Work–family spillover and metabolic syndrome indicators: Findings from a national sample

Abstract

This study examines the link between negative work–family spillover and metabolic risk factors over a 9-year period. Data from two waves of the Midlife in the United States Survey were used to explore relationships between negative work–family spillover and four indicators of metabolic syndrome—blood pressure, triglycerides, body mass index, and glucose levels. In a sample of full-time working men and women (N = 630), increased negative spillover at baseline significantly predicted higher body mass index nearly a decade later, with a marginally significant effect for triglyceride levels. Increases in spillover also body mass index and glucose levels at follow-up. This study extends research tying work–life spillover to health and suggests that further investigation is needed to fully understand the long-term effects of work stress.

Keywords

body mass index longitudinal metabolic syndrome negative spillover work–family conflict

The labor force has changed significantly in recent decades, and research suggests that work–family balance is a concern for working families (Allen and Martin, 2017; Clark, 2000; Michel et al., 2011). Although some studies indicate that firm boundaries between work and personal life can be beneficial, it is often the case that these boundaries are crossed, potentially creating conflict and strain between the two domains (Clark, 2000; Grandey and Cropanzano, 1999). Such interference may include job demands that interfere with household responsibilities or the inability to devote sufficient time and energy to family and friends outside of work (Greenhaus and Beutell, 1985; Kahn et al., 1964).

A number of studies have found that job demands are less important than the degree of role spillover in predicting stress and negative health outcomes (Kossek and Ozeki, 1998; Voydanoff, 2004). Negative spillover is defined as the bidirectional tension between roles and obligations of being a caregiver, spouse and an employee (Frone et al., 1997; Kossek and Ozeki, 1998). Negative work–family spillover (WFS) occurs when job responsibilities influence workers’ attitudes, capabilities, or energies toward family, creating difficultly in meeting obligations and expectations (Kanter, 1977). Since both men and women manage dual and often competing demands, negative spillover has become increasingly common, leading to distress, feeling overwhelmed, and burned out (Allen et al., 2000; Barnett and Baruch, 1985; Dettmers et al., 2016; Kelly et al., 2014).

Negative spillover, role strain, and health

Previous research finds that working conditions that induce stress (including those that interfere with personal life) are associated with poorer health outcomes (Beehr and Newman, 1978; Ganster and Rosen, 2013; Greenhaus and Allen, 2011; Rosenthal and Alter, 2012). Rooted in an interrole conflict framework, negative spillover is considered a chronic stressor (or strain) that depletes resources through competing demands on time and energy (Allen et al., 2000; Dich et al., 2015). A perceived or real depletion of resources gives rise to several negative psychological states (e.g. dissatisfaction, distress, depression, and anxiety), which may have long-term consequences (Grandey and Cropanzano, 1999; Greenhaus et al., 2006; Martire et al., 2000; Woods-Giscombé, 2010). In the absence of changes in role stress or environment, health can be compromised (Halbesleben et al., 2014). We draw upon role-strain theory in addition to a broad ecosocial paradigm (Krieger, 1994, 2001) to better understand the effect negative spillover may have on metabolic risk factors over time.

Metabolic syndrome and work–family relationships

The metabolic syndrome (or MetS) is comprised of multiple risk factors including high blood pressure, impaired glucose control, weight, and lipid dysregulation (Alberti et al., 2006; Grundy et al., 2004). Together, these indicators increase the risk for cardiovascular disease (CVD) and type 2 diabetes mellitus, two chronic conditions reflective of health decline and disparities in the United States (Haffner, 2006; Kaur, 2014). In general, research has only begun to link environmental and workplace stressors with physiological outcomes. The current project examines the effect of negative spillover on metabolic syndrome indicators.

A growing body of research links role strain to negative health (Fransson et al., 2015; Frone, 2000; Frone et al., 1992; McMunn et al., 2015; Nyberg et al., 2014; Spruill, 2010; Van Hedel et al., 2016). The mechanism linking role strain to metabolic risk factors is believed to operate through a stress response system, in which hormones are released in response to stressful stimuli, resulting in increased heart rate, cardiac activity, and blood pressure (McEwen, 2012; McEwen and Seeman, 1999; Terrill et al., 2012). While sympathetic responses to acute stress are well documented, whether chronic stress exposure (e.g. WFS) contributes to the elevation of risk factors over time is not clear (Hamer and Malan, 2010; Rozanski et al., 1999; Spruill, 2010). Some studies have shown that individuals who exhibit significant (or “exaggerated”) metabolic responses to acute stressors are at greater risk for CVD than those who show less reactivity (Low et al., 2009; Matthews et al., 1993; Treiber et al., 2003).

