This study examines the influence of work characteristics—conceptualized as demands and resources—on the preventive health behaviors and subjective health outcomes of married parents with preschool age children. Although personal health is often thought to be the result of conscious choices made by individuals, organizational characteristics play an important role in determining the context within which these choices are made (Lundstrom et al. 2002). Indeed, the workplace organization is the site in which many adults spend the greatest portion of their waking hours; more hours than ever before (Bianchi et al. 2006; Jacobs and Gerson 2004). The traditional male provider, female homemaker model exists in only 19 % of married couple households in the US, the two-earner couple now being the modal household type (US Bureau of Labor Statistics 2013). Though parenthood significantly shapes women’s participation in the workforce (Gibb et al. 2014), over 60 % of mothers are employed by the time their child is 3 years old and the majority of them work full-time (US Bureau of Labor Statistics 2012). In short, most married parents are working parents and this is no different for those with preschool age children.

Understanding the link between work and health for employed parents with young children is particularly important because of the many demands they face. Indeed, most working parents describe their days as busy and their energies as spread thin (Hochschild 1997). Caring for young children is time consuming and stressful (Hetherington and Parke 1999), with parenting norms emphasizing devotion and intensive caretaking (Hays 1996). In light of busy schedules, personal health often takes a back seat to more pressing responsibilities (Nomaguchi and Bianchi 2004), especially in a culture that prioritizes work and family at the expense of personal leisure (Hochschild 1997). With multiple, often competing obligations, working parents are more likely to identify themselves as experiencing conflict between work and family (Haas 1999). Although those who are partnered tend to have better health and use preventive health care more frequently (Kohn and Averett 2014; Miller and Pylypchuk 2014), compared to non-parents, those with children are less physically active, with mothers getting less exercise than fathers (Bellows-Riecken and Rhodes 2008). Mothers also have poorer dietary intake and higher BMI than women without children (Berge et al. 2011). With little time for exercise or sleep (Berge et al. 2011; Sayer et al. 2004), work characteristics are likely to influence whether a mother or father has a healthy diet, goes for a walk, or finds time for relaxation. Furthermore, past research has shown that paid work is often more strongly associated with personal health than domestic responsibilities (Hunt and Annandale 1993), with authors stressing the importance of workplace measures in considerations of health (Lutz 1989).

What follows is a discussion of the job demands and resources approach, or JD-R, an overarching theoretical perspective that cuts across occupational settings to frame the work experiences of individuals and their potential impacts on non-work life (Demerouti and Bakker 2011). On the basis of this model, five job demands and resources (work hours, work pressure, face-time norms, job flexibility, and coworker support) were considered in models examining preventive health behaviors and subjective health outcomes (measured as days per week of adequate sleep, exercise, relaxation, and healthy diet; days per week of worry and stress, feeling overwhelmed, and feeling healthy and full of energy). Data were taken from a sample of married parents with preschool-age children, all employed and living in a Midwestern city (N = 144). In the sections to come, the literature for each of the job demands and resources included in the models is reviewed, and then the current study is described.

Literature Review

As the location in which most adults spend the majority of their waking hours (Bianchi et al. 2006; Jacobs and Gerson 2004), workplace environments are increasingly recognized as sites of influence that can promote or hinder personal health (Kelloway and Day 2005). Noted by Perry-Jenkins et al. (2011, p. 1118), “job conditions have the potential to be both a source of stress as well as a source of support and empowerment.” Employees are routinely exposed to stressors as occupants of work roles, but they also gain access to resources such as social support and income (Roxburgh 1996). This enrichment model of work recognizes that resources gained at work can improve the quality of life of employees both on the job and off. Participation in work brings not only hassles, but also rewards beyond income including self-esteem, empowerment, social support, and satisfaction (Greenhaus and Powell 2006; Grzywacz and Bass 2003; Perry-Jenkins et al. 2011; Sorensen et al. 1985). Together, work demands and resources influence individual employees for better or worse, behaviorally, physically, and mentally. This is the central premise of the job demands and resources approach or JD-R (Bakker et al. 2007; Fredriksen-Goldsen and Scharlach 2001; Voydanoff 2004).

The first type of work characteristics, demands, are physical, social, psychological, and organizational aspects of the job that require a sustained effort or skills from the employee, or in other words, result in the use of resources (Carlson et al. 2011; Demerouti et al. 2001). Examples of job demands include heavy workloads, irregular work hours, difficult clients, and negative or stressful work environments. Though they are not necessarily negative in nature, job demands can result in stress, strain or psychological exhaustion when they exceed an employee’s ability to cope (Bakker et al. 2007). When exposure to demands is prolonged and unrelieved, employee health suffers (Karasek and Theorell 1990; Roxburgh 1996). For instance, employees with heavy job demands are more likely to experience distress and depression (Bromet et al. 1992).

Job resources are also physical, social, psychological, or organizational in nature. But in contrast to job demands, resources facilitate performance or achievement of work goals, reduce demands, or enhance personal growth and resources (Bakker et al. 2007; Voydanoff 2004). Thus, as noted by Bakker et al. (2007), they are important both in their own right, as a means to achieve other resources, and for successful responsiveness to job demands. Examples of job resources include job flexibility, support from colleagues, job satisfaction, income, career opportunities, clarity of expectations, and the like. Many resources are fixed to some degree, such that when consumed within one domain, they are less likely to be available for use in another domain. For instance, when work hours are long, there is less time available outside of working hours for personal leisure or relaxation—especially if limited time is devoted to interacting with children (Bianchi et al. 2006). Job demands may preclude the utilization of job resources, whereas resources can reduce job demands (Bakker et al. 2007).

