An extensive body of research exists on facilitators and moderators of work-family conflict and stress among employees caring for minor-age children and older adults. Less researched are the work family experiences of parents of children with disabilities. This gap is significant since as Brown and Clark (2017) note in their review of related work-family literature, “Over 16.8 million Americans provide care to children with disabilities under the age of 18 years” and “nearly 70% of these caregivers are employed” while they are actively providing care. It is clear that parents may greatly value the positive changes that parenting a child with special health care needs has brought about in their family life such as an increased sense of personal strength, appreciation for their child’s accomplishments, and development of meaningful relationships with other parents raising children with disabilities (Beighton & Wills, 2017; Farrell & Krahn, 2014). However, employed parents providing exceptional care for their children with disabilities or chronic conditions may struggle to meet both work and family demands. Understanding how workplace supports affect employed parents of children with disabilities is important given that the U. S. Department of Health and Human Services (2013) has reported that approximately 25% of parents caring for a child with a special health care need or disability indicate that this care responsibility affects their ability to work. Hence, the field needs a better understanding of specific personal, family, and workplace supports that may serve as protective factors for working parents of youth with disabilities and chronic health conditions (Brown & Clark, 2017). More research is also crucial for understanding how supports may differ for working parents caring for typically-developing children.

Using data from the 2016 National Study of the Changing Workforce (NSCW) this study investigated whether personal, family, and workplace characteristics differ between parents providing disability-related care compared to those providing typical family care. A test of a moderation model determined the influence of type of child care demand (typical versus disability-related) and workplace supports (job flexibility, work autonomy, organizational support, supervisor support, coworker support) on work and family conflict and perceived stress

Theoretical Background

Two theoretical frameworks are used to situate our study: conservation of resources theory and the continuum of dependent care. These theories allow for an understanding of exceptional care responsibilities compared to typical care responsibilities, and how personal, family, and workplace resources and demands affect conflict and stress among employed parents.

Conservation of Resources Theory: Demand-Resource Gains and Losses

Conservation of resources (COR) theory (Hobfoll, 1989, 2011) can be used to explain how working parents use instrumental and social resources to manage family and work demands and to specify resources needed to meet demands. Understanding the complexity of, and interactive relationships among, resources and demands is necessary to better comprehend the strategic choices available to parents of children with disabilities and chronic health conditions and to examine how those choices impact stress and well-being (Sellmaier, 2019). COR proposes that individuals seek out resources to buffer against loss. Stress results from resource loss, threat of resource loss, or the absence of resource gain after a loss. COR also argues that individuals can develop a surplus of resources in low-stress times. This surplus of resources in contrast adds to a person’s well-being and health. Resources can be: conditions for example employment, objects like a house or car, personal resources including skills, and energies such as money. Resources are defined broadly as anything that helps an individual achieve their goal (Halbesleben et al., 2014). Resource loss typically occurs at a higher rate than resource gain, and once resources are lost, they are difficult to regain (Hobfoll, 1989). For example, owning a car with a lift to transport an electric wheelchair is an object-type resource. Access to this resource reduces stress, and the loss of this resource could increase stress if the loss cannot be buffered by the purchase of a similar car. Access to workplace flexibility can be a condition that supports parents to meet both work and care demands, reducing conflict and stress. The experience of caring for a child with disabilities or chronic health conditions is therefore affected by the family’s access to resources, potentially increasing stress or enhancing well-being.

The Continuum of Dependent Care

Dependent care is defined as the provision of emotional, developmental, or physical resources to a person who must rely on others (Rothausen, 2016). The Continuum of Dependent Care model (Stewart et al., 2018) places typical caregiving demands at one end of the continuum, and exceptional (i.e., disability-related) caregiving demands at the other. Movement along the continuum shifts as care demands, resources, strategies, and contexts change. Activities involved in caring for children with typical development, and to a limited extent, older adults are known as typical care demands (Stewart, 2013). Exceptional family care demands are part of the experience of caring for a family member with a disability or chronic health condition (Roundtree andand Lynch 2006), and are characterized by crises and heightened physical, emotional, social, and financial demands on families that reflect the trajectory of the disability or chronic health problem (Brennan et al., 2016).

Several key factors shape the extent to which dependent family care is typical or exceptional (Stewart et al., 2018). These include: (1) the intensity of the care demands (e.g., number of dependents, age of dependents, direct care activities, indirect care activities); (2) the typology of the care demands (onset, course, outcome, timing); (3) the complexity of the care demands (type and degree of knowledge and skills about caring tasks); and (4) type of resource demands (family, workplace, community). For example, a carer could have a dependent with a disability that requires very minor specialized knowledge, care demands, or interface with health resources in the community. Conversely, one might provide care for a dependent with health care needs that involves specialized knowledge related to the disability and requires a high degree of coordination and interface with health and specialized community support services (Bradshaw et al., 2019). The Continuum of Dependent Care model is useful for understanding how unique disability characteristics might influence the demands-resources gain and loss cycles, which in turn influence conflict and perceived stress among employed parents of children with disabilities (see Stewart et al., 2018 for conceptual model).

