they would rate their health as very bad, bad, poor, fair, good, or very good (6-point scale). Self-rated health has been shown to predict mortality and to provide valuable health information above and beyond objective physician assessments (Idler & Benyamini, 1997; Jylhä, Volpato, & Guralnik, 2006).

Analytic Strategy

Pursuant to our first research question, we generated descriptive statistics on caregiver variables and the distribution of unmet need across levels of work interruption. To respond to our second research question, concerning the predictors of unmet need for workplace support, we contrasted the individual characteristics and situations of informal elder caregivers who reported at least one unmet need from those who reported no unmet needs. To answer our third research question, we contrasted caregiver variables across levels of work interruption (none, mild, severe). Finally, to answer our fourth research question, regarding the nature of the association between unmet need and work interruption, multinomial logistic regression models were used to estimate the odds of experiencing a mild work interruption compared with no work interruption and the odds of experiencing a severe work interruption compared with no work interruption. In our first model, we estimated the odds of mild or severe work interruption versus no interruption linked to unmet need for workplace support. Here, we included only caregiver characteristics as predictor variables. Model 2 adds aspects of caregiver situation to the model. Finally, in Model 3, caregiver psychological well-being and physical well-being are entered. Although this set of three nested models is not meant to serve as a conclusive test of pathways leading from unmet need to work interruption, it sequentially identifies any robust predictors of both unmet need and work interruption. Across all these analyses, we used only those caregivers who provided valid responses on all our key independent variables (N = 642). Results using all valid cases on a variable-by-variable basis led to the same conclusions. The multinomial regression analyses produced the same substantive conclusions under multiple imputation (available on request).

Results

Employees held about 16 years of education on average (69.5% had a 4-year degree or higher) and predominantly were female, married, and White. Whereas providing care for one family member was most common (67.7%), a sizable number of caregivers oversaw two (26.5%) or three (5.8%) family members. Furthermore, about two fifths (41.4%) of caregivers reported being involved in their role for 10 or more hours per week. Most provision of care occurred within the household or within a short drive (60.8%; ≤30-min drive). Caring for family members with chronic health conditions or physical limitations was quite common in the sample (70.7% and 80.2%, respectively), which attests to the intensity of informal caregiving undertaken by employees within this firm.

The percentage of respondents who reported an unmet need was significantly higher for those caregivers experiencing either a mild or severe work interruption (29.79% and 35.00%) than for those reporting no interruption (16.87%, ps < .05).

Table 1 summarizes caregiver demographic background and situational characteristics by unmet need status. About one quarter (27.26%, n = 175) of employed caregivers reported one or more unmet needs (i.e., “needing but not using”) for either university- or community-based caregiver support programs. Among caregiver demographic characteristics, caregivers reporting at least one unmet need have fewer years of schooling on average (15.71 vs. 16.26 years, p < .05). Also, White racial status is more common among those reporting an unmet need (94.43% vs. 98.86%, p < .05). Caregivers reporting at least one unmet need care for a greater number of recipients on average (1.32 vs. 1.49 care recipients, p < .05). However, all other caregiver situation aspects do not show significant links to unmet need status.

Table 1.

Descriptive Statistics, by Level of Unmet Need (Caregivers).

Informal elder caregiver needs	No needs unmet (n = 467)		1+ unmet needs (n = 175)
Informal elder caregiver needs	M	SD	M	SD
Caregiver characteristics
Education (years)	16.26	2.40	15.71^a	2.41
Age (years)	54.70	8.82	54.42	8.26
White (%)	94.43		98.86^a
Female (%)	78.80		78.86
Married (%)	73.45		70.29
Tenured or tenure-track faculty (%)	8.14		8.57
Caregiver situation
Number of recipients	1.32	0.55	1.49^a	0.65
Spends 10+ hr caregiving (%)	42.40		42.86
Distance from recipient (1-5)	3.07	1.41	2.95	1.44
Recipient health issues (%)
Chronic condition	70.45		70.29
Mental illness/cognitive impairment	48.18		52.00
Independence/physical limitations	80.73		83.43
Caregiver well-being
Psychological well-being	0.02	0.40	−0.05^b	0.45
Self-rated health = “very good” (%)	34.69		28.00

Source. 2012 Iowa Employee Survey.

Note. Means, standard deviations, and sample percentages are shown.

Differs from no needs unmet at α = .05.