Central to the current study, the JD-R model highlights that when employees have inadequate resource access and/or are engaged in work that is very demanding, they will be at greater risk for physical and psychological exhaustion and motivational loss (Demerouti and Bakker 2011). Thus, employees who have few resources or whose job resources are outstripped by heavy job demands can be expected to have a lower level of participation in preventive health behaviors and experience a reduction in subjective well-being. Rather than treating all job demands or resources as interchangeable, however, it is likely that specific job characteristics operate differently, either alone, or in tandem with other demands and resources. For instance, long work hours can negatively impact health, whereas long work hours combined with job flexibility or high levels of coworker support may not be as harmful. In the current analysis, five job demands and resources (work hours, work pressure, face-time norms, job flexibility, and coworker support) are considered in models regressing preventive health behaviors and subjective perceptions of health (days per week of adequate sleep, exercise, relaxation, and healthy diet; days per week of worry and stress, feeling overwhelmed, and feeling healthy and full of energy). Existing literature for each of the five job characteristics and employee health is considered in turn.

Job Demands

Work Hours

Previous work documents the relationship between long work hours and poor health, using both objective and subjective measures (Caruso 2006; Sparks et al. 1997; Taris et al. 2011; Van der Hulst 2003). The impact of long work hours on health has been theorized to operate through two different mechanisms. First, long hours limit time and motivation for recovery, leading to adverse health consequences. Second, long hours lead to behavioral lifestyle decisions resulting in poor health, such as reduced physical activity and poor diet (Taris et al. 2011). Working long hours can result in feelings of stress or role conflict, reducing one’s motivation to exercise, especially if these hours are seen as taking away from family time (Grzywacz and Marks 2001). Long work hours are associated with poorer and shorter sleep (Åkerstedt et al. 2007; Dahlgren et al. 2006; Lallukka et al. 2010; Metlaine et al. 2005), less time for and participation in physical activities (Mattingly and Bianchi 2003; Nock and Kingston 1989), reduced intake of fruits and vegetables, and lower levels of subjective health (Taris et al. 2011). Given the heavy work–family obligations of employed parents with young children, longer work hours are expected to be negatively associated with participation in preventive health behaviors and subjective health outcomes.

Work Pressure

In addition to experiencing the burden of long work hours, time spent at work is sometimes marked by intense productivity pressure. The experience of work pressure occurs when employees feel rushed, short of time, or have a frequent sense of urgency about work tasks and deadlines (Roxburgh 2004); in short, they have what is sometimes described as a “time crunch” (Robinson and Godbey 1997). Work pressure is more commonly experienced by women employees among the well-educated, and by men and women equally among workers with average or below average educational levels. To effectively navigate workplace settings marked by high levels of work pressure, employees must have the skills to simultaneously coordinate multiple demands and obligations (Roxburgh 2004). Yet, the pressure to meet frequent deadlines combined with demanding time constraints indicates a discrepancy between the demands of work and employee resources (House 1974). Thus, work pressure limits an employee’s perception of control over work activities, thereby increasing feelings of stress (Sorensen et al. 1985). As such, the workload and stress associated with jobs with high levels of work pressure diminishes participation in physical activity and produces sleepless nights (Eriksen et al. 2008; Lallukka et al. 2010; Roth and Roehrs 2003; Sorensen et al. 1985). Other studies link heavy workloads and time pressure to mental strain and cardiovascular disease, particularly when social support is lacking (Karasek and Theorell 1990), depression (Perry-Jenkins et al. 2011; Roxburgh 2004), and premature death (Sorensen et al. 1985). Research examining specific occupational groups has shown that work pressure is related to heart disease among middle-aged blue-collar working men (Siegrist et al. 1990), and emotional exhaustion among nurses (Robinson et al. 1991).

Face Time Norms

“Face time” is a term used in the popular business press, referring to the amount of time an employee is seen at work or in the office, whether interacting with others or not, and whether being productive or not (Elsbach et al. 2010). In essence, it is simple validation or “getting credit” for being physically present at work. For instance, being seen sitting at one’s desk by another employee who walks by is evidence of face time, regardless of whether the individual is actually engaged in the work process. Face time norms are powerful because of their connection to how employees are perceived and evaluated by others. Face time is frequently used as a criterion for making inferences about workers, such as whether an employee is dedicated, takes initiative, or has leadership ability (Brandel 2002; Gopinath 2003). The most devoted employees are often believed to be those who arrive first, and leave last (Bailyn 1993; Fried 1998; Kanter 1977). New employees who want to be seen as hard workers must often put in extensive face time (Ryan 2005). As well, face time is important among professional workers who want to be viewed by coworkers and supervisors as responsible and committed (Elsbach et al. 2010; Kossek and Van Dyne 2008). Among professional workers whose output is often difficult to quantify, face time has been used to judge employee contributions and outcomes (Kossek and Van Dyne 2008; Maume and Houston 2001). The power of face time norms is evidenced by limited career progress among remote site employees (Duxbury 1999; Joyce 2002) and negative evaluations of those who do not put in weekend overtime (Perlow 1997)—regardless of how productive those employees are (Krzystofiak et al. 1988). Further, strong face time norms result in a culture of overwork marked by employees who routinely put in 70–80 h workweeks to show commitment (Fried 1998). Sennett (1998) aptly referred to the norms and rules surrounding work time as “time cages,” constricting employees’ behavior both during and outside of working hours. As such, the norms for visibility and face time may bleed into nonwork hours to shape health behaviors and perceptions of well-being. Indeed, scholars have suggested that a change in the culture of time at work is needed to enhance employee health (Moen et al. 2011a, b), although few studies empirically examine this question.