Personal, Family, and Workplace Characteristics as Resource Loss and Gains

Research has identified a number of personal, family, and workplace characteristics that affect work-family conflict and stress (Brown & Clark, 2017; Hammer et al., 2011; Kossek & Thompson, 2016; Powell & Greenhaus, 2010; Sellmaier, 2019; Stewart, 2013). The following sections will discuss these factors identified as salient for both work-family conflict and stress for parents with typical and exceptional care demands. In general, the presence of characteristics such as supervisor support or family flexibility, can be considered resource gains for families, which can reduce work-family conflict (Hammer et al., 2011). The absence of certain characteristics can act as resource loss, aggravating negative consequences of workplace and family demands (Martinengo et al., 2010). Personal, family, and workplace characteristics also interact with each other, creating patterns of resource loss and gain. Less understood is how these characteristics influence these outcomes for parents of children with disability care demands. For example, the risk that caring for a child with a disability may become onerous might be mitigated by workplace organizational supports that protect families against increased work-family conflict and stress (Rosenzweig et al., 2008; Stewart, 2013). Comparisons are needed regarding how these characteristics might differ as a result of the type of care demands that a parent experiences.

Personal Characteristics

Current research has identified several individual characteristics that may influence work-family conflict and stress. Gender differences, in particular, have been identified as a key predictor of work-family conflict for both employed parents of children with typical and exceptional care demands (Aumann et al., 2011; Stewart, 2013). Certain workplace or family resources and demands also have a differential effect on work-family conflict and stress based on gender. For example, a lack of workplace support was a greater predictor of negative work-to-family spillover for women, but not for men, and low levels of family burden were correlated with more positive family-to-work spillover for women and not for men (Grzywacz & Marks, 2000).

A second personal characteristic thought to predict work-family conflict and stress among employed parents of children with child care demands is workers’ age. Matthews et al. (2010) reported younger employees experienced significantly more work-family conflict than older employees, but employees between ages 29–45 experienced the highest levels of family-work conflict. Sellmaier (2019) found less negative work-family spillover for older fathers caring for children with special health care needs. These results align with theoretical linkages of work-family conflict across the lifespan (Moen, 2011).

Research on the influence of parental race and ethnicity on work-family conflict and stress established that differences do exist between groups (Bianchi & Milkie, 2010). Ammons et al. (2017) found lower levels of work-family conflict among Hispanic employees compared to non-Hispanic White and Black employees. A recent investigation by Kim et al. (2020) revealed that family resilience reduced parenting stress for African American and for White caregivers of children with autism spectrum disorders, but not for Hispanic caregivers. More research is needed to clarify the impact of race/ethnicity on exceptional care.

Education can influence levels of work-family conflict and stress. In an analysis of the 2002 NSCW, Glavin and Schieman (2012) found workers with higher levels of education tended to work in professional level jobs with higher work demands and pressures. Employees with lower levels of education worked in jobs with less control over working conditions and variable shifts. Further research is needed to clarify the role of personal characteristics to better target workplace supports designed to mitigate conflict experienced by employed parents with and without exceptional care demands (Stewart et al., 2018).

Family Characteristics

Several family characteristics are thought to influence or moderate the impact of work-family conflict experienced by employed parents with typical or exceptional care demands. Exceptional care responsibilities can be understood as additional demands on family resources (Stewart et al., 2018). Exceptional care demands such as more time required to provide direct and indirect child care, as well as the need for complex and specialized knowledge and skills may affect parent well-being (Hilbrecht et al., 2017; Moorman & Macdonald, 2012). Intensity of the child’s symptoms was a negative predictor of positive family-work spillover for fathers caring for children with special health care needs (Sellmaier, 2019), and increased parental caregiver strain (Brannan & Heflinger, 2001). Families with children under age 5 or with exceptional care responsibilities have been characterized as having low home control and high home demands (Moen et al., 2008). Employees in these families reported more work-family conflict and negative work-family spillover compared to employees with low home demands and high home control. Higher family demands also increased family-work conflict for a sample of employees providing care for a family member with a disability (Li et al., 2015). Having more children in the home can also be an additional family demand (Michel et al., 2011). While exceptional care demands without access to supportive resources can increase stress, it is critical to add that this is only one side of the care experience. For example, fathers in one study reported that they value their time spent with their children and that this is worth adjusting work and leisure activities (Sellmaier & Buckingham, 2021). Families with exceptional care responsibilities thrive especially when they have adequate socio-economic supports (Farrell & Krahn, 2014; McConnell et al., 2014).

Marital status may affect conflict and stress. Sharing care responsibilities with a partner may be a key family resource for employees caring for dependent children. DeRigne and Porterfield (2010) found that single mothers provided more hours in direct home care and care coordination than married mothers. Marital status also affected positive family-to-work spillover for men and women generally (Grzywacz & Marks, 2000), and family-to-work conflict for families with exceptional care demands (Stewart, 2013). More hours worked by a partner have also been found to increase work-to-family conflict for families with exceptional care demands (Brown, 2014). However, this also contributes to greater household income, which can be an important resource (Emlen, 2010; Hilbrecht et al., 2017; Moen et al., 2008).

