Mental Health of Employed Family Caregivers in Canada: A Gender-Based Analysis on the Role of Workplace Support.

This study examines the effect of gender differences in mental health outcomes among employed family caregivers, focusing on the role of workplace support in balancing work and caregiving. Guided by the social role theory, this study analyzes nationally representative data from the 2012 Canada General Social Survey, with a sample of 2,426 participants. Women experience worse mental health outcomes than men when they require employment adjustment to fulfill their caregiving responsibilities. Workplace support could offset the negative effects of employment adjustment on mental health either directly or indirectly through family-work conflict, but gender difference is apparent in terms of the effect of workplace support. In general, women require more supportive workplace than men. Further study of the effects of various types of workplace support on the mental health among women who are employed family caregivers, and on more tailored support, is recommended.

Introduction

It is well documented that women are at more of a disadvantage in terms of caregiving than men, since women are more likely to be family caregivers, take on more caregiving tasks, and experience more negative outcomes (Lee & Tang, 2015; Li & Lee, 2020; Pinquart & Sörensen, 2003; Wakabayashi & Donato, 2006). Although an increasing number of men are taking on caregiving responsibilities, women still constitute the majority of family caregivers, and the proportion is as high as 70%–75% in countries such as the United States or Japan (Sakka et al., 2016; Sharma et al., 2016). Also, women spend more time in “traditionally female” caregiving tasks such as personal care, meal preparation, and house chores than men who are family caregivers (Fast, 2015). Evidence shows that women who provide care to family members suffer more negative outcomes from caregiving, especially higher levels of caregiving stress, more depressive symptoms, and worse quality of life (Revenson et al., 2016).

However, the surge of women in the labor force in Canada, which reached 82% in 2014 (Statistics Canada, 2018), has also brought new challenges, as women take on the roles of both family caregiver and worker. Women are still socially and culturally expected to be caregivers when their family members or friends need care due to chronic diseases or frailty. The demand for family caregiving has increased as the population ages, and in Canada, the percentage of women who are caring for family members or friends has been as high as 81% (Sharma et al., 2016). Women are estimated to be more likely to contribute more than 20 h per week toward family caregiving than men, spending an average of 5.8 years caregiving compared to 3.4 years for men (Fast et al., 2013; Vanier Institute of the Family, 2017).

The competing demands of caregiving and work affect both women's and men's employment, as previous studies have noted (Bertogg et al., 2020; Ciccarelli & Van Soest, 2018; Lee & Tang, 2015). Due to increased awareness of the challenges of balancing caregiving and work, many companies have developed workplace support for employees who are engaged in family caregiving. Supportive workplace policies help family caregivers adjust their work schedules, so they can provide care to family members and friends. However, there have been relatively few studies of the role of gender in the mental health outcomes of balancing work and caregiving in relation to work adjustment and workplace support. Therefore, this study examines the relation of gender to mental health in employed family caregivers, focusing on the role of workplace support in balancing work and caregiving.

Combining Work and Caregiving: A Social Role Theory

Social, political, and cultural changes have led to shifts in the division of labor and in gender roles in modern society (Eagly & Wood, 2012). However, even though women have taken on more active social and/or political roles, they still take more responsibility for domestic work and providing care for family members or friends (Sharma et al., 2016). The social role theory posits that gendered differences in people's behavior or personalities are affected by gender role beliefs and expectations in society (Eagly & Wood, 2012). One such belief is the gendered division of labor, the notion that women are socialized and expected to take on family roles such as homemakers and caregivers, whereas men are expected to take on occupational roles and focus on employment. Koenig and Eagly (2014) have pointed out that social role theory is useful for understanding gender stereotypes of social groups, since each familial, occupational, or other role is accompanied by certain expectations, norms, and behaviors within a particular social context. In the family caregiving context, the essence of caregiving is generally characterized as emotional support, caring, and nurturing, behaviors that are often stereotypically expected of women. For example, Miller and Kaufman (1996) have noted that male caregivers were less likely to associate caregiving with emotional work, but both men and women relate caregiving to attributes that are consistent with social role theory, including caring, emotional strength, and patience.