Job Resources

Job Flexibility

As noted, the JD-R model allows that job demands generally occur in tandem with job resources to exert influence on non-work aspects of employee life. There are two job resources considered in the current study. The first of these is job flexibility, a form of control over work and work time that helps reduce perceptions of overload among employees (Nijp et al. 2012). By allowing employees to take a break when they feel they need to recover, have influence in selecting leave days, and/or control starting and ending times of work, job flexibility serves as a buffer against fatigue and stress and promotes employee well-being (Donovan and Halpern 2002; Nijp et al. 2012). This is especially true for women and employees with heavier family responsibilities (Jang et al. 2012). Flexible scheduling can benefit employee health by reducing conflict between work and non-work demands (Thomas and Ganster 1995), and by providing feelings of control and empowerment (Hall and Atkinson 2006). Job flexibility has been empirically linked to both physical and mental health (Frone et al. 1996, 1992; Glass and Finley 2002; Jang 2009; Jang et al. 2011; Thomas and Ganster 1995). In one study, a truly flexible work schedule and control over scheduling resulted in an additional hour of sleep on work nights and increases in exercise (Moen et al. 2011a, b). Participants engaging in a “results-only work environment” that prioritizes outcomes over time clocks and face time, reported less emotional exhaustion and psychological distress, and more energy. Meta-analytic studies and reviews also indicate that flexible work arrangements are beneficial to employee health (Nijp et al. 2012). Among nursing staff, the relationship between job flexibility and improved health has been documented (Lea and Bloodworth 2003).

Coworker Support

The second job resource considered in this study is coworker support. Supportive relationships are an important factor shaping health behaviors and outcomes, making up part of the psychosocial environment of the workplace (Hammer et al. 2004). Coworkers provide a wide range of supportive functions, including those that are practical, informational, emotional, and evaluative in nature (Sorensen et al. 1998). Coworkers can provide a buffer against poor working conditions or difficult managers (Hodson 2001), and help prevent feelings of stress and emotional exhaustion (Jenkins and Elliott 2004; Thoits 1995; Thompson et al. 2005). Coworker support has been linked to motivation to make healthy choices, including readiness to make dietary changes (Sorensen et al. 1998), as well as participation in preventive health behaviors including fruit and vegetable consumption, exercise, and cervical cancer screening (Kelsey et al. 2000). A supportive work environment has been linked to perceptions of well-being among employed adults (Heaney et al. 1995; Hochschild 1997; Treiber and Davis 2012), and for members of specific subgroups including employed mothers with preschool-aged children (Greenberger et al. 1989), and those in particular occupations, including nurses (Bradley and Cartwright 2002; Treiber and Davis 2012) and civil servants (Griffin et al. 2007). Indeed, the greatest declines in health status have been found among those with the lowest level of coworker support (Cheng et al. 2000). Past research indicates that those who are socially isolated are less likely to engage in preventive health behaviors, including exercise and eating a healthy diet; they are more likely to smoke, drink heavily, and be overweight (Berkman and Syme 1979). Thus, for those who are employed, coworker support often plays a significant role in health behaviors and perceptions of well-being (Baker et al. 1996; Daltroy et al. 1993; Johnson and Hall 1988).

Additional Factors

In addition to considering the five job demands and resources outlined above, this study will consider the role of gender. Given differences between men and women in parenting expectations and employment obligations, it is expected that the link between work characteristics and health will be experienced differently by mothers and fathers. Since 1965, married women’s leisure time has declined (Bianchi et al. 2006). In turn, time spent with children and at work increased, resulting in more married women than men reporting that they had “too little time for oneself” (Bianchi et al. 2006, p. 136). Because women have greater responsibility for household tasks and child care, they are more likely than men to scale back “discretionary” activities such as sleep, leisure, and exercise, especially in response to work demands (Bianchi et al. 2006; Bittman and Wajcman 2000; Shaw 1985). Hislop and Arber’s (2003, 2006) work illustrated that the role definitions and priorities set by women and men, that is, how they define work, family, and time, results in different sleep patterns. Whereas men in their study believed that sleep is necessary to recuperate from work and a discrete use of time that is the right of a family breadwinner, women viewed the nighttime hours as continuous with the day and its many obligations. Likewise, other researchers have also found that women’s daytime obligations are more likely to extend into hours that would otherwise be devoted to sleep (Maume et al. 2009). Furthermore, it is not only the quantity of free time but also the quality of free time that varies by gender. Men have access to greater periods of uninterrupted free time, whereas women’s leisure is more frequently interrupted and combined with other tasks, such as childcare and household chores (Mattingly and Bianchi 2003), making it more difficult for them to relax or participate in health behaviors. Compared to men, women spend about twice as much of their “free time” supervising and attending to children and their activities (Bianchi et al. 2006; Lareau 2003). Mattingly and Sayer (2006) argued that the time pressure faced by women limits their ability to recuperate from stressors that negatively affect health. In addition to women’s heavy workload as household managers, women are often relegated to positions in the workforce that offer lower pay and less room for advancement than those held by men (Valian 1999). For these reasons, gender is likely to be a central factor shaping the relationship between job demands and resources and health outcomes, especially among those with young children. Because the data taken from this sample are non-independent—some of the mothers and fathers were married to each other—separate regression equations are estimated for women and men. Thus, gender cannot be examined for its significance as an independent variable, but the different experiences of mothers and fathers can be considered.