Family flexibility refers to family-based resources to meet work and care demands through sharing care responsibilities with a partner or another family member and can be an important factor when examining work-family conflict and stress. For example, a meta-analysis found a relationship between spousal support in completing daily parenting tasks and family-to-work conflict in the general population (Michel et al., 2011). Disability research revealed that parents with exceptional care demands have reported less family flexibility (Emlen, 2010). These families may have maximized their boundary spanning solutions so when care demands increase, they are not able to draw additional resources from the family (Emlen, 2010; Hilbrecht et al., 2017). Because of this, employed parents of children with exceptional care demands are more likely to draw on workplace supports to meet heightened care demands (Brennan et al., 2016). At the same time access to family-based resources can strengthen positive family-work spillover for parents with both typical and exceptional care demands, resulting in parents drawing on their family experiences to meet work challenges (Grzywacz & Marks, 2000; Sellmaier, 2019)

Workplace Characteristics

For employees with family care demands, the supervisor is often the primary source of support at work and the gatekeeper to informal resources in the workplace (Matthews et al., 2014). Family supportive supervisors show empathy and understanding about employee family care demands, provide information about organizational supports, act as examples of how to balance work and family life, and organize work to accommodate both employee and organization (Hammer et al., 2009). Kossek et al. (2011) contend that work-family conflict cannot be addressed by supervisor and organizational empathetic support only but requires specific, concrete, and relevant supports. The degree of supervisor support is a significant predictor of work-family conflict for parents of children with autism spectrum disorder (Brown, 2014). Coworker support, which is defined broadly as the support from coworkers to manage work and care demands, was also found to be an important resource for addressing work-family conflict for parents (Aumann et al., 2011). Additionally, Allen (2001) demonstrated that employees’ perceptions of the level of organizational family support in their workplace and the actual benefits offered by the organization reduced work-to-family conflict for employees. Organizational support effects were also found for families with exceptional care demands (Brown & Clark, 2017). Workplace flexibility, which can include control over one’s schedule or over work location, is another important resource for working parents. Based on a meta-analysis of 58 studies, Allen et al. (2013) found that having flexible workplace arrangements was significantly associated with less work interfering with family (WIF). The same effect was not found for family interfering with work (FIW). Greater use of schedule flexibility was the strongest predictor of having less WIF. Greater availability of schedule flexibility, and availability and use of place flexibility were also related to less WIF. Interestingly, a study of employed fathers with exceptional care responsibilities found that access to job flexibility was a positive predictor for negative spillover, with fathers with greater access to job flexibility reporting more negative work to family spillover (Sellmaier, 2019). The same study also reported that greater use of job flexibility was a predictor of greater negative family to work spillover. Yet a national comparison of employees found those caring for dependents with higher disability-related care demands reported less use of flexible work arrangements compared to those providing typical child care (Stewart, 2013).

Figure 1 depicts how the relationships among the family and work variables discussed above will be represented in this study. Exceptional caregiver impacts resources through the cost of child services and reduction of work as a result of the child-caring demands. Exceptional caregiving impacts work outcomes (work-family conflict, family-work conflict, perceived stress) directly and that work variables provide some mediation of those relationships. Exceptional caregiving influences the types of jobs that caregivers seek and maintain. Exceptional caregivers are more likely to work in jobs with more flexibility, autonomy, and support. Exceptional caregiving also directly influences employment-family outcomes (i.e., work-family conflict, family-work conflict, and perceived stress). Workplace conditions (e.g., job flexibility and autonomy) and workplace support (i.e., organization, supervisor, co-worker) are expected to directly influence conflict and stress. They may also mediate the relationship between exceptional caregiving and employment-family outcomes. This suggests that the greater job flexibility, autonomy, and support employees experience, the less conflict and stress exceptional caregivers experience. While family resources (e.g., parent-employee education, income) have a direct impact on employment-family outcomes, workplace characteristics are expected to have a greater influence on outcomes. Building from the theoretical models and research literature that guide this research, we examined the following hypotheses. The first two hypotheses address the differences in personal, family, and workplace characteristics between employed parents of children with typical child care demands and parents with exceptional care demands and their influences on work-family conflict (work-family, family-work) and perceived stress. The subsequent hypotheses address the potential moderating effect of having exceptional child care responsibility on the use of workplace supports and the reported effect on work-family conflict (work-family, family-work) and perceived stress.

Fig. 1
figure 1

Conceptual Model of Exceptional Caregiving and Employment-Family Outcomes

Full size image

Ha1: Employed parents of children with exceptional care demands will report lower levels of workplace supports compared to employed parents of children with typical child care demands.

Ha2: Employed parents of children with exceptional care demands will report higher levels of work-to-family conflict, family-to-work conflict, and perceived stress.

Ha3: Having exceptional care demands will be associated with work and family conflict, and perceived stress, when workplace supports are controlled.

Ha4: Having exceptional child care demands will moderate the effects of workplace supports on reported levels of conflict (work-family, family-work), and perceived stress.