α = .10.

Table 2 summarizes caregiver demographic background and situational characteristics by work interruption status. About one quarter (25.9%, n = 166) of caregivers reported no work interruption and the remainder reported either mild (57.2%, n = 376) or severe (15.6%, n = 100) interruption. Higher education is negatively associated with reports of work interruption, given fewer years of schooling, on average, among those reporting mild or severe work interruption (relative to no interruption, ps < .05). Similarly, tenured or tenure-track faculty are less represented among those experiencing work interruption (mild interruption contrast, p < .05). Hours spent caregiving also is a distinguishing factor, given that more than half (54%) of caregivers experiencing severe interruption report serving 10 or more hours per week (vs. 37.4% for noninterrupted caregivers, p < .05). Otherwise, trends suggested that recipient health problems (chronic conditions or mental illness/cognitive impairment) may be linked to work interruption. Caring for an individual suffering from a chronic condition showed a link to mild work interruption in particular (74.73%, vs. 60.24% for noninterrupted caregivers; p < .05). Finally, individuals who reported severe work interruptions were less likely to report “very good” self-rated health than those who reported no work interruption (25%, vs. 37% for noninterrupted caregivers; p < .05).

Table 2.

Descriptive Statistics, by Level of Work Interruption.

Level of work interruption	None (n = 166)		Mild (n = 376)		Severe (n = 100)
Level of work interruption	M	SD	M	SD	M	SD
Caregiver characteristics
Education (years)	16.71	2.43	15.89^a	2.37	15.93^a	2.40
Age (years)	54.88	10.07	54.40	8.00	55.05	8.64
White (%)	95.78		96.28		93.00
Female (%)	75.90		80.05		79.00
Married (%)	73.49		73.40		68.00
Tenured or tenure-track faculty (%)	12.05		6.38^a		9.00
Caregiver situation
Number of recipients	1.31	0.56	1.38	0.59	1.41	0.59
Spends 10+ hr caregiving (%)	37.35		41.76		54.00^a
Distance from recipient (1-5)	3.05	1.43	3.01	1.38	3.10	1.54
Recipient health issues (%)
Chronic condition	60.24		74.73^a		71.00^b
Mental illness/cognitive impairment	43.37		49.47		58.00^b
Independence/physical limitations	83.73		80.05		83.00
Caregiver well-being
Psychological well-being	0.06	0.40	−0.01	0.42	−0.08^b	0.43
Self-rated health = “very good” (%)	36.75		33.24	0.47	25.00^a	0.44

Source. 2012 Iowa Employee Survey.

Note. Means, standard deviations, and sample percentages are shown.

Differs from none at α = .05.

α = .10.

Table 3 displays results from multinomial regression models predicting mild or severe work interruption during one’s time as an elder caregiver. In total, nearly three quarters (73.4%) of the caregiver sample experienced either mild or severe work interruptions. In these models, no work interruption serves as the reference or baseline category. As shown in Model 1, unmet need has a significant and positive impact on the likelihood of experiencing work interruption. Specifically, net of caregiver demographic characteristics, caregivers who report one or more unmet needs show substantially elevated odds of reporting a mild (odds ratio [OR] = 2.06, confidence interval [CI] = [1.28, 3.31]) or severe work interruption (OR = 2.57, CI = [1.42, 4.67], ps < .01). Meanwhile, education negatively predicts work interruption in this model, with each additional year being linked to an 11.2% reduction (1 – 0.888 = 11.2) in odds of mild interruption and 14.2% reduction in odds of severe interruption.

Table 3.

Multinomial Logistic Regression Predicting Mild or Severe Work Interruption Among Informal Elder Caregivers (N = 642).