Education and occupational status are included as control variables in the analyses to follow. The relationship between social class and health is longstanding. Strong evidence links educational attainment and professional occupational status to both preventive health behaviors and better subjective health outcomes (Adler and Newman 2002; Lynch et al. 1997; Ross and Wu 1995; Salmon et al. 2000).

Method

Sample and Data Collection

To examine the relationship between work characteristics and preventive health behaviors, data were taken from a sample of parents whose preschool-age children were enrolled in licensed child care centers (day care or preschool). Three providers were randomly selected from among eight licensed daycare centers located within a mid-sized Midwestern city (population of approximately 50,000). Home-based centers were excluded from the sampling frame. Questionnaires were distributed to parents at each of the three centers through their child’s mailbox or cubby. Two questionnaires were distributed to each of the approximately 325 enrolled children in the event that the child lived with or had contact with two parents. Surveys were prefaced by a letter of introduction, and a children’s storybook was provided as compensation for parents’ time. Parents returned completed surveys—placed in sealed envelopes—in a drop box located near the entrance of their child’s center.

A total of 206 questionnaires were returned. Because some families enrolled more than one child in the center and the family structure of each child was unknown, it is hard to determine a precise response rate. For instance, if a married couple enrolled two children in a participating center, they would have received four surveys, at least two of which would have been discarded or handed in blank. Single/divorced (10.8 %), cohabiting (3.4 %), and married (85.8 %) parents all participated in the study. Among married and cohabiting respondents (N = 184), full data were available from both partners for 65 couples (70.7 % of married and cohabiting participants). For the current analysis the sample was limited to married parents who were working, a total of 66 fathers and 78 mothers (N = 144), some of whom were married to each other. Representative of the community from which the sample was drawn, the majority of the sample (89 %) was white. The average age of fathers was 36.56 (SD = 5.10), and mothers were 35.71 years old (SD = 4.80). Parents had about two children on average. The majority of parents (68 % of fathers, 73 % of mothers) had a bachelor’s degree or higher. Fifty-five percent of fathers were working in jobs classified as professional/managerial, whereas 40 % of mothers had positions that were professional/managerial. Median family income was approximately $85,000 per year.

Measures

Dependent Variables

Seven items were used to measure preventive health behaviors and subjective health outcomes. These items were adapted from the Center for Disease Control’s Behavioral Risk Factor Surveillance System (BRFSS), an annual survey distributed to adults residing in all US states and territories. It is currently the largest continuously conducted survey system of health in the world (CDC 2014). Whereas the BRFSS asks respondents to report frequencies for the previous month, in the current survey respondents were asked, “Thinking about a typical week, on how many days do you: (a) Get at least 7 or more hours of sleep; (b) Exercise for 30 min or more; (c) Have time to yourself to relax; (d) Eat a healthy, balanced diet; (e) Feel worried or stressed; (f) Feel overwhelmed by all the things you have to do; and (g) Feel healthy and full of energy.” Respondents indicated the number of days per week, ranging from 0 to 7, for each statement.

Independent Variables

Three job demands were measured. First, work hours per week was operationalized using a question that asked respondents to write in the number of hours they typically worked in a given week. Work pressure was measured using a subscale of the Work Environment Scale created by Moos and Insel (1974). The Work Environment Scale has been used extensively since its debut four decades ago in clinical, research, and occupational contexts. The instrument has been used both nationally and internationally, showing validity across occupational settings including among them the military, government, health care, and education (Moos 1994). Eight items made up the index of work pressure: “(a) There is constant pressure to keep working; (b) There always seems to be an urgency about everything; (c) People cannot afford to relax; (d) It is very hard to keep up with your work load; (e) People often have to work overtime to get their work done; (f) There are always deadlines to be met; (g) You can take it easy and still get your work done (reverse coded); and (h) There is no time pressure (reverse coded).” Responses for the items included in the work characteristics indices ranged from strongly disagree to strongly agree along a five point scale, with higher scores indicating greater agreement. To create the indices, the items were summed and then averaged, in order to conform to the original coding and for ease of interpretation. An alpha reliability coefficient of .86 was obtained for the work pressure index. The index measuring face time norms included four items (Moos and Insel 1974): “(a) I’d be afraid to take extra time off to spend with my family; (b) Employees are afraid to take time off because they might be perceived as lacking commitment to the job; (c) You have to put in a lot of “face time” to be appreciated; and (d) If you aren’t visible at work, your commitment is questioned.” The index had a high level of reliability (α = .73).