Method

We used data from the 2016 Society for Human Resource Management (SHRM) National Study of the Changing Workforce, conducted by the Data Recognition Corporation (DRC) Inc., for the Families and Work Institute, which became a SHRM project in 2016. The NSCW has been designed to include a nationally representative sample of the U.S. workforce through a random digit dialing (RDD) method. Data were collected in 2015 through online phone interviews and an online questionnaire. One member from each household was selected for the interview. The sample was augmented through surveys completed by members of NORC and the Amerispeak panel (SHRM, 2020). The 2016 NSCW response rate was 16%, which meets the standard for similar workforce studies (between 5–15%). The total sample for the study was 1510 employed workers. The data are available for public use in a format that does not include information permitting identification of individual participants. The current study was approved and monitored by the Institutional Review Board of Portland State University.

Participants

Participants for this study were a subsample (n = 862) of the total sample (N = 1510) of employed parents. We used the following inclusion criteria to draw our sample for analysis: (1) working in government, for-profit, or non-profit sector, (2) having a minor child living at home for at least half the time; and (3) having complete data on the variables included in these analyses.

Employed parents providing typical or exceptional care were identified using two items: (a)“How many of your children have a chronic illness or medical condition, a disability or other notable health problems?” and (b)“How many of your children had emotional or developmental problems in the past year?” Parents reporting 0 for both items were coded as having typical care demands (n = 642; 74%). Parents whose responses were greater than 0 on either item were coded as having exceptional care demands (n = 220; 26%). Table 1 provides the demographic characteristics of the sample of individuals who were identified as parents with typical and exceptional care demands.

Table 1 Percentages, means, and standard deviations for key variables
Full size table

Measures

In this study, the primary outcomes of interest are work-family conflict and stress. Predictor variables included work-related factors, personal factors, and family factors.

Conflict and stress

Work-to-family conflict (WFC) measures assessed the extent to which parents felt their work created conflict at home through four items using a 5-point Likert type scale (1 = very often to 5 = never). Items were reverse-coded so higher scores indicated more WFC. This scale demonstrated strong internal consistency with this sample (Cronbach alpha = 0.93), as well as in Voydanoff’s 2005 study. Sample items are “How often have you not had enough time for your family or other important people in your life because of your job?” and “How often has work kept you from doing a good job at home?” Family-to work conflict (FWC) measures the extent to which workers reported that family demands interfered with their job (α = 0.94; see Voydanoff, 2005; Nichols & Swanberg, 2018). FWC was assessed through four items using the same Likert scale as WFC. Items included:: “How often has your family or personal life kept you from doing as good a job at work as you could?” and “How often have you not had enough time for your job because of your family or personal life?” Again, items were reverse coded so higher scores indicated more FWC. Stress was measured through seven items using a Likert-type scale (Cohen et al., 1983). Respondents were asked to indicate their level of stress in the past month. Sample items include “Been bothered by minor health problems such as headaches, insomnia, or stomach upsets?” and “Felt that difficulties were piling up so high that you could not overcome them?” Response categories for the items ranged from 1 = never to 5 = very often (α = 0.78).

Work-related measures

Workplace support measures used in primary analyses included job flexibility, job autonomy, organizational support, supervisor support, and co-worker support. Job flexibility was constructed from four items: (1) difficulty taking time off, (2) ability to choose start and end time, (3) ability to temporarily change start and end time, and (4) ability to arrange part-time and full-time work schedule. Items (2) to (4) were dichotomous variables including 0 = no and 1 = yes. Difficulty taking time off was recoded from a 5-point Likert scale (i.e., 1 = very hard to 5 = it depends) into a dichotomous variable, with participants indicating that taking time off was very hard or somewhat hard as having no job flexibility, and participants reporting difficulty taking time off as not too hard, not hard at all, or it depends as having job flexibility. The sum of these four scheduling variables was used to measure overall job flexibility. The job autonomy scale was calculated as the mean of three items rated on a 4-point scale (1 = strongly disagree, to 4 = strongly agree). Items included were: “I have the freedom to decide what I do;” “I have a lot of say about what happens at my job;” and “I decide when I take breaks” (Winfield & Rushing, 2005). Internal consistency reliability for this scale was acceptable with this sample (Cronbach alpha = 0.68). Overall job satisfaction was also measured with one item measured on a 4-point Likert scale that asked, “How satisfied are you with your job?” with 1 = very satisfied and 4 = not satisfied at all, which was reversed coded for this analysis. Organizational support for work life balance variable was constructed as the mean of three items that were rated on a 4-point Likert scale (1 = strongly disagree; 4 = strongly agree). These items included: “My job lets me create and sustain healthy relationships with the people in my life;” “My job lets me deal with my feelings about work and life in a healthy way;” and “I have the schedule flexibility I need at work to manage my personal and family responsibilities.” Items demonstrated good internal consistency reliability (Cronbach alpha = 0.80). The supervisor support scale was calculated as the mean of 11 Likert-scale items rated on a 4-point scale (1 = strongly disagree to 4 = strongly agree). Sample questions are: “I am comfortable bringing up family/personal business with my supervisor/manager;” and “My supervisor or manager really cares about the effect that work demands have on my personal and family life” (Winfield & Rushing, 2005). Internal consistency reliability this scale was good in this sample (Cronbach alpha = 0.83). Coworker support was measured as one item rated on a Likert scale (1 = strongly disagree; 4 = strongly agree). Participants indicated their agreement with the statement, “I have support from coworkers that helps me to manage my work and personal or family life.” Higher scores on these work-based measures indicate higher levels of flexibility, autonomy, satisfaction, or support. Other work-related characteristics are reported for descriptive purposes such employer type, access to time off, work hours, and number of jobs.