Reference: no work interruption	Model 1		Model 2		Model 3
Reference: no work interruption	Mild interruption	Severe interruption	Mild interruption	Severe interruption	Mild interruption	Severe interruption
Unmet need	2.059**	2.574**	2.165**	2.739**	2.162**	2.671**
Unmet need	(0.499)	(0.783)	(0.539)	(0.855)	(0.541)	(0.841)
Caregiver characteristics
Education (years)	0.888*	0.858*	0.883*	0.857*	0.879*	0.863
Education (years)	(0.047)	(0.062)	(0.048)	(0.064)	(0.049)	(0.076)
Age (years)	0.997	1.005	0.999	1.006	1.001	1.013
Age (years)	(0.011)	(0.016)	(0.017)	(0.016)	(0.012)	(0.017)
White	0.828	0.390	0.759	0.319	0.756	0.337
White	(0.410)	(0.228)	(0.381)	(0.191)	(0.385)	(0.203)
Female	1.131	1.083	1.087	1.028	1.123	1.171
Female	(0.270)	(0.355)	(0.263)	(0.343)	(0.279)	(0.397)
Caregiver situation
Number of recipients			0.917	0.918	0.920	0.926
Number of recipients			(0.168)	(0.222)	(0.169)	(0.226)
Spends 10+ hr caregiving			1.113	1.956*	1.126	1.992*
Spends 10+ hr caregiving			(0.229)	(0.541)	(0.232)	(0.556)
Distance from recipient			1.008	1.128	1.005	1.126
Distance from recipient			(0.073)	(0.109)	(0.073)	(0.110)
Chronic condition			2.152***	1.672	2.129***	1.546
Chronic condition			(0.456)	(0.485)	(0.457)	(0.456)
Mental illness/cognitive impairment			1.450	2.022*	1.452	2.015*
Mental illness/cognitive impairment			(0.293)	(0.557)	(0.294)	(0.559)
Independence/physical limitations			0.851	0.925	0.840	0.889
Independence/physical limitations			(0.222)	(0.334)	(0.220)	(0.324)
Caregiver well-being
Psychological well-being					0.639	0.446*
Psychological well-being					(0.168)	(0.153)
Self-rated health = “very good”					1.180	0.841
Self-rated health = “very good”					(0.260)	(0.265)
Pseudo R²	.037		.06		.066

Note. For each model, reference or baseline category is no work interruption. Odds ratios (relative to reference category) are shown and robust standard errors are given in parentheses. Models also control for marital status (not significant, not shown) and include fixed effects for university job position (not significant, not shown; faculty, professional/scientific staff, union-protected staff, research/clinical or adjunct faculty, other employee).

p < .05. **p < .01. ***p < .001 (two-tailed significance tests).

Model 2 adds in caregiver situation. Relative to Model 1, coefficients remain significant and quite similar.⁴ In particular, unmet need remains linked to increased odds of mild (OR = 2.17, CI = [1.32, 3.53]) or severe interruption (OR = 2.74, CI = [1.49, 5.05]) relative to no interruption (ps < .01). Meanwhile, certain aspects of caregiver situation are linked to work interruption, namely, caring for individuals with chronic conditions (mild interruption: OR = 2.15, CI = [1.42, 3.26], p < .001) or mental illness or cognitive impairment (severe interruption: OR = 2.02, CI = [1.18, 3.47], p < .05). Also, odds of severe work interruption are almost twice as high among those providing more than 10 hr of care per week (OR = 1.96, CI = [1.14, 3.37], p < .05).

In Model 3, caregiver well-being (psychological well-being and self-rated health) is added to the work interruption model. Of key interest, unmet need remains linked to heightened odds of mild (OR = 2.16, CI = [1.32, 3.53]) or severe (OR = 2.67, CI = [1.44, 3.95]) work interruption (ps < .01). Meanwhile, psychological well-being is linked to lower odds of incurring a work interruption (OR = 0.45, CI = [0.23, 0.87], p < .05), which is consistent with the resilience of caregivers who possess psychological resources (Pearlin et al., 1997).

In sum, the persistent association of unmet need with work interruption across all models (Models 1-3) suggests that the link between unmet need and either mild or severe work interruption is not produced by the caregiver background characteristics or caregiving situations included in our models (Models 1 and 2), or is this relationship explained by differences in our measures of psychological or physical well-being (Model 3).

Discussion

Caregiving resources, whether provided by employers or in the community, often go underutilized. Previous studies have identified numerous workplace, cultural, or personal factors that may stand in the way of caregivers gaining the support they need for their role, such as lack of supervisor encouragement, ethnic differences in cultural understandings of the provision of care, or personal factors such as marital status or number of dependents (e.g., Brown & Chen, 2008; Robinson, Buckwalter, & Reed, 2013; Scharlach et al., 2006; Toseland, McCallion, Gerber, & Banks, 2002). However, these models leave unclear how unmet needs are associated with work interruption among employed caregivers.