According to social role theory, people's behavior in a social setting is affected by the social expectations of their gender roles, and individuals may internalize gender role beliefs and behave according to social role expectations (Eagly et al., 2000). However, social role expectations might confirm gender stereotypes, because the expectations and demands of social roles may be determined more by gender than by other factors. On the one hand, even in work environments, women are expected to, and often tend to, engage in communal and/or nurturing behaviors such as providing support to colleagues, rather than assertive or task-oriented activities (Clow & Ricciardelli, 2011; McGinn & Oh, 2017). Due to this gender role stereotype, women are often perceived to lack the necessary attributes to succeed in traditionally male-dominated positions or professions. Thus, women must work harder to meet job requirements, or perform more duties than men do. On the other hand, social gender belief places different expectations on women and men at home when family caregiving is required. Women internalize these expectations and behave as though their natural role is taking on caregiving responsibilities, whereas men are reluctant to assume such responsibilities because they are not consistent with social expectations of masculine behavior (Lee et al., 2001; Saito, 2017).

Indeed, caregiving responsibilities and tasks affect employment among both women and men (Schulz, 2020). Due to gender role expectations in both the work and family domains, women require extra effort to perform roles as both employee and caregiver, and face greater pressure to balance these responsibilities. Women are more likely to experience imbalance between work and caregiving than men, and are therefore more likely to adjust work or careers to meet their caregiving needs (Lee & Tang, 2015). Combining caregiving and work is challenging, and often results in adverse social, financial, and/or mental outcomes. Women are more likely to experience negative social and personal consequences of conflict between work and family duty, role strain, and tensions over the division of care and work responsibilities than men (Converso et al., 2020; Fan et al., 2019). Women who are employed family caregivers spend less time socializing with friends, exercising, engaging in emotional and social self-care, and taking part in education and training opportunities as a result of caregiving duties, than their male counterparts (Lahaie et al., 2013). Therefore, this study proposes Hypothesis 1: Women who are employed family caregivers reported worse mental health than their male counterparts.

Workplace Support: A Dilemma?

Nowadays, many employers are aware of employees’ caregiving responsibilities for family members, and offer workplace support for family caregivers, such as flexible work schedules or alternative work locations, to help them meet both employment and caregiving needs (Bainbridge & Townsend, 2020; Shabo, 2015). In Canada, the most commonly offered support options include flexible work schedules, part-time employment, offsite work locations, and extended career breaks (Lero et al., 2012). Several studies have confirmed these types of workplace support as beneficial to family caregivers (Bainbridge & Townsend, 2020; Brown & Pitt-Catsouphes, 2016; Duncan & Pettigrew, 2012). When workplace support is available, employed family caregivers have more options to balance work and caregiving, which prevents job absences or work interruptions. More importantly, workplace support can buffer the adverse effects of caregiving on work–family role conflict, and also mitigate the negative effects of role conflict on health and well-being, such as depression or life dissatisfaction (Wayne et al., 2013).

Despite these potential benefits, employed family caregivers are reluctant to take advantage of relevant family-friendly or caregiver-friendly workplace policies due to the influence of normative views and expectations of gendered social roles (Chung & Van der Lippe, 2018). Wu (2018) has pointed out that men were less likely than women to apply for flexible working arrangements when family-related concerns arise, because men are afraid of being perceived as unmasculine or not serious about work. However, women are faced with more challenges than men in similar situations, due to social expectations, when deciding to apply for workplace support. Women who must balance work and caregiving are “in a unique double bind” (Blake-Beard et al., 2010, p. 412), because they need workplace support for family care but must balance being good employees and providing good care for their families. Since women still face the challenge of the glass ceiling or sticky floor in promotion or career development, they have more concerns than men about career advancement and use workplace support less often (Minnotte et al., 2010). In Canada, among those employed family caregivers who have access to flexible working arrangements, about 47% of women believe that applying for this benefit will negatively affect their careers (Vanier Institute of the Family, 2017). Thus, this study proposes Hypothesis 2: Women who are employed family caregivers benefit less from available workplace support when comparing to men who are employed family caregivers.

Role Conflict: A Consequence of Employment Adjustment

With formal workplace support, employed family caregivers adopt an employment adjustment strategy in order to save more time and energy to meet all their responsibilities (Bauer & Sousa-Poza, 2015). Women who are employed family caregivers are more easily influenced by their caregiving responsibilities to make adjustments in their working arrangements than men are (Lee & Tang, 2015; Smith et al., 2020). For example, an American study (National Alliance for Caregiving and AARP, 2009) found that women who are employed family caregivers are more likely than their male counterparts to take on less demanding jobs (16% of women vs. 6% of men), become unemployed (12% vs. 3%), or give up job-related benefits (7% vs. 3%) due to caregiving duty. Dentinger and Clarkberg (2002) also reported that women supporting their husbands had a much higher likelihood of choosing early retirement than noncaregiver wives, or men caring for their wives.