While useful in illuminating a potential pathway between stress and health, work–family studies to date have largely focused on acute, rather than chronic exposure to negative spillover (Berkman et al., 2015; Shockley and Allen, 2013). Advancing research in this area requires examining the effects of negative spillover over time. Understanding whether a delayed response exists, that is, whether exposure to negative spillover over the course of several years compromises health, is important in evaluating the relationship between the chronicity of spillover effects and negative health risk (Boylan and Ryff, 2015; Mottillo et al., 2010).

WFS and health over time

In considering previous research on work–family interference, 94 percent of these studies have taken a cross-sectional approach (Casper et al., 2007). Yet research indicates that trajectories of negative WFS may remain relatively stable over time (or increase for certain subgroups) before decreasing in later life (McMunn et al., 2015; Rantanen et al., 2012). No study to our knowledge has examined the effects of work–family conflict (WFC) on multiple metabolic syndrome risk factors over a nearly 10-year period. This is a critical gap in the literature, given that work–family tensions are likely to be ongoing. Therefore, the extent to which chronic stress from work–home responsibilities impact individual risk factors over time warrants additional investigation as several questions remain. First, does negative spillover, an indicator of WFC, carry the same negative effects over an extended period of time that has been indicated by cross-sectional studies? Second, what is the relationship between negative spillover over time and the metabolic syndrome, beyond those studied to date examining only one indicator? The focus of this study examines the effect of negative spillover over a 9-year span on four primary indicators of metabolic syndrome—body mass index (BMI), triglycerides, blood glucose, and systolic blood pressure.

Current study

This article contributes to the existing literature in several ways. First, we examine the change in negative spillover over nearly a decade. The majority of work–family studies tied to health have been cross-sectional, which constrains the researchers’ abilities to draw inferences about the relationship between work and family stressors and health considerably (Matthews et al., 2014). Second, we focus on primary indicators that are likely to reflect an increased risk in chronic conditions, such as CVD and type 2 diabetes mellitus, controlling for prior history of CVD-related conditions.

This study focuses on negative WFS as the exposure of emphasis, given the potential of this body of research to influence organizational policy broadly at the workplace level. Specifically, we hypothesize that higher levels of negative spillover at baseline predict higher levels of blood pressure, BMI, triglycerides, and glucose at time two, controlling for covariates (Hypothesis 1). Second, we hypothesize a positive relationship between change in negative spillover and biomarkers at time two, such that increases in WFS will be associated with higher levels of blood pressure, BMI, triglycerides, and glucose (Hypothesis 2). Because separate analyses did not find significant interactions by sex and marital status, we do not include these variables as potential moderators. The non-significant effect of marital status and sex on negative spillover has been documented in previous studies (Grzywacz, 2000; Panisoara and Serban, 2013), suggesting that work–family strain can present burdens for both men and women and is not necessarily limited by marital conventions. To account for any unintended cross-level effects, we control for both variables. Small numbers of racial subgroups were inadequate for longitudinal analysis or drawing conclusions regarding the effect of race. We present findings for a national sample of working adults with complete data at baseline and follow-up (N = 630).