Personal and family characteristics

Based on the theoretical frameworks that guide this research, several employee and family characteristics served as control variables in primary analyses. Personal employee variables included age, gender, race/ethnicity, and education. The key family variable of interest was whether the respondent was engaged in typical or exceptional caregiving, as described in the Participant section. Other family variables included partner status, number of children under 18 in the home, and age of youngest child living in the household and total annual household income from all sources (see Table 1). Average number of hours worked per week by the respondent’s partner was also included. Respondents with no live-in partners were assigned zero partner hours worked.

Analyses

Hierarchical regression analyses were conducted to test the direct and moderated effects of exceptional caregiving on outcome variables. In these analyses, WFC, FWC, and perceived stressed served as outcome variables. Workplace characteristics (i.e., job flexibility, job autonomy, and organizational, supervisor, and co-worker support) served as the moderator variables. Separate regression analyses were conducted for each of the outcome variables. Personal, family, and workplace factors were entered in progressive steps of the hierarchical regression model. Step 1 tested the influence of demographic and family characteristics on the outcome variables. Step 2 added workplace characteristics including job flexibility, job autonomy, organizational support, supervisor support, and coworker support. Moderated relationships were tested in step 3 using interaction terms constructed from type of care (i.e., typical versus exceptional care) and job characteristics (i.e., job flexibility, job autonomy, organizational support, supervisor support, coworker support). To construct the interaction terms, workplace continuous variables were mean-centered in keeping with recommended practice (Hayes, 2018).

Results

Characteristics of Employed Parents with Typical and Exceptional Care Demands

Table 1 shows the frequencies, means, and standard deviations of the demographic and job characteristics for employed parents with typical and exceptional care demands. For the t-tests we report the mean difference (mean diff.) which is a statistic that measures the absolute difference between the mean values in our two groups. Bivariate tests of demographic characteristics indicated several significant differences between the two groups of participants with associations between being a parent of children with exceptional care demands and older age (mean diff. = −1.70; d = 0.15), being female (χ2 (1, n = 862) = 5.95, p < 0.05, phi = 0.08), and less likely to be currently married (χ2 (3, n = 862) = 15.65, p < 0.001, Cramer’s V = 0.13) compared to participants giving typical care. Small but significant differences were found between the two groups with those giving exceptional care having older youngest children on average (mean diff. = −1.29; d = 0.18) and more children under 18 years of age in their home (mean diff. = −0.23; d = 0.22), compared to typical care.

Workplace Characteristics of Employed Parents with Typical and Exceptional Child Care Demands

Table 2 summarizes the workplace conditions of caregivers in this sample, The majority of parents in the sample with typical child care demands (61.7%) and those with exceptional care demands (59.4%) worked in the for-profit sector (see Table 2). Parents worked an average of 43.28 h per week (SD = 14.01). Most had access to certain leave benefits, such as paid sick leave and vacation time. Interestingly, significantly fewer parents of children with exceptional care demands agreed with the statement that they could take five or more days off for a sick child without penalty than did parents with typical care demands (46.7 and 56.1% respectively; X2 = 4.01, p < 0.05; phi = 0.08). Accordingly, 32% of parents with exceptional care demands reported that it was somewhat hard to take time off and 13.7% reported that it was very hard. A similar percentage of parents with typical care demands reported that it was very hard to take time off (13.9%), but only 20% reported that it was somewhat hard. More than 75% of parents in both groups reported that they had a partner/spouse who worked for pay, with partners/spouses working approximately 39 h per week on average across both groups. Examining outcomes related to Ha1, parents of typically developing children scored higher on all job measures than parents of children with exceptional care demands (see Table 3). Parents of children with exceptional care demands reported significantly lower levels of job flexibility (mean diff. = 0.2287, p < 0.05, d = 0.19) and job autonomy (mean diff. = 0.1334, p < 0.05, d = 0.15). A significant difference was found for organizational support with parents caring for children with exceptional care demands indicating lower levels of organizational support (mean diff. = 0.1164, p < 0.05; d = 0.16). Parents in both groups reported fairly high levels of job satisfaction. Non-significant differences between the two groups were found on supervisor support (mean diff. = 0.0563, p < 0.36) and coworker support (mean diff. = 0.1015, p < 0.12).

Table 2 Job-related characteristics by type of care demand
Full size table
Table 3 Hierarchical regressions predicting work-family, family work conflict (N = 482)
Full size table

Work and Family Conflict and Perceived Stress Outcomes

Bivariate tests of differences between groups on conflict (work-family conflict, family-work conflict) and perceived stress measures are presented in Table 2 and test Ha2. Significant small to medium differences were found between groups. Parents of children with exceptional care demands reported higher levels of work-family conflict (mean diff. = −0.2009; p < 0.01; d = 0.20) and family-work conflict (mean diff. = −0.2134; p < 0.05; d = 0.19). Parents providing exceptional care also had higher levels of perceived stress (mean diff. = −0.3213, p < 0.001; d = 0.43).