Employment adjustment leads to both expected and unexpected outcomes. Expected outcomes include wage loss and the potential for poverty in later life. One study reported that caregiving responsibilities incur an average of productivity loss around $10,000 per year due to work restriction, and about $5,669 in out-of-pocket expenditures due to traveling or other activities (Ganapathy et al., 2015). Hopkins et al. (2010) also estimated that employed family caregivers sacrifice about 23% of their working hours per month, which are worth about $2,887. Some evidence has also shown that adverse financial status among women in later life can result from their involvement in family caregiving when they were younger (Wakabayashi & Donato, 2006). Given the vicious circle of women becoming caregivers and experiencing lower financial status (Lee et al., 2014), women who are family caregivers are at much higher risk for long-term financial disadvantage than men who are family caregivers when they take on the responsibility of caregiving.

The unexpected consequences of employment adjustment include an increase in role conflict between work and family, particularly the family-to-work role conflict. Li & Lee (2019) have reported that employment adjustment is associated with worse mental health outcomes among family caregivers; its adverse impact on mental health results from an increased feeling of family-to-work role conflict due to employment adjustment. Another longitudinal study (Hammer et al., 2005) identified similar findings among female caregivers only; Hammer et al. (2005) suggested that in a gendered society, women are expected to be the family caregivers, and such expectations may increase if women make changes in their jobs to allow for caregiving responsibility. Sometimes women who are family caregivers report feeling depressed due to having retired early to support their spouses in need due to disability (Szinovacz & Davey, 2004).

The Current Study

With social role theory and conceptual foundation, this study examined the gender differences in the relationship between employment adjustment and mental health outcomes of employed family caregivers in Canada, focusing on the effect of workplace support. This study followed the Moderated-Mediation Model of Employment Adjustment and Mental Health (see Li & Lee, 2019 for details). Li and Lee's previous work (2019) has pointed out that when employed family caregivers make employment adjustments, they are more likely to experience worse mental health outcomes due to increased family-to-work role conflict; however, workplace support can buffer the mediating effect of family-to-work role conflict. This study further added gender as a secondary moderator in the model (Figure 1) and examined whether the moderated effect of workplace support varies by the gender of employed family caregivers.

Figure 1.

Gendered-moderated-mediation model of employment adjustment and mental health.

Methods

Data and Sample

This study was conducted using secondary data from the public use microdata file (PUMF) of the 2012 Canada General Social Survey (GSS) Cycle 26: Caregiving and Care Receiving. GSS 26 is a nationwide Canadian study conducted by Statistics Canada, with a total of 23,093 participants finishing the survey between 2011 and 2012. The target population of GSS 26 are Canadians aged 15 years and older residing in private homes in all ten provinces of Canada, and a stratified design of probability sampling at the province/census metropolitan area level was applied. Through computer-assisted telephone interviewing, GSS 26 collected information related to caregiving, care receiving, health and well-being, employment, family and social life, and demographic and socioeconomic information. For the purpose of this study, we included a sample of 2,426 participants who identified working as their main activity in the previous 12 months before they took the survey, and also had provided care to an individual (such as an aging parent, sibling, child with special needs, friend, etc.) at the same time. When the survey was conducted, the participants were still caring for the person in need.

Measurement

Mental health was represented in this study by two indicators of self-rated mental health and caregiver stress. Participants rated their mental health status, with the options including excellent, very good, good, fair and poor, on a scale of 1 to 5. This single-item measurement of mental health has been widely used with evidence of validity (Ahmad et al., 2014). With regard to caregiver stress, the participants indicated whether they found caregiving stressful, according to nine different items, including managing their own emotions, meeting the needs of the care receiver, making decisions for the care receiver, dealing with the care receiver's declining health, managing family conflict about caregiving, finding services for the care receiver, getting along with the care receiver/managing the care receiver's moods, balancing caregiving and other responsibilities, and other factors, during the past 12 months, with 1 indicating “Yes” and 0 indicating “No.” Caregiver stress, ranged from 0 to 9, is an aggregated variable summarized based on these nine items (Cronbach's Alpha  =  0.648), in which a higher number means more stressful life circumstances due to caregiving.

The key independent variables include employment adjustment, family-to-work role conflict, workplace support, and gender. Employment adjustment was measured according to five questions about employment-related actions, including “take one or more days off from your job,” “turn down a job offer or promotion,” “take a less demanding job,” “reduce regular weekly hours of employment,” or “quit a job” (ranging from 0 to 5), indicating the number of employment adjustments made by the participants. Family-to-work role conflict was measured according to a 4-point scale in answer to the question “How often has it been difficult to concentrate on work because of family responsibilities in the past 12 months?” with 1 indicating “Never” and 4 indicating “Most of the time.” Workplace support was calculated according to six questions as to whether participants’ workplaces have specific workplace policies, including flexible schedules and options to work part-time, taking leave to care for children, taking leave to care for spouse/family members, taking extended leave for personal reasons, and telework. The workplace support variable ranges from 0 to 6, indicating the amount of workplace support available. Higher number of available types of workplace support indicates that the workplace is more flexible and supportive of employee's family or personal responsibilities. Gender was grouped as men and women.