In this study, we contributed to this important line of work by drawing on a diverse employee sample at a large U.S. firm. We uncovered a considerably high amount of mild or severe work interruption among informal caregivers, with about three fourths of caregivers incurring at least a mild interruption and about one quarter of caregivers expressing an unmet need for workplace support. These targeted statistics lend a more nuanced perspective on labor force participation than a typical approach, which simply examines whether the caregiver is or is not currently employed (Lee & Tang, 2015; Lilly et al., 2007) while extending approaches that focus solely on severe interruption (e.g., retirement or early retirement; Dentinger & Clarkberg, 2002; Kubicek et al., 2010) or milder forms of interruption such as leave or unanticipated time off (Robison et al., 2009).

We found that unmet need was about twice as common among those who interrupted work relative to those who did not. Unmet need became less common with increasing education, and was more common among White respondents and those with a greater number of care recipients. Meanwhile, mild or severe interruption was predicted by less education, faculty job status, self-rated health, time spent caregiving, and recipient health issues. These findings indicate which groups of employed caregivers may be vulnerable to mild or severe work interruptions: those who provide numerous hours of help, who have diminished well-being, or who support a care recipient with a mental illness or cognitive impairment. However, the relationship between unmet need and work interruption remained strong even after adjusting for all caregiver factors considered: characteristics, situation, and psychological or physical well-being.

To contextualize these findings, it is important to recognize some significant limitations. First, while relying on a single employer allows us to focus on unmet need in a context of universal support access, this also unfortunately limits the generalizability of our results. Next, although our study focuses on work interruptions incurred while one serves as a caregiver, our survey still cannot discern the relative timings of unmet needs and work interruptions, because we do not know whether interruptions are in the present or in the recent past of one’s caregiving career. Another related issue with time ordering is that certain caregiver characteristics may shift in response to interrupting work, such as when interrupting work enables one to increase one’s hours devoted to caregiving. We were unable, in this survey, to ascertain specific reasons surrounding work interruptions. A design inquiring about caregiver work history, the logistical nature or difficulty of informal care provided, and stated reasons for interruption certainly would be valuable to contextualizing the present findings with greater nuance.

As in other studies, particularly burdened caregivers may be less likely to complete surveys, which may have constrained the amount of variation in caregiver situation, unmet need, and work interruption that we observed in the recruited sample. However, our conclusions regarding the prevalence of unmet need and work interruption and their association are likely to be conservative, as past employees who ceased employment, perhaps due to intensive or unmet caregiving needs, were not part of our sampling frame. Unmet workplace support need often results in taking extended time off or quitting altogether, making full variation in unmet need hard to achieve in workplace surveys; meanwhile, population or multisite surveys rarely ask targeted questions about workplace supports.

Because our study draws upon a large firm with consistent access to caregiver workplace support across its units, it provides a strong test of whether caregiver variables matter to needs-based work interruptions. Because we find little evidence to suggest they do, other contextual factors, such as organizational culture or workplace relationships, may induce needs-based interruptions, but this possibility will need to be measured and tested explicitly by future research.⁵ For instance, coworker relationships may create a sense of obligation that would make interrupting work difficult, as in work teams where absent members would stifle progress on important projects. Another well-known factor influencing employee outcomes is supervisor attitudes, as these are linked to workplace climates that foster competitiveness or total work commitment on one hand or self-care and work–life balance on the other (Chesley & Moen, 2006; Pavalko, 2011). Finally, despite universal, firm-wide access to caregiving supports, workplaces may still lack clear guidance at the unit or departmental level on obtaining caregiving support, which may be responsible for producing unmet need as well as work interruption. Indeed, despite workplace initiatives around the United States, employers often do not adequately describe support options for caregivers to their employees, or clarity and communication regarding available supports is uneven across work units or departments within firms (Anderson, Coffey, & Byerly, 2002; Kelly, Moen, & Tranby, 2011; Lapierre & Allen, 2006; Pavalko & Henderson, 2006).

There is growing evidence that self-care and caring informally for others need not be mutually exclusive (Arora & Wolf, 2014). Employers may creatively help restructure work time or place employees in touch with counseling resources for optimizing and enjoying personal time despite challenging social obligations associated with informal caregiving (N. Kim & Gordon, 2014). Through counseling, it may be helpful to recast caregiving as a meaningful or purposeful social role that, despite its frequent demands or frustrations, allows individuals to convey their unwavering support and affection for significant others by being a source of reliable support during a time of struggle (Marks, Lambert, & Choi, 2002).