Effects of Exceptional Care Responsibilities and Workplace Supports on Work and Family Conflict and Perceived Stress Outcomes

There were a number of strong associations between workplace supports and both work-family and family-work conflict with organizational support having both a strong and negative association with work-family conflict (r = −0.36, p < 0.001) and family-work conflict (r = −0.11, p < 0.001). All workplace supports had negative and significant associations with perceived stress as reported in Supplementary Table 2.

Tables 3 and 4 present the results of the regression models for work-family and family-work conflict and perceived stress that address Ha3 and Ha4. Step 2 was the best-fitting model for work-family conflict as indicated by the change in R2 estimates (i.e., increase in the amount of variance explained in each successive step). For the step 2 model, the main effects model accounted for 14% of the variance in work-to-family conflict (F = 5.99, R2 = 0.14, p < 0.001). Having exceptional care demands was positively related to work-to-family conflict (i.e., β = 0.10, p < 0.01). Job autonomy (β = 0.21, p < 0.01) and organizational support (β = −0.42, p < 0.001) were the only two workplace supports that significantly contributed to the explanation of variance in the main effects model.

Table 4 Hierarchical regressions predicting perceived stress (N = 474)
Full size table

The demographic, family, and job characteristics accounted for 4% of variance in the main effects model for family-work conflict. Step 1 demographic and family characteristics predicted 4% of the variance in family-to-work conflict (Ha3). One demographic and one family characteristic significantly predicted family-to-work conflict: not being married (β = −0.15, p < 0.05), and having exceptional care demands (β = 0.13, p < 0.01). Step 2 job characteristics accounted for a 3% increase of variance explained for family-to-work conflict (∆ F (5, 466) = 3.14; ∆ R2 = 0.03, p < 0.01). Not being married and having exceptional care demands remained statistically significant predictors when the workplace supports were added to the model. Organizational (β = −0.15, p < 0.01) and supervisor (β = −0.12, p < 0.05) supports were significantly associated with lower levels of family-work conflict. The interaction model (Ha4) was significant at the level of a trend (∆ F (5, 461) = 2.12, ∆ R2 = 0.02, p < 0.06). Simple slopes tests on the moderated effects of organizational support (see Fig. 2) established that family-to-work conflict was moderated by organizational support for employed parents of children with exceptional child care demands (t (17,474) = 4.52, p < 0.001) but not for those with typical child care demands (t (17,474) = −1.26, p < 0.20). The results of a simple slopes test (see Fig. 3) established family-to-work conflict was moderated by coworker support for employed parents of children with exceptional child care demands (t (17,494) = 3.35, p < 0.001) but not for those with typical child care demands (t (17,474) = 0.38, p < 0.70).

Fig. 2
figure 2

Moderating Effect of Type of Care Demand on Organizational Support and Family-Work Conflict

Full size image
Fig. 3
figure 3

Moderating Effect of Type of Care Demand on Coworker Support and Family-Work Conflict

Full size image

The main effects model for perceived stress accounted for 17% of the variance in stress. Step 1 demographic and family characteristics predicted 10% of the variance in perceived stress: Being female (β = 0.12, p < 0.05), having less income (β = −0.16, p < 0.01) and having exceptional care demands (β = 0.20, p < 0.001) were associated with greater levels of reported stress. In Step 2 job characteristics accounted for an 8% increase in the main effects model (∆ F (5, 458) = 9.38; ∆ R2 = 0.08, p < 0.001). Lower organizational support was associated with higher levels of perceived stress (β = −0.23, p < 0.001). Having exceptional care demands remained significant in Step 2 (Ha3). In Step 3, the moderated effects model (Ha4) for perceived stress significantly improved the explanation of the variance in stress (i.e., ∆ F (5, 453) = 2.76; ∆ R2 = 0.02, p < 0.05). Being female (β = 0.16, p < 0.001), having less income (β = −0.13, p < 0.05) and having a child with a disability (β = 0.19, p < 0.001) remained significant personal and family predictors. Organizational support remained a negative predictor of stress in Step 3 (β = −0.27, p < 0.001). Findings from simple slopes analysis (see Fig. 4) indicated that greater job flexibility was associated with higher levels of stress for parents with exceptional child care demands (t (17,474) = 2.72, p < 0.05) but not for those with typical care demands (t (17,474) = 0.38, p < 0.13). Simple slope tests (see Fig. 5) also demonstrated that organizational support was associated with significantly lower levels of stress for parents with typical child care demands (t (17,474) = −4.28, p < 0.001) but not for those with exceptional care demands (t (17,474) = −1.47, p < 0.14).