This study also controlled the sociodemographic variables of the participants and caregiving contextual factors (Pinquart & Sörensen, 2003, 2007). Sociodemographic variables include age (younger than 44 years old/45–64 years old), marital status (unmarried/married or common-law), highest education attainment (lower than high school/high school/college diploma or equivalent/university degree), country of birth (in Canada/outside of Canada), and personal annual income ($30,000 or less/between $30,001 and $60,000/$60,001 or more). Caregiving contextual factors include living arrangements between participants and their main care receivers (not living together/living together), caregiver–receiver relationship (parents/other family members/others), amount of community support available for participants, average caregiving hours per week provided by participants, and amount of caregiving tasks participants in which participants were involved.

Data Analysis

The SPSS version 26 was used to perform the data analysis. Descriptive statistics were first generated to indicate the characteristics of selected participants, illustrated in Table 1. Bivariate analysis based on gender (men and women) was carried out to compare gender differences in the study variables (Table 1). In order to examine the potential influence of gender on the moderating effect of workplace support on family-to-work role conflict between employment adjustment and mental health, a moderated-moderated-mediation analysis was conducted based on the Model 73 of SPSS macro PROCESS 3.3 (Hayes, 2018). This process was meant to assess the direct relationship between employment adjustment and mental health according to different amounts of workplace support for women and men, and the moderating effect of workplace support on the indirect effect of family-to-work role conflict between employment adjustment and mental health for both men and women simultaneously. The SPSS macro PROCESS uses a bootstrapping procedure to show the conditional effect of studied variables (Hayes, 2018), and for the purposes of this study, both gender and workplace support are the conditions. The results of the bootstrapping procedure provide the 95% confidence interval (CI) of the coefficients, and if the 95% CI does not include zero, then the result is significant. All the selected variables related to caregiver characteristics and caregiving context were controlled in the analysis. Sampling weight was applied to the descriptive analysis, and standardized weight was applied to the bivariate and multivariate analyses.

Table 1.

Demographic and Socioeconomic Characteristics of Participants (N  =  2,426, Weighted N  =  2,220,696).

Variables	All participants(N  =  2,426)	Gender		χ2(df)/t-test
		Men(n  =  990)	Women(n  =  1436)
Gender (%)		–	–	–
Men	49.11
Women	50.89
Age (%)				0.16 (1)
15 to 44	46.14	46.55	45.74
45 to 64	53.86	53.45	54.26
Marital status (%)				6.53 (1)*
Unmarried	29.15	26.74	31.47
Married	70.85	73.26	68.53
Educational attainment (%)				48.76 (3)***
Lower than high school	7.09	9.49	4.78
High school and equivalent	29.35	33.27	25.57
College diploma/certificate and equivalent	38.54	33.32	43.59
University degree and above	25.01	23.93	26.07
Country of birth (%)				0.01 (1)
Not Canada	16.29	16.23	16.36
Canada	83.71	83.77	83.64
Annual personal income				163.29 (3)***
$30,000 or less	19.08	13.26	24.7
Between $30,001 and $60,000	36.82	31.4	42.04
$60,001 and more	33.2	45.21	21.61
Not stated	10.9	10.13	11.65
Employment adjustmentMean (SD)	0.76 (0.90)	0.70 (0.85)	0.83 (0.93)	−3.55***
Workplace supportMean (SD)	2.61 (1.39)	2.63 (1.41)	2.59 (1.38)	0.65
Family-to-work role conflictMean (SD)	1.68 (0.66)	1.65 (0.68)	1.72 (0.64)	−2.48*
Community supportMean (SD)	0.29 (0.60)	0.33 (0.66)	0.24 (0.53)	3.43**
Caregiving tasksMean (SD)	6.66 (0.72)	6.68 (0.67)	6.63 (0.76)	1.58
Caregiver hours per week	9.48 (13.67)	8.52 (10.93)	10.41 (15.80)	−3.44**
Living arrangement with care receiver (%)				20.78 (1)***
Not living together	72.73	68.51	76.79
Living together	27.27	31.49	23.21
Relationship with care receiver (%)				32.66 (2)***
Parents	48	47.02	48.93
Other family members	42.04	46.17	38.08
Others (e.g., friends)	9.96	6.81	12.98
Self-rated mental healthMean (SD)	2.18 (0.94)	2.11 (0.93)	2.24 (0.95)	−3.42**
Caregiver stressMean (SD)	1.18 (1.57)	0.89 (1.20)	1.47 (1.82)	−9.29***

Note. *p < .05, **p < .01, ***p < .001.