Two measures of job resources were considered. First, an eight item index was used to measure job flexibility. This measure was created and piloted by researchers from a cooperative regional research project focusing on work and family linkages (Western Regional Project W-167 2002) funded by the US Department of Agriculture. The items making up the scale were generated using information from business and policy sources (Ferber et al. 1991; “Is Your Company” 1990; Nelton 1989), and the scale was validated across participating research sites with samples of working adults. Respondents were asked, “On days when you are working, how easy or difficult is it for you to arrange time to do the following: (a) To go to work later than usual if needed; (b) To go to health care appointments; (c) To run errands; (d) To go shopping; (e) To take care of household chores; (f) To participate in community or church activities; (g) To have some meals together as a family; and (h) To have some time to myself.” The alpha reliability coefficient for the job flexibility index was .86. Coworker support was measured using a subscale from the Work Environment Scale (Moos and Insel 1974). These seven items were summed and then averaged: “(a) People go out of their way to help other employees feel comfortable; (b) Employees take a personal interest in each other; (c) People are usually honest and open about how they feel; (d) We often eat lunch together or chat in the break area; (e) Employees often talk to each other about their personal problems; (f) The atmosphere is somewhat impersonal (reverse coded); and (g) People often talk about each other behind their backs (reverse coded).” The alpha reliability coefficient for the index was .81.

Control variables in the analysis included education and occupational status. Education was measured ordinally along an eight point scale (1 = some high school to 8 = professional degree or doctorate). A dummy variable was created for occupational status (1 = managerial/professional).

Analytic Technique

To examine the combined influence of job demands and resources on preventive health behaviors and subjective health outcomes, separate OLS regression models were estimated for mothers and fathers, with each health outcome as a criterion variable. In the regression models, independent and control variables were entered at once to simultaneously consider job resources and demands while holding education and occupational status constant. Before doing so, t tests were generated to examine whether differences in job demands and resources varied on the basis of gender or occupational status; t tests were also used to examine differences in preventive health behaviors and subjective health outcomes on the basis of gender and occupational status.

Results

Descriptive Statistics

Descriptive statistics for the variables included in the analysis are reported in Table 1. The average number of work hours per week for mothers was 34.04 (SD = 11.19), whereas fathers worked significantly more hours—more than full time, on average (Mean = 46.40, SD = 9.27) (t = 7.07, df = 142, p < .000). Mothers (Mean = 2.56, SD = .63) and fathers (Mean = 2.58, SD = .64) indicated that they had a fair amount of job flexibility (on a scale from 1 to 5) and face time norms were fairly true of their workplaces (Mean = 2.58, SD = .71 for mothers; Mean = 2.75, SD = .77 for fathers). On the same scale, they reported higher average levels of work pressure (Mean = 3.23, SD = .78 for mothers; Mean = 3.42, SD = .65 for fathers). Mothers (Mean = 3.48, SD = .71) reported significantly higher levels of coworker support than fathers (Mean = 3.23, SD = .73) (t = 2.18, df = 142, p < .031).

Table 1 Descriptive statistics (N = 78 mothers, 66 fathers)
Full size table

With regard to the health variables, during a typical week, respondents reported a higher average number of days—about four—for which they ate a healthy diet (Mean = 4.37, SD = 1.83 for mothers; Mean = 3.85, SD = 1.96 for fathers) and got adequate sleep (Mean = 4.10, SD = 2.17 for mothers; Mean = 3.92, SD = 1.99 for fathers). The average number of days per week for which respondents reported 30 min or more of exercise (Mean = 2.44, SD = 1.96 for mothers; Mean = 2.09, SD = 1.73 for fathers) and time to relax (Mean = 2.04, SD = 1.61; Mean = 2.41, SD = 1.62 for fathers) was just over two, although variation within the sample was high. Averages for subjective health outcomes were relatively similar across the three variables: respondents reported an average of three to about three and a half days per week for which they felt overwhelmed (Mean = 3.69, SD = 2.00 for mothers; Mean = 3.20, SD = 1.85 for fathers), and around three and a half days per week for which they felt worried or stressed (Mean = 3.72, SD = 1.66 for mothers; Mean = 3.48, SD = 1.92 for fathers) but also healthy and full of energy (Mean = 3.63, SD = 1.70 for mothers; Mean = 3.65, SD = 1.74 for fathers). T tests revealed no significant differences in the measures of preventive health behaviors or subjective health outcomes for mothers and fathers. T tests were also generated by occupational status, with significant differences in the means found for work hours (t = 3.40, df = 142, p < .001), work pressure (t = 3.20, df = 142, p < .002), job flexibility (t = 2.01, df = 142, p < .046), and days per week of adequate sleep (t = 2.21, df = 142, p < .029). Professional workers had higher means for each of these variables with the exception of job flexibility. In this case, nonprofessional/nonmanagerial workers reported higher average flexibility of work time.

Regression Models

Results for the regression analyses are presented in Table 2 for mothers and Table 3 for fathers. Separate models were estimated for each preventive health behavior and rating of subjective health. All seven models reached statistical significance for mothers. For fathers, four of the seven regression models reached statistical significance, the exceptions being days per week of adequate sleep, time to relax, and eating a healthy diet—which nearly reached statistical significance (p < .105). In the regression models the variance inflation factors were low (the highest for mothers being 1.438, and for fathers, 1.743). Tolerance statistics ranged from .70 to .83 for mothers and from .57 to .85 for fathers. None of these results suggested that multicollinearity was of concern (Field 2005; Licht 1997). Some combination of job demands and resources was influential in most of the significant regression models, with the exception of days per week eating a healthy diet for mothers and fathers. In these cases, neither job demands nor resources were significant for fathers and only job resources were significant for mothers. Furthermore, among mothers, face time norms were not significantly associated with any of the preventive health behaviors or subjective health outcomes.