Fig. 4
figure 4

Moderating Effect of Type of Care Demand on Job Flexibility and Perceived Stress

Full size image
Fig. 5
figure 5

Moderating Effect of Type of Care Demand on Organizational Support and Perceived Stress

Full size image

Discussion

Employed parents raising children with disabilities or notable health problems encounter both rewards and challenges as they provide exceptional care and focus on supporting their children to thrive and reach their full potential (Beighton & Wills, 2017; Farrell & Krahn, 2014). Our study revealed that although parents providing exceptional care might receive support in their workplaces, they still report significantly more work to family and family to work conflict and higher levels of stress than do parents of children with typical development and health care needs. Regression analyses found that exceptional caregivers experienced more WFC, FWC, and stress than typical caregivers, when the analysis controlled for other potential contributors, even when the analyses controlled for work-related supports. This suggests that, though some work-related supports appear to reduce work-related conflict and stress for both groups of workers, they are not sufficient to close the gap between exceptional and typical caregivers. The resource-demands proposition in the Continuum of Dependent Care Model (Stewart et al., 2018) suggests that resources need to fit specific disability demands. In the case of parents with exceptional care demands, needs can be crisis-driven, demanding a high level of resources for long periods of time (Brennan et al., 2016).

Results further support both COR theory (Hobfoll, 1989; 2011) and the Continuum of Dependent Care Model (Stewart et al. 2018). When FWC levels were higher, employed parents caring for children with disabilities rated both co-workers and the organization itself as more supportive, unlike parents providing typical care whose ratings of those workplace supports did not reflect their FWC ratings. Because parents providing exceptional care may have disclosed their family situation to coworkers when high care demands conflicted with job requirements, they may have rated co-workers as more supportive than others providing typical care having less FWC conflict. Disclosure at work of family issues can be trying for parents struggling to manage exceptional care, but coworker support can provide resources that make job retention possible (Rosenzweig et al., 2011). Organizational support was also rated higher by parents providing exceptional care who reported higher levels, rather than lower levels of FWC. When organizations have key personnel who disseminate crucial information about work-life policies and supports, mediate issues between beleaguered workers and their supervisors, and build networks of support among co-workers, even difficult family demands can be managed. (Brennan et al., 2016). Consistent with COR theory (Hobfoll, 1989; 2011), workplace resources can be more impactful when demands for resources are high.

Findings about elevated levels of stress in families with exceptional care demands are consistent with the Continuum of Care Model (Stewart et al., 2018) which suggests that family care demands occur on a continuum influenced by not only the intensity of the care demand but also the type and complexity of the demands influencing the well-being of those providing family care. On the well-established measure of perceived stress (Cohen et al., 1983) employees responsible for exceptional care rated their stress levels as substantially higher than those giving typical care. There are promising stress-reducing and health-promotion interventions being developed and tested with parents providing exceptional care, including health coaching, fostering personal coping skills, and linking with support networks (Ogourtsova et al., 2019; Whiting et al., 2019). To reduce stress, it will be important for workplaces to link employees to such critical support through workplace wellness programs.

Findings from regression analyses indicated that work-related factors appear to provide some protection from negative outcomes for both groups of workers, overall. Although the effect sizes were small, greater organizational support was associated with lower WFC, while co-worker support was related to FWC. Organizational support also appeared to protect employed caregivers from stress. Analyses of the moderation effects of type of caregiving indicated that the health-related benefits of workplace support accrue primarily to typical caregivers. A surprising result of the bivariate and regression analyses was the lower job flexibility reported by parents giving exceptional care, compared to those providing typical care. In fact, perceived stress was elevated for parents with exceptional care responsibilities who had higher levels of flexibility at work. Although obtaining flexible work hours has been reported as a strategy used by some parents raising children with special needs (Scott, 2010), job flexibility also could have the effect of blurring boundaries between work and home (Glavin & Schieman, 2012) and actually increase, rather than reduce, stress levels. Given both the lower job flexibility and lower ratings of job satisfaction reported by exceptional caregivers, it is possible that these parents may have had to obtain employment that fits with care responsibilities rather than find work that matches their interests and qualifications (Warfield, 2005). Fostering family-supportive supervisory behavior can be an important organizational health and wellness strategy to improve the job satisfaction and retention of all employees, but may be a particularly important resource for parents providing exceptional care (Matthews et al., 2014).

Last, findings from regression analyses also indicate that some personal and family factors appeared to protect caregivers from negative work-related and health outcomes. Being married reduced family-to-work conflict. Higher income was related to lower stress. These findings are consistent with the Continuum of Dependent Care Model (Stewart et al., 2018) which suggests that personal predictors can influence both demands and available resources for employed parents providing exceptional care to children and youth.

Study Limitations

All of the data used in this study were self-reported by employees and were gathered at a single point in time. As a result, relationships may be exaggerated. For example, it is possible that caregivers who had higher levels of stress at the time of the study, also indicated that they had higher workplace flexibility, since they had a greater need to find and make flexible work arrangements than other respondents. Because of the comprehensive nature of the NSCW, some of the key variables were measured through single items or with scales with small numbers of items, which were not standardized. Our study also found that 26% of parents reported raising a child or youth with a health condition or disability, a relatively high proportion when compared with standard census or national survey estimates. For example, Child Health USA 2014 reported 19.8% of children under the age of 18 had special health care demands (U.S. Department of Health and Human Services, 2015). The 2016 NSCW included six standard questions that were used to identify respondents who had disabilities; Galinsky et al. (2020) found that 13% of their nationally-representative sample identified themselves as a person with a disability and/or indicated having a disabling condition. Questions on their own disability status may have affected the employed parents’ reports of the disability status of their children. Unfortunately, the data set did not include measures that reliably represented the intensity of care provided for children with exceptional care needs. Without this measure, we were not able to estimate the time devoted to care, but instead focused on the presence or absence of a child’s condition requiring exceptional care and could only explore the relationship of type of care given to other major study variables rather than include care intensity in our analysis. The survey also did not measure the positive spillover of work which might relieve the family stress experienced by parents providing high levels of exceptional care. As Morris (2014) has noted, mothers providing care for children with disabilities had more positive spillover from work on their personal mental health than did mothers providing care for children developing typically.