Results

The sample of 2,426 participants included 1,436 women and 990 men, weighted to represent 2,220,696 persons, with 1,090,591 men (49%) and 1,130,105 women (51%). The majority of participants were middle-aged (45–64 years old, 54%), married (71%), well-educated with a college diploma or university degree (64%), earning a decent annual personal income between $30,000 and $60,000 (37%), and born in Canada (84%). Most of the participants were not living with their main care receivers (73%), and roughly half of them provided care to their parents (48%). Participants undertook about an average of seven different types of tasks, and about ten caregiving hours per week.

The results of gender difference based on bivariate analyses are shown in Table 1. Men and women who were employed family caregivers show significant differences in most of the variables of interest, except for age, country of birth, workplace support, and the amount of caregiving tasks in which they were engaged. The mental health status of women who were employed family caregivers was significantly worse than their male counterparts. Women reported worse self-rated mental health (t = −3.42, p < .01) and higher caregiver stress (t = −9.29, p < .001) than men. Women also contributed more weekly caregiving hours than men (t = −3.44, p < .01), made more employment adjustments (t = −3.55, p < .001), and experienced a higher level of family-to-work role conflict (t = −2.48, p< .05) than men who are employed caregivers. These results tend to support Hypothesis 1.

Table 2 shows the results of the moderated-moderated-mediation analysis based on the SPSS macro PROCESS (Hayes, 2018). The SPSS macro PROCESS assessed the moderator at each amount of workplace support on a scale from 0 to 6, and illustrated the direct effect and indirect effect according to each amount of workplace support. The direct relationship between employment adjustment and self-rated mental health was significant only for male family caregivers when workplace support is not available (b  =  0.079, 95% CI: [0.004, 0.153]). The magnitude of direct effect of the employment adjustment on caregiver stress depends on the amount of workplace support, and gender difference is also observed. For women, the direct effect decreased from 0.366 (95% CI [0.246, 0.426]) when workplace support was not available to 0.291 (95% CI [0.031, 0.549]) when four types of workplace support were present. For men, the direct effect of employment adjustment on caregiver stress was only significant when workplace support was not available (b  =  0.193, 95% CI: [0.074, 0.312]) or when only one type (b  =  0.169, 95% CI: [0.029, 0.310]) of workplace support was present.

Table 2.

Moderating Effect of Workplace Support on the Direct Effect and the Indirect Effect of Family-to-Work Role Conflict Between Employment Adjustment and Mental Health Between Men and Women.

	Direct effect (b) [95% CI]
	Self-rated mental health		Caregiver stress
Workplace support	Women	Men	Women	Men
0	0.027 [−0.030/ 0.083]	0.079 [0.004/0.153]	0.366 [0.246/0.426]	0.193 [0.074/0.312]
1	0.015 [−0.054/0.083]	0.072 [−0.016/0.160]	0.324 [0.216/0.434]	0.169 [0.029/0.310]
2	0.002 [−0.093/0.098]	0.065 [−0.059/0.190]	0.313 [0.162/0.465]	0.146 [−0.053/0.344]
3	−0.010 [−0.138/0.118]	0.059 [−0.111/0.229]	0.302 [0.099/0.505]	0.121 [−0.148/0.392]
4	−0.022 [−0.185/0.140]	0.052 [−0.167/0.271]	0.291 [0.032/0.549]	0.098 [−0.250/0.446]
5	−0.035 [−0.233/0.164]	0.045 [−0.223/0.314]	0.279 [−0.036/0.595]	0.074 [−0.354/0.502]
6	−0.047 [−0.282/0.188]	0.039 [−0.281/0.359]	0.268 [−0.106/0.642]	0.051 [−0.459/0.560]
	Indirect effect of family-to-work role conflict (b) [95% CI]
	Self-rated mental health		Caregiver stress
Workplace support	Women	Men	Women	Men
0	0.067 [0.047/0.090]	0.073 [0.047/0.102]	0.079 [0.050/0.112]	0.065 [0.034/0.102]
1	0.062 [0.040/0.088]	0.062 [0.032/0.093]	0.081 [0.047/0.119]	0.050 [0.015/0.091]
2	0.058 [0.029/0.094]	0.052 [0.010/0.095]	0.083 [0.036/0.138]	0.036 [−0.014/0.095]
3	0.054 [0.017/0.102]	0.042 [−0.014/0.099]	0.084 [0.025/0.160]	0.023 [−0.045/0.104]
4	0.050 [0.006/0.113]	0.033 [−0.042/0.104]	0.086 [0.011/0.186]	0.011 [−0.079/0.116]
5	0.047 [−0.003/0.125]	0.024 [−0.071/0.111]	0.087 [−0.003/0.213]	−0.001 [−0.119/0.126]
6	0.043 [−0.013/0.136]	0.016 [−0.103/0.118]	0.087 [−0.018/0.243]	−0.012 [−0.162/0.139]