Conclusion

This study offers a preliminary test of the importance of basic caregiver factors to unmet need and work interruption. By taking the perspective of unmet caregiver needs and by focusing on a single large firm, we provided new and specific insights not only into caregiver background and situation aspects linked to unmet need but also into potential reasons why unmet need might precipitate work interruption. According to our findings, work interruption linked to unmet need is explained neither by basic caregiver or caregiving differences nor by caregiver well-being. Although more work still is needed, future research may profit from redoubling explanatory efforts focused on contextual factors beyond personal caregiver factors, to better understand pathways by which unmet need triggers work interruption.

Footnotes

Declaration of Conflicting Interests

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

Funding

The authors disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: This research was supported by an award from the TIAA-CREF Institute (“Successful Aging Within Academic Institutions”; 01/01/10-06/30/15). Program on Aging seminar participants at the Yale School of Medicine provided valuable feedback. The authors acknowledge additional support provided by the University of Iowa, the Iowa Center on Aging, and the Iowa Social Science Research Center.

Notes

References

Anderson

S. E.

Coffey

B. S.

Byerly

R. T.

(2002). Formal organizational initiatives and informal workplace practices: Links to work-family conflict and job-related outcomes. Journal of Management, 28, 787-810.

Applebaum

A. J.

Breitbart

(2013). Care for the cancer caregiver: A systematic review. Pallative & Supportive Care, 11, 231-252.

Arora

Wolf

D. A.

(2014). Is there a tradeoff between parent care and self-care? Demography, 51, 1251-1270.

Baird

C. L.

Reynolds

J. R.

(2004). Employee awareness of family leave benefits: The effects of family, work, and gender. Sociological Quarterly, 45, 325-353.

Baruch

Holtom

B. C.

(2008). Survey response rate levels and trends in organizational research. Human Relations, 61, 1139-1160.

Berg

Kossek

E.E.

Misra

Belman

(2014). Work-life flexibility policies: Do unions affect employee access and use? Industrial and Labor Relations Review, 67, 111-137.

Black

B. S.

Johnston

Rabins

P. V.

Morrison

Lyketsos

Samus

Q. M.

(2013). Unmet needs of community-residing persons with dementia and their informal caregivers: Findings from the Maximizing Independence at Home Study. Journal of the American Geriatrics Society, 61, 2087-2095.

Bainbridge

H.T.J.

Cregan

Kulik

C.T.

(2006). “The effect of multiple roles on caregiver stress outcomes.” Journal of Applied Psychology, 91, 490-497.

Brown

Chen

S.-L.

(2008). Help-seeking patterns of older spousal caregivers of older adults with dementia. Issues in Mental Health Nursing, 29, 839-852.

10.

Budd

J.W.

Mumford

K.A.

(2006). Family-friend work practices in Britain: Availability and perceived accessibility. Human Resources Management, 45, 23-42.

11.

Burke

P. J.

Stets

(2009). Identity theory. New York, NY: Oxford University Press.

12.

Casado

B. L.

van Vulpen

K. S.

Davis

S. L.

(2011). Unmet needs for home and community-based services among frail older Americans and their caregivers. Journal of Aging and Health, 23, 529-553.

13.

Chesley

Moen

(2006). When workers care: Dual-earner couples’ caregiving strategies, benefit use, and psychological well-being. American Behavioral Scientist, 49, 1248-1269.

14.

Cook

Heath

Thompson

R. L.

(2000). A meta-analysis of response rates in web- or Internet-based surveys. Educational and Psychological Measurement, 60, 821-836.

15.

Cutler

D. M.

Lleras-Muney

(2010). Understanding differences in health behaviors by education. Journal of Health Economics, 29, 1-28.

16.

Dentinger

Clarkberg

(2002). Informal caregiving and retirement timing among men and women: Gender and caregiving relationships in late midlife. Journal of Family Issues, 23, 857-879.

17.

Fredriksen

K. I.

Scharlach

A. E.

(1997). Caregiving and employment. Journal of Gerontological Social Work, 28, 3-22.

18.

Gaugler

J. E.

Anderson

K. A.

Leach

C. R.

Smith

C. D.

Schmitt

F. A.

Mendiondo

(2004). The emotional ramifications of unmet need in dementia caregiving. American Journal of Alzheimer’s Disease and Other Dementias, 19, 369-380.