Table 2 Statistically significant findings for models regressing preventive health behaviors and subjective health on work characteristics for mothers (N = 78)
Full size table
Table 3 Statistically significant findings for models regressing preventive health behaviors and subjective health on work characteristics for fathers (N = 66)
Full size table

Preventive Health Behaviors

Examining these models in turn, preventive health behaviors will first be considered. Looking at the model regressing days per week of adequate sleep for mothers, in addition to the positive association with occupational status (β = .32*), indicating that the professional women workers in the sample reported more days of adequate sleep, a statistically significant negative association with work pressure was also found (β = −.25†). That is, higher levels of work pressure were reported alongside fewer days of adequate sleep. Thus, only job demands were associated with days of adequate sleep for mothers. (As noted, this model did not reach statistical significance for the fathers in the sample.) In contrast, for days per week of adequate exercise, job resources and demands were found to work in tandem for mothers and fathers. Days of adequate exercise was negatively associated with work hours (β = −.31* for mothers and −.40** for fathers), as expected, and positively associated with job flexibility (β = .28* for mothers and .25† for fathers). Those who worked more hours reported fewer days of adequate exercise, but this was tempered by job flexibility.

The third regression model examined number of days respondents had time to relax, and the same combination of job demands and resources that was associated with adequate exercise was significantly associated with time to relax among mothers. Work hours was negatively associated with time to relax (β = −.38**), whereas job flexibility was positively associated with the outcome variable (β = .25*). This model did not reach statistical significance for fathers. The final model examining preventive health behaviors regressed days per week respondents ate a healthy diet on job demands and resources. For mothers, job flexibility was again significant (β = −.31*). Additionally, a positive association between education and eating a healthy diet existed for mothers (β = .31*) and fathers (β = .43**). Job demands and resources were not statistically significant in this model for fathers. Overall, the predicted variance in the models regressing preventive health behaviors was greater among mothers, although the model for days of adequate exercise was stronger for fathers.

Subjective Health Outcomes

Models examining subjective health outcomes indicated the importance again of the same combination of job demands and resources for mothers: work hours and job flexibility were statistically significant across all three models. In contrast to findings for mothers, coworker support emerged as significant in all three models for the subjective health outcomes of fathers. In the first subjective health model, in addition to work hours (β = .27*) and job flexibility (β = −.21†), the number of days that mothers felt worried or stressed was positively associated with work pressure (β = .28*) and negatively associated with education (β = −.21†). Mothers who had higher levels of education and job flexibility reported fewer days per week of feeling worried or stressed, whereas those with greater work hours and work pressure reported more days of worry and stress. Fathers reported more days of worry and stress when they had lower levels of job flexibility (β = −.22†) and coworker support (β = −.34**), and greater work pressure (β = .35**).

The second subjective health outcome considered was days per week that the respondent felt overwhelmed by all they had to do. Work hours (β = .24*) and job flexibility (β = −.33**) were significantly associated with this outcome for mothers; work pressure (β = .31*) and coworker support (β = −.43***) were statistically significant in the model for fathers. For both mothers and fathers, then, a combination of job demands and resources was important to this subjective health outcome. This was also the case in the last regression models, which examined days per week respondents felt healthy and energetic. Again, the combination of work hours (β = −.32*) and job flexibility (β = .29*) was important to mothers, along with a positive association with occupational status (β = .32*). This suggests that the mothers who felt the most healthy and energetic were those of professional/managerial status who worked less hours and had greater job flexibility. Among fathers, those with greater coworker support (β = .41***) and less reported exposure to face time norms (β = −.22†) reported more days of feeling healthy and energetic. This was the only regression model in which face time norms was statistically significant. For the mothers and fathers in this sample, the job demands and resources examined here were more predictive of subjective health outcomes than preventive health behaviors, accounting for between 15 and 21 % of the variance in mothers’ subjective health outcomes and roughly 31 % of the variance in fathers’ subjective health outcomes.

Discussion

The current study examines how job demands and resources are associated with preventive health behaviors and subjective health outcomes among working parents of young children, a group of adults for whom free time is limited and work–family obligations are high and competing (Haas 1999; Hochschild 1997). Of the job demands and resources considered here (work hours, work pressure, face time norms, job flexibility, and coworker support), all were significantly associated with the dependent variable in at least one of the statistical models. Job demands and resources were more useful in models examining subjective health outcomes (feeling worried or stressed, feeling overwhelmed, and feeling healthy and energetic) than models examining preventive health behaviors (days of adequate exercise, sleep, and healthy diet). This provides limited evidence to suggest that mental health outcomes are more directly associated with work characteristics than behavioral outcomes, although support for both behavioral and subjective health outcomes was evidenced here. Working together, job demands and resources were associated with all of the examined health outcomes for mothers, with the exception of days per week eating a healthy diet. In this case, only job resources were significant. For fathers, the combination of job resources and demands was important in models for adequate exercise, and the three subjective health outcomes. Eating a healthy diet was associated with neither job demands nor job resources; rather, education was significantly associated with how often fathers reported healthy nutrition practices. Neither the model regressing days per week of adequate sleep nor the model regressing time to relax achieved statistical significance among the married fathers in this sample.