Note. The effects in grey are nonsignificant.

Table 2 also shows the results of the moderating effect of workplace support, represented by the change of indirect effect of family-to-work role conflict between employment adjustment and mental health outcomes for both women and men. Gender difference was also identified with regard to the indirect effect of family-to-work role conflict corresponding to the amount of workplace support. The indirect effect of family-to-work role conflict between employment adjustment and self-rated mental health was statistically significant for female participants, who worked in a workplace with four or fewer types of support (b  =  0.067, 95% CI: [0.047, 0.090] for no workplace support, to b  =  0.050, 95% CI: [0.006, 0.113] for four types of workplace support). Similar results were identified for caregiver stress (b  =  0.079, 95% CI: [0.050, 0.112] for no workplace support, to b  =  0.086, 95% CI: [0.011, 0.186] for four types of workplace support). When men worked in a workplace with at least two workplace support options (caregiver stress, b  =  0.065, 95% CI: [0.034, 0.102] for no workplace support, and b  =  0.050, 95% CI: [0.015,0.091] for one type workplace support), or three (self-rated mental health, b  =  0.073, 95% CI: [0.047, 0.102] for no workplace support, to b  =  0.052, 95% CI: [0.010, 0.095] for two types of workplace support), the relationship between employment adjustment and negative mental health outcomes due to family-to-work role conflict becomes statistically insignificant. Therefore, for women who were employed by a workplace with at least five types of support, their feeling of family-to-work role conflict did not mediate the employment adjustment and negative mental health, but their male counterparts only needed a workplace with two or three types of support.

To summarize, when a certain amount of workplace support is available, workplace support is a significant moderator in the relationship between employment adjustment and worse mental health outcomes, both directly and indirectly, among employed family caregivers. Gender difference is an important factor, with women needing a more supportive workplace than men to moderate the direct effect of employment adjustment on mental health, as well as the mediating effect of family-to-work role conflict. These results tend to support Hypothesis 2.

Table 3 illustrates the effects of all the controlled variables in the gendered-moderated-mediation model. Married family caregivers were less likely to rate their mental health poorer than those not in a partnered relationship (coeff = −0.12, p < .01). When compared to family caregivers with an annual income over $60,000, those with lower income reported higher caregiver stress (coeff = 0.17, p < .05 for $30,000 or less, and coeff = 0.30, p < .01 for between $30,001 and $60,000). Caregiving hour per week was positively correlated with higher caregiver stress (coeff = 0.01, p < .01). So were caregivers living with their care receivers compared to those not living together (coeff = 0.21, p < .05). Compared to family caregivers helping their parents, those who providing care for other family members (coeff = −0.14, p < .05) or others (e.g., friends) (coeff = −0.26, p < .05) reported lower levels of caregiver stress. Family caregivers for others (e.g., friends) also reported better self-rated mental health (coeff = −0.13, p < .05) than those supporting parents.

Table 3.

Effects of the Controlled Variables on Caregiver Stress and Self-Rated Mental Health in the Gendered-Moderated-Mediation Model.

	Caregiver stress	Self-rated mental health
Age (15 to 44)
45 to 64	0.06	0.01
Marital status (Unmarried)
Married	0.05	−0.12**
Educational attainment (Lower than high school)
High school and equivalent	0.07	−0.02
College diploma/certificate and equivalent	0.12	−0.01
University degree and above	0.16	−0.16
Country of birth (Not Canada)
Canada	−0.14	0.07
Annual personal income ($60,001 and more)
$30,000 or less	0.17*	−0.08
Between $30,001 and $60,000	0.30**	−0.09
Community support	0.05	0.07*
Caregiving tasks	−0.01	−0.01
Caregiver hours per week	0.01**	0.01
Living arrangement with care receiver (Not living together)
Living together	0.21*	0.02
Relationship with care receiver (Parents)
Other family members	−0.14*	−0.08
Others (e.g., friends)	−0.26*	−0.13*

Note. *p < .05, **p < .01, ***p < .001.