19.

Given

Sherwood

P. R.

Given

C. W.

(2008). What knowledge and skills do caregivers need? American Journal of Nursing, 108, 28-34.

20.

Idler

E. L.

Benyamini

(1997). Self-rated health and mortality: A review of twenty-seven community studies. Journal of Health and Social Behavior, 38, 21-37.

21.

Jeon

Y.-H.

Brodaty

Chesterson

(2005). Respite care for caregivers and people with severe mental illness: Literature review. Journal of Advanced Nursing, 49, 297-306.

22.

Jylhä

Volpato

Guralnik

J. M.

(2006). Self-rated health showed a graded association with frequently used biomarkers in a large population sample. Journal of Clinical Epidemiology, 59, 465-471.

23.

Kaplowitz

M. D.

Hadlock

T. D.

Levine

(2004). A comparison of web and mail survey response rates. Public Opinion Quarterly, 68, 94-101.

24.

Karlson

K.B.

Holm

Breen

(2012). “Comparing regression coefficients between same-sample nested models using logit and probit: A new method.” Sociological Methodology, 42, 286-313.

25.

Kelly

E. L.

Moen

Tranby

(2011). Changing workplaces to reduce work-family conflict: Schedule control in a white-collar organization. American Sociological Review, 76, 265-290.

26.

Kim

Ingersoll-Dayton

Kwak

(2013). Balancing eldercare and employment: The role of work interruptions and supportive employers. Journal of Applied Gerontology, 32, 347-369.

27.

Kim

Gordon

J. R.

(2014). Addressing the stress of work and elder caregiving of the graying workforce: The moderating effects of financial strain on the relationship between work-caregiving conflict and psychological well-being. Human Resource Management, 53, 723-747.

28.

Kim

Schulz

(2008). Family caregivers’ strains: Comparative analysis of cancer caregiving with dementia, diabetes, and frail elderly caregiving. Journal of Aging and Health, 20, 483-503.

29.

Knight

B.G.

Lutzky

S.M.

Macofsky-Urban

(1993). A meta-analytic review of interventions for caregiver distress: Recommendations for future research. The Gerontologist, 33, 240-248.

30.

Kossek

E. E.

Pichler

Bodner

Hammer

L. B.

(2011). Workplace social support and work-family conflict: A meta-analysis clarifying the influence of general and work-family-specific supervisor and organizational support. Personnel Psychology, 64, 289-313.

31.

Kubicek

Korunka

Hoonakker

Raymo

J. M.

(2010). Work and family characteristics as predictors of early retirement in married men and women. Research on Aging, 32, 467-498.

32.

Lapierre

L. M.

Allen

T. D.

(2006). Work-supportive family, family-supportive supervision, use of organizational benefits, and problem-focused coping: Implications for work-family conflict and employee well-being. Journal of Occupational Health Psychology, 11, 169-181.

33.

Lee

Tang

(2015). More caregiving, less working: Caregiving roles and gender difference. Journal of Applied Gerontology, 34, 465-483.

34.

Lilly

M. B.

Laporte

Coyte

P. C.

(2007). Labor market work and home care’s unpaid caregivers: A systematic review of labor force participation rates, predictors of market withdrawal, and hours of work. The Milbank Quarterly, 85, 641-690.

35.

Lima

J. C.

Allen

S. M.

Goldscheider

Intrator

(2008). Spousal caregiving in late midlife versus older ages: Implications of work and family obligations. Journal of Gerontology, Series B: Social Sciences, 63, S229-S238.

36.

Marks

N. F.

Lambert

J. D.

Choi

(2002). Transitions to caregiving, gender, and psychological well-being: A prospective U.S. national study. Journal of Marriage and Family, 64, 657-667.

37.

Mitra

Bogen

Long-Bellil

L. M.

Heaphy

(2011). Unmet needs for home and community-based services among persons with disabilities in Massachusetts. Disability and Health Journal, 4, 219-228.

38.

Moen

Roehling

(2005). The career mystique: Cracks in the American dream. Lanham, MD: Rowman & Littlefield.

39.

Pavalko

(2011). Caregiving and the life course: Connecting the personal and the public. In Settersten

R. A.

Jr. Angel

J. L.

(Eds.), Handbook of Sociology of Aging (pp. 603-618). New York, NY: Springer.