Considering each of the job demands and resources, work hours appeared to be important in helping working mothers and fathers find time for exercise. Additionally, work hours are influential in helping mothers find time for relaxation, feeling healthy and energetic, and avoiding feelings of being overwhelmed, worried, and stressed. This is consistent with previous empirical work documenting the negative impact of long work hours on motivation and time for exercise (Grzywacz and Marks 2001; Mattingly and Bianchi 2003; Nock and Kingston 1989), participation in physical activity (Taris et al. 2011), and higher levels of stress (Grzywacz and Marks 2001). The greater importance of work hours for mothers’ preventive health behaviors and subjective health outcomes likely speaks to the gendered nature of work–family obligations and decision-making faced by working women. Mothers are more likely to bend their work hours to fit the needs of family (Arendell 2001; Daly 2002; Gibb et al. 2014; Hilbrecht et al. 2008), and the obligations of both work and family are more likely to bleed into and shape the personal time of women (Bianchi et al. 2006; Bittman and Wajcman 2000; Maume et al. 2009; Shaw 1985). As noted, this high degree of time pressure felt by working mothers limits their ability to recuperate from the stressors they face in the work and family arenas. Ultimately, it can have a negative impact on their health (Mattingly and Sayer 2006).

Similar to past empirical work, feelings of work pressure were also positively associated with feelings of worry and stress (for mothers and fathers), as well as feeling overwhelmed (for fathers) (Karasek and Theorell 1990), and reporting fewer nights of adequate sleep (for mothers) (Eriksen et al. 2008; Lallukka et al. 2010; Roth and Roehrs 2003). The JD-R model warns that work characteristics can have behavioral, physical, and mental impacts, particularly when exposure to demands becomes routine (Bakker et al. 2007; Fredriksen-Goldsen and Scharlach 2001; Karasek and Theorell 1990; Roxburgh 1996; Voydanoff 2004). There is real concern that among those who face long work hours or high levels of work pressure on a regular basis, reduced participation in physical activity or sleep and increased stress will lead to more serious health consequences, such as heart disease (Karasek and Theorell 1990; Siegrist et al. 1990) or serious depression (Perry-Jenkins et al. 2011; Roxburgh 2004). Mothers and fathers in this study reported getting adequate exercise and time to relax about 2 days a week, feeling worried, stressed or overwhelmed about 3 days a week, and getting adequate sleep about 4 nights a week.

The last job demand considered was the experience of face time norms, which was significantly and negatively associated with days per week fathers felt healthy and energetic. Although work hours were not significant in this model, it is possible that the presence of face time norms is a signal of overwork among fathers. Those who face strong face time norms are often part of a culture of overwork that challenges the commitment of individual employees (Fried 1998). The routine mandate to “prove it” and climate of distrust surrounding workers who experience demands for face time creates an environment that appears to drain fathers’ energy and perceptions of health. Though there is limited empirical work that focuses on the experience of face time and its effects, this finding is consistent with the suggestion that a change in the culture of time at work is needed to enhance employee health (Moen et al. 2011a, b).

Optimistically, job resources appear to have the ability to help counter the negative impact of job demands, at least for some of the outcomes reviewed here. While controlling for job demands, job flexibility emerged as a significant variable in all but one of the regression models for mothers. Mothers who were able to control their work hours, whether they worked long hours or not, were better able to find time to themselves for leisure pursuits, ate a healthier diet more frequently, felt less overwhelmed, worried or stressed, and were more likely to report feeling healthy and energetic. Among fathers, job flexibility was important for getting adequate exercise and limiting feelings of worry and stress. By allowing employees to have some fluidity in work and work time, job flexibility can enhance employee well-being (Donovan and Halpern 2002; Nijp et al. 2012), particularly among women and employees with family responsibilities by reducing conflict between work and non-work demands (Jang et al. 2012; Thomas and Ganster 1995), and by providing feelings of control and empowerment (Hall and Atkinson 2006). But the benefit of job flexibility is generally most accessible and beneficial to professional workers with high levels of autonomy (Golden 2001). For instance, Tuttle and Garr (2012) found that schedule control among nonprofessional/nonmanagerial women who worked nonstandard shifts increased work-to-family conflict—perhaps because despite working schedules women remain responsible for family life (Maume and Sebastian 2012). In the regression models, feeling healthy and energetic was also associated with occupational status for mothers, suggesting that professional working mothers with high levels of job flexibility reported the highest levels of subjective health. It is interesting to note, however, that professional/managerial workers in this sample reported less job flexibility than parents in non-professional/managerial positions. Generally speaking, employees in higher status positions with greater pay and authority have more access to schedule flexibility, and access to flexibility does not significantly vary by gender (McCrate 2005). Thus, the current sample varied distinctly in this regard.

The second job resource included in the statistical models, coworker support, emerged as an important factor in all the models examining subjective health outcomes among fathers. This is interesting, as the fathers in this sample reported significantly less coworker support than the mothers who were studied, a finding consistent with other empirical studies of employed men and women (Schieman 2006; Thompson and Cavallaro 2007). Past work also indicates that though both men and women benefit from high levels of coworker support, its absence is more detrimental to working men (Geller and Hobfall 1993, 1994). Similarly, a study by Väänänen et al. (2003) found that coworker support was more important among male than female employees in the private industrial sector for reducing long term sickness absence. Roxburgh (1999) reported that for fathers, support in the workplace was more important than partner support in shaping satisfaction with work. Thus, although there are gender differences in the amount of coworker support received, it plays a more central role in shaping the quality of work and nonwork experiences of employed men.