Discussion

Framed by the social role theory, this study examines the influence of gender differences on the effect of workplace support in moderating the relationship between employment adjustment and mental health, both directly and indirectly, among employed family caregivers. This study has provided empirical evidence relating to the complex situation of balancing family and employment dynamics, particularly for women. The findings highlight the gender difference in the need for workplace support in managing both work and caregiving responsibilities.

A more supportive working environment is necessary for women to be able to balance work and family caregiving needs. Women who are employed family caregivers are sometimes reluctant to use workplace support options for fear of negative impressions and reactions from their co-workers and supervisors (Berkman et al., 2010). As Hegewisch and Gornick (2011) have pointed out, the possibility exists that “work-family policies could worsen some women's labor market outcomes, especially job/occupational segregation and/or the gender wage gap” (p. 130). Also, many women believe that they must work harder to avoid the perception that they are taking advantage of available workplace accommodations if they choose to work at home (Eales et al., 2015). For example, Chung and Van der Lippe (2018) identified the association between use of family-friendly flexible working arrangements and working overtime later among women. Therefore, even though they can adjust their work schedules because of workplace support, they may still need other time for their work. Moreover, given the gender gap in financial rewards when employees use workplace support (Lott & Chung, 2016), women are less likely to take advantage of available workplace support. As a previous study has suggested, workplace policies that provide flexibility to help women remain in the labor force do not have a direct effect on employees’ psychological distress due to caregiving (Pavalko & Henderson, 2006).

When women have flexible work schedules, they are expected to take more family responsibilities (Hilbrecht et al., 2013; Power, 2020). It is also possible that when workplace support to accommodate family caregiving is available, women are even more likely to become family caregivers. This is because workplace support may increase family members’ expectations that these women will take the majority of the caregiving responsibilities (Hammer et al., 2005). This also resonates with the premises of gendered stereotypes, under which women are expected to be carers or nurturers in a gendered society even if they are in the labor force. In such a situation, women who are employed family caregivers may need more support from the workplace to accommodate family caregiving responsibilities; otherwise, they may experience greater family–work role conflicts.

The results show that women who are employed family caregivers are more likely to rate their mental health as worse and to experience greater caregiver stress than men who are employed family caregivers. Given that women are already in more vulnerable positions with regard to employment and financial adequacy in relation to family caregiving, adverse mental health outcomes following employment adjustment are another challenge that women might have to face. This finding further emphasizes the salience of studying employment adjustment among employed family caregivers. In addition to the widely recognized negative financial consequences (Hopkins et al., 2010), adverse mental health conditions related to employment adjustment should receive more attention in family caregiving study.

The results of this study show that women who are employed family caregivers undertook higher-intensity caregiving, made more employment adjustments, and reported higher family-to-work role conflicts, but received less community or informal support. These findings are consistent with previous studies (Pinquart & Sörensen, 2003; Rodríguez-Madrid et al., 2019; Schwarzer & Gutiérrez-Doña, 2005), and further highlight the imbalance in caregiving, with women bearing more of the burden than men. This resonates with the tenets of social role theory that women are expected to take more responsibility in the family domain by becoming caregivers (Eagly et al., 2000). Under such an expectation, it is taken for granted that women take care of others in need, and there is no need to provide support to help them in the family role. Therefore, women who are family caregivers usually need to be “wonder women” to do it all (Eales et al., 2015).

This study also supports the significant association between several demographic and socio-economic factors and mental health outcomes among employed family caregivers. Married family caregivers tend to rate their mental health better than those not. This finding is consistent with previous studies (e.g., Penning & Wu, 2016). The support from a spouse or partner can help family caregivers with care responsibilities, and better handle the caregiving stress or burden. The identified association between low-income and greater caregiver stress also resonates with existing literature (Ferrara et al., 2008; Penning & Wu, 2016). A higher level of socio-economic status is closely related to the resources and support family caregivers can get, such as paid home services, to help them manage caregiving tasks. In this study, considering that higher proportions of women are under the lower-income groups than men, and women make more employment adjustments to accommodate caregiving, this finding highlights the disadvantaged position of women who need to combine care and employment.