40.

Pavalko

Henderson

K. A.

(2006). Combining care work and paid work: Do workplace policies make a difference? Research on Aging, 28, 359-374.

41.

Pearlin

L. I.

(2010). The life course and the stress process: Some conceptual comparisons. Journal of Gerontology, Series B: Social Sciences, 65, 207-215.

42.

Pearlin

L. I.

Aneshensel

C. S.

Leblanc

A. J.

(1997). The forms and mechanisms of stress proliferation: The case of AIDS caregivers. Journal of Health and Social Behavior, 38, 223-236.

43.

Pearlin

L. I.

Menaghan

E. G.

Lieberman

M. A.

Mullan

J. T.

(1981). The stress process. Journal of Health and Social Behavior, 22, 337-356.

44.

Pinquart

Sörensen

(2005). Ethnic differences in stressors, resources, and psychological outcomes of family caregiving: A meta-analysis. The Gerontologist, 45, 90-106.

45.

Pinquart

Sörensen

(2007). Correlates of physical health of informal caregivers: A meta-analysis. Journal of Gerontology, Series B: Psychological Sciences, 62, P126-P137.

46.

Pinquart

Sörensen

(2011). Spouses, adult children, and children-in-law as caregivers of older adults: A meta-analytic comparison. Psychology and Aging, 26, 1-14.

47.

Robinson

K. M.

Buckwalter

Reed

(2013). Differences between dementia caregivers who are users and nonusers of community services. Public Health Nursing, 30, 501-510.

48.

Robison

Fortinsky

Kleppinger

Shugrue

Porter

(2009). A broader view of family caregiving: Effects of caregiving and caregiver conditions on depressive symptoms, health, work, and social isolation. Journal of Gerontology, Series B: Social Sciences, 64, 788-798.

49.

Savla

Almeida

D. M.

Davey

Zarit

S. H.

(2008). Routine assistance to parents: Effects on daily mood and other stressors. Journal of Gerontology, Series B: Social Sciences, 63, S154-S161.

50.

Scharlach

A. E.

Kellam

Ong

Baskin

Goldstein

Fox

P. J.

(2006). Cultural attitudes and caregiver service use: Lessons from focus groups with racially and ethnically diverse family caregivers. Journal of Gerontological Social Work, 47, 133-156.

51.

Scheier

M. F.

Carver

C. S.

Bridges

M. W.

(1994). Distinguishing optimism from neuroticism (and trait anxiety, self-mastery, and self-esteem): A reevaluation of the Life Orientation Test. Journal of Personality and Social Psychology, 67, 1063-1078.

52.

Schulz

Sherwood

P. R.

(2008). Physical and mental health effects of family caregiving. American Journal of Nursing, 108, 23-27.

53.

Silva

A. L.

Teixeira

H. J.

Teixeira

J. C.

Freitas

(2013). The needs of informal caregivers of elderly people living at home: An integrative review. Scandinavian Journal of Caring Sciences, 27, 792-803.

54.

Silverstein

Giarrusso

(2010). Aging and family life: A decade review. Journal of Marriage and Family, 72, 1039-1058.

55.

Skaff

M. M.

Pearlin

L. I.

Mullan

J. T.

(1996). Transitions in the caregiving career: Effects on sense of mastery. Psychology and Aging, 11, 247-257.

56.

Szinovacz

M. E.

Davey

(2005). Predictors of perceptions of involuntary retirement. The Gerontologist, 45, 36-47.

57.

Taylor

S. E.

Stanton

A. L.

(2007). Coping resources, coping processes, and mental health. Annual Review of Psychology, 3, 377-401.

58.

Thoits

P. A.

(2011). Mechanisms linking social ties and support to physical and mental health. Journal of Health and Social Behavior, 52, 145-161.

59.

Toseland

R. W.

McCallion

Gerber

Banks

(2002). Predictors of health and human services use by persons with dementia and their family caregivers. Social Science & Medicine, 55, 1255-1266.

60.

U.S. Census. (2010). Age and sex composition: 2010. Retrieved from https://www.census.gov/population/age/

61.

Wagner

D. L.

Hunt

G. G.

(1994). The use of workplace eldercare programs by employed caregivers. Research on Aging, 16, 69-84.

62.

Wang

(Ed.). (2012). The Oxford handbook of retirement. New York, NY: Oxford University Press.