Coworker support was found to be significantly associated with days of adequate sleep for mothers. Thus, as a job resource, for this sample coworker support had benefits primarily related to mental health for fathers and limited benefits to mothers’ preventive health behaviors. Perhaps the lack of findings in regard to preventive health behaviors is related to whether formalized approaches to promoting wellness were occurring within the workplace. As shown by others, social support from coworkers can be critical for accelerating behavioral commitment to healthy choices (Sorensen et al. 1998). Some have argued that given Americans’ significant time spent at work, coworkers are becoming a sort of “pseudo-family,” providing the same kind of support that family members do at home (Treiber and Davis 2012). Although fathers in this study reported less coworker support than mothers, the importance of that support varied significantly. Work-site health promotion programs that intervene at the social network level have been shown to be particularly effective in motivating and providing support for preventive health behaviors and well-being (Kelsey et al. 2000). These social support networks can be tapped into when attempting to build an organizational culture that values health (Treiber and Davis 2012). This sort of programming can have additional benefits for employee morale through enhanced workplace cultures. Indeed, a recent survey of employees found that over three-quarters of employees believed that health and wellness programs positively impact workplace climate (Martin 2013). The current study suggests that these reforms can help protect the emotional health of working fathers.

There are limitations to this work that should be addressed. First, there are commonly held concerns about self-reporting bias in health measures that occur on the basis of gender and education (Dowd and Todd 2011). Further, the use of cross-sectional data from a sample with limited diversity limits the study’s generalizability. As noted, this sample may vary in significant ways from the larger population of employees. Further, a precise response rate could not be determined for the sample used in this study. Nonetheless, findings from the current analysis have direct implications for employers seeking to create a positive work environment and enhance employees’ quality of life. Improvements in workers’ health result in better health for the organization (Holzer 2005).

Conclusion and Implications

The current study makes a contribution to the important literature examining workplace measures in considerations of health (Lutz 1989) by focusing on how job demands and resources work in tandem to shape the preventive health behaviors and subjective health outcomes of parents with young children. As they struggle to find time to meet work–family needs during their busy days (Haas 1999; Hochschild 1997), working mothers and fathers face challenges when attempting to prioritize their own health (Nomaguchi and Bianchi 2004). In addition to the intensive mothering and involved fathering culture of parenthood that employed parents face (Coltrane 1996; Hays 1996), subjective perceptions of well-being, such as heightened feelings of stress and worry and being overwhelmed can make it even more difficult for parents to feel they can make time for exercise, relaxation, and other leisure pursuits. Gender further complicates the negotiation between work and family needs and prioritizing personal health. This has implications for the structure and nature of work environments.

Findings suggest that it is the scheduling of working mothers’ time that matters most for their preventive health behaviors and subjective health outcomes. For working fathers, it is coworker support that most shapes subjective health. The multiple roles that men hold are generally defined in a more complementary manner than those held by women (Simon 1995). Thus, it comes as no surprise that temporal job resources are critically important for working mothers to help them manage multiple roles and avoid overload (Roxburgh 1997; Warren and Johnson 1995). The JD-R model highlights that when employees have inadequate resource access, they will be at greater risk for physical and psychological exhaustion and motivational loss (Demerouti and Bakker 2011). Women remain principally responsible for domestic work and child care, even as their labor market participation has risen (Coltrane 2000; Robinson and Godbey 1997). As noted by Daly (2002, p. 327), “men and women live with different sociotemporal expectancies that result in women continuing to give priority to their families.” When job flexibility is available, mothers are better able to coordinate personal and family needs and find time for health, especially in light of rising work hours. As such, employers who care about their workers’ health would do well to consider allowing employees to have flexible work times and personal control over work schedules to the degree possible in given work environments.

One recent workplace initiative that challenges standard time practices, including face time norms, is the “Results Only Work Environment” (ROWE). Instead of valuing face time and rigid scheduling practices, a ROWE workplace shifts its focus to employee productivity and quality of outcomes (Perlow and Kelly 2014). To date, the implementation of ROWE has been linked to less negative spillover of home into work life, fewer physical symptoms, and lower turnover. These effects hold regardless of the gender, age, or family life stage of employees (Moen et al. 2011a, b), although women employees have been more enthusiastic about its implementation (Kelly et al. 2010)—perhaps because they need it most. Because the ROWE model is an organizationally adopted work strategy, it may bypass the usual personal, work, and firm characteristics that combine to shape whether employees will adopt flexible job arrangements (Goñi-Legaz and Ollo-López 2014). It is one possible avenue to be considered, particularly for professional/managerial workers.

And how to increase coworker support to benefit the subjective health outcomes of employees, particularly fathers? The argument that occupational self-direction, including the degree to which work is routinized and employees have access to job autonomy, influences the ability of employees to form close relationships with their coworkers (Sloan et al. 2013) points to a few specific starting points. Again, models such as ROWE that allow workers a great deal of flexibility and autonomy in how work is performed may have an unintended and positive influence on relationships between employees, both directly and indirectly influencing employees’ health. But this model is not applicable in many nonprofessional work settings. Across a variety of workplaces, employers can offer training in open communication and team building exercises to encourage employees to work together more positively and meaningfully. These efforts will have long-term benefits for both the employee and the employer. Indeed, newer management theories place peer relationships at the center of organizational processes because of their importance to individual and organizational outcomes (Sias 2009). The current study shows that this emphasis is well placed, given the role of coworker support in shaping fathers’ health. As organizations move forward and become more progressive, recognizing that the personal health and satisfaction of employees has a direct impact on productivity and the bottom line (Chenoweth 2014), perhaps these and other reforms will become more commonplace. Flexibility and support will be of the utmost importance for working mothers and fathers.