Also, caregiving hour is positively related to caregiver stress. The time spent on caregiving tasks is an essential indicator of caregiving intensity, which is related to caregiver burden and stress. This study, along with others suggesting the impact of higher caregiving intensity on reducing labor force participation (Lilly et al., 2010), further calls for programs or services to support women with higher levels of care responsibility. Living with the care receiver is associated with higher caregiver stress. Existing evidence suggest that when living with the care receivers, family caregivers tend to have limited choice in undertaking the caregiving role, performing more intensive care, and providing more energy- and time-consuming personal care (Li & Lee, 2019; Thrush & Hyder, 2014). In addition, family caregivers providing care to aging parents tend to experience worse mental health than those supporting other family members or non-kin others. The association between the caregiver–receiver relation and mental health outcomes has been addressed in detail in previous studies (Li & Lee, 2019; Penning and Wu, 2016). The participants in this study are mainly middle-aged individuals, and their care receivers are usually aging parents at this life-course stage (Silverstein et al., 2006). Given their multiple and competing responsibilities from caring for aging parents, career development, and parenting young children, it is understandable that adult children providing care to aging parents report considerable caregiver stress and poor perceived mental health.

Implications

This study supports that gender difference persists among employed family caregivers, since women take more caregiving responsibilities, experience more burdens, and show worse mental health outcomes as a consequence of family caregiving. However, studies of the mental health consequences of employment adjustment are still sparse, and only certain aspects of these outcomes, such as financial loss and role conflict, have been investigated to date (Li & Lee, 2019). Future studies should explore the effects of employment adjustment on family caregivers in more detail. Other studies have examined the effects of various workplace support options, such as individual and organizational work-life balance strategies (Zheng et al., 2015), organizational support and support from supervisors or colleagues (Yucel & Minnotte, 2017), and family-friendly workplace policies (Vuksan et al., 2012). However, relatively few studies have examined the role of gender difference in the use of workplace support or potential benefits (Minnotte et al., 2010). Based on the gender difference with regard to the function of workplace support identified in this study, future research on gendered experience regarding the use of workplace support and the perception thereof is encouraged.

In addition, the findings of this study provide empirical evidence for policymakers to take gender stereotypes into consideration when developing relevant family-oriented or workplace-based programs to support employed family caregivers. It is also particularly relevant to the current situation due to the COVID-19 pandemic. The provision of family caregiving is affected to a greater degree by the workplace support policy, community service delivery, and family and societal expectations. It is unfair to expect women to take the majority of caregiving responsibilities when working from home is commonplace. Therefore, more programs and interventions should be provided for women who have both working and caregiving responsibilities. Also, social education programs are needed to emphasize the importance of sharing caregiving responsibility between men and women, and to promote more positive images of caregiving among men.

This study also has its limitations. Due to the nature of the GSS 26, only the participants who identified their main activity as working in the past 12 months before the survey were included in the data analysis. This data selection process might have excluded family caregivers who stopped working in order to devote themselves full-time to caregiving. Therefore, the scope of generalization of the findings is not adequate for covering family caregivers who do not work, or for whom work is not the main activity. Also, the GSS 26 PUMF does not provide participants’ job-related information, including types of job, job position, or job security, which are closely related to family caregivers’ capability in combining care and employment. For instance, Hill and colleagues (2008) pointed out that low job security is the main reason for leaving employment among family caregivers. Fredriksen and Scharlach (1997) also found that among university employees, family caregivers in staff positions provided more care, but experienced lower flexibility, control or support at the workplace than academics and administrators. Therefore, further study is needed to comprehensively examine the effects of job characteristics on the mental health outcomes among employed family caregivers.

Another potential limitation is the measure of workplace support, which is calculated as the amount of workplace support for caregivers caring for aging parents, the child with special needs, other family members and friends, and so on. Thus, some sort of support might be needed or not be needed at all for some caregivers depending on their caregiving situation. For instance, family caregivers who care for aging parents but do not have any child to care for might not need to use the workplace support related to taking leave to care for children. However, the amount of available types of workplace support would still indicate the level of perceived supportiveness of the workplace. Ideally, it would be more sophisticated to analyze both perceived and utilization of workplace support, but the GSS 26 does not collect information related to the utilization of each type of support. In addition, because the GSS 26 is a cross-sectional survey, the results of this study must be interpreted carefully with regard to any causal relationship. Future studies incorporating longitudinal data are needed to explore in greater detail the effects of employment adjustment on mental health among family caregivers while considering the influence of workplace support.

Conclusion

Traditional gender role beliefs and social norms lead people to expect that women will assume the major responsibility in caring for family members, even when they are active in the workforce. The findings of this study reveal that, even with workplace adjustment and support, gender differences in mental health outcomes and work–family dynamics continue to exist. The findings suggest the desirability of gender equality in terms of role beliefs and expectations in family life and employment, and greater opportunities for both women and men to balance work and family lives.