The set of resources included in this study was limited by the availability of measurements in the existing data set. For example, community supports that may have affected levels of work-family conflict and stress were not measured in this specific data set. While the data collected in NSCW provided measurement of key variables relevant for work-family conflict, they did not include items on community resources like inclusive child care, school supports for children with special needs, and child health care coordinated services that are very relevant to the experiences of families providing exceptional care. The data set did also not include measures of positive work-family spillover, focusing the results on stress and strain without assessing areas of thriving and enhancement related to participants’ work and care responsibilities. It should also be noted that the data were collected in 2015, and represent the employment, health, and family support conditions of that time.

Finally, a number of main and indirect effects between predictors and our outcome variables were small to moderate. Whether these associations are plausible can be assessed using the magnitude of the difference using Cohen’s d which is interpreted as 0.20 as small, 0.50 as medium, 0.80 as large effects (Cohen, 1988). Future research with this population of workers using the same predictor and outcome measures would increase confidence that the magnitude of the effects are indeed feasible (Matthay et al., 2021).

Implications

Findings from our study suggest that exceptional care demands are directly related to work family conflict and stress. Results also connect workplace supports for those with exceptional care demands to their wellbeing. These findings offer some direction for the way forward. Work-related support may improve work-life balance and reduce stress. Findings from this study, however, indicate that workplace supports may need to be tailored to the specific requirements of exceptional caregivers.

Clearly, the current results point to some key reasons for organizations to provide support for the substantial proportion of their workers who are parents caring for children and youth with disabilities. As found in previous studies, employed parents responding to the 2016 NSCW who provided exceptional care had higher levels of work and family conflict than those raising children with typical development (Brown, 2014; Brown & Clark, 2017; Stewart, 2013). For parents balancing employment with responsibilities for children with disabilities or special health needs, family supportive organizational cultures have proven to help them stay employed and manage the work-family conflict they experience (Brown, 2014; Stewart, 2013).

Another provocative finding was that parents providing exceptional care also reported less access to the use of five or more days of annual family leave to care for their dependents, compared with those providing typical care. Combined with the lower job flexibility experienced by some exceptional caregivers, the consequences could be quite serious, including unexcused work absences, reduction of work hours, and loss of talented employees (Earle & Heymann, 2012). Particularly important are policies allowing employees to take leave to care for their children’s special needs, and to manage both work and caring demands through control over their work timing and location (Wakefield et al., 2014). Clearly, workplaces need to incorporate specific strategies into their organizational policies and health and wellness programs that will support stabilizing the employment of this substantial group of exceptional caregivers (DeRigne & Porterfield, 2017; Stewart, 2016).

The current study provides important insights into the effects of individual, family, and workplace resources on work family conflict, and stress. Future research should expand on these findings, by including measures of community resources. The theoretical concepts of conservation of resources and the continuum of care framework acknowledge the relevance of resource loss or gain across a number of different ecologies including informal and formal community supports, and research has examined these connections empirically as well (Sellmaier, 2019; Stewart, 2013). This body of research should be expanded specifically by comparing the relevance of community resources and social supports for parents with and without exceptional care responsibilities (Woodman, 2014). A closer look at certain individual level factors such as gender, race, disability status of the parent, single-parent households, and same–sex couples could also provide a more nuanced understanding of how experiences of conflict and stress might differ or align across a variety of identities and family settings. This cross-sectional study is limited, and implementation of longitudinal studies is warranted to better understand how resource losses and gains change across the lifespan for those giving exceptional care while maintaining employment. Responsibilities often do not get easier or less intensive as children grow older, and resources such as after-school care for high school-aged youth are likely fewer, potentially increasing work-family conflict and stress over time. The current study points to the importance of supervisor and organizational support for employed parents. Accessing flexible workplace options often requires disclosure of personal and family needs, and supervisors play a critical role in either facilitating or hindering disclosure (Rosenzweig et al., 2008). Future research should look at the disclosure processes, examining how these might differ across gender, race, and occupation. Levels of exceptional care responsibilities can vary based on type of disability, symptomology, supportive services, and family flexibility. Diary studies or time use surveys could provide nuanced understandings of the nature of exceptional care and how different levels of intensity might have differing effects on conflict and stress and disrupt work careers or add to the positive experiences of parents providing exceptional care. Finally, the effectiveness of efforts to improve communication about disability care needs, and enhance organizational, supervisor, and coworker support for employees providing exceptional care could be the focus of future experimental research that may guide employers to change practices and provide a more inclusive workplace.