Method

Participants and procedures

Data from the National Survey of Midlife Development in the United States (MIDUS) survey were used, which includes non-institutionalized, English-speaking adults. In the first wave of data collection (1995–1996), respondents were recruited via random digit dialing (RDD) from the 48 contiguous states. MIDUS I data collection (hereafter referred to as T1) occurred between January 1995 and September 1996. The second wave (MIDUS II) began in 2004 (hereafter referred to as T2) and continued until August 2005, with 75 percent of T1 respondents participating. Detailed information on the MIDUS assessments and longitudinal retention has been previously reported (Brim et al., 2004; Radler and Ryff, 2010). The MIDUS T2 survey added several new components (named projects) to collect additional data related to physiological characteristics and sample diversity. Projects 1–5 included a variety of assessments collected during T2 only. Biological data collected from a subset of T2 respondents comprised the Biomarker Project (P4), which aimed to investigate the long-term consequences of behavioral and psychosocial factors on health and illness. The Biomarker Project contained data from 1255 respondents. Excluding participants who did not complete survey data at MIDUS T1, data from the Biomarker Project at T2 was attained from 633 participants in the longitudinal sample. Participants with stroke (n = 3) at either of the two time points were removed; leaving 630 participants in the final analyses. Compared with the full longitudinal MIDUS sample, the subsample was younger (M = 42.7 in subsample, M = 46.8, in the original sample) and had a higher educational level. Approximately 93.2 percent of the subsample identified as White, 2.5 percent as African-American/Black, 0.3 percent as Asian, 0.6 percent as Native American, 0.5 percent as multiracial, and 2.1 percent as other. Women (49.8%; N = 314) and men (50.2%; N = 316) were nearly equally represented in the sample.

Measures

Negative spillover

Negative spillover, assessed at T1 and T2 by a 4-item scale created for the MIDUS study (Grzywacz and Marks, 2000), measures perceptions that work interferes with home functions and responsibilities (Grzywacz, 2000). Sample items include “Your job reduces the effort you can give to activities at home” and “Job worries or problems distract you when you are at home.” Responses were rated on a 5-point Likert-type scale. Appropriate items were reverse coded so that higher scores indicated a greater degree of spillover. A composite score summed the scores of all four individual items and was used to represent negative spillover in the analysis (see Table 1 for means and standard deviations). Internal consistencies at each time point was T1 α = .82 and T2 α = .81, respectively.

Table 1.

Mean, standard deviations, and correlations of study variables.

	M (SD)	1	2	3	4	5	6	7	8	9	10	11	12	13	14	15	16	17	18	19	20	21
1. Systolic BP T2	123.56 (17.47)	–
2. BMI T2	29.29 (5.93)	.24**	–
3. Triglycerides T2	136.89 (151.0)	.07	.17**	–
4. Glucose T2	99.86 (25.61)	.06	.17**	.45**	–											1
5. WFC T1	10.64 (2.64)	−.05	.11*	.05	.002	–
6. WFC change	−0.30 (2.83)	−.10*	−.04	.06	.03	−.53**	–
7. Age T1	43.73 (9.77)	.27**	−.02	0	.11**	−.08	−.16**	–
8. Sex (male)	0.50 (0.50)	.12**	.07	.16**	.17**	.02	−.02	.04	–
9. Separated/divorced T1	0.16 (0.36)	.01	0	−.03	−.05	−.06	−.03	.13**	−.17**	–
10. Not married T2	0.14 (0.35)	−.07	.02	−.01	−.09*	.03	.06	−.30**	.04	−.17**	–
11. Education level T1	7.83 (2.34)	−.11**	−.11**	−.06	−.04	.08*	−.05	.02	.03	−.04	.13**	–
12. Financial situation T1	6.18 (2.06)	.03	−.10*	−.04	−.01	−.17**	−.01	.19**	.07	−.11**	−.08*	.09*	–
13. Smoking history T1	0.59 (0.49)	−.01	0	.06	.06	−.04	−.02	.16**	.01	.09	−.08	−.19**	0	–
14. Smoking history T2	0.58 (0.49)	.01	.01	.07	.05	−.04	−.03	.13**	.03	.14**	−.09	−.22**	−.02	.91**	–
15. Drinking history T1	0.41 (0.49)	.01	.05	.07	.12**	.08	.02	−.07	.26**	−.04	.08	.02	−.08	.20**	.17**	–
16. Drinking history T2	0.71 (0.46)	−.09*	−.13**	.02	−.03	.05	−.01	−.01	−.03	−.04	.01	.10*	.14**	.01	−.01	.13**	–
17. Heart trouble T1	0.07 (0.26)	.11**	.04	.03	−.01	.13**	−.02	.05	.03	.05	−.08	−.01	−.04	−.03	.03	0	.03	–
18. Heart trouble T2	0.11 (0.32)	.10*	.06	.10*	.10*	.08	−.05	.25**	.05	.01	−.09*	−.04	.01	.05	.07	.02	.03	.27**	–
19. BP medication	0.07 (0.26)	.11**	.06	−.03	.07	−.06	−.07	.23**	.01	−.03	−.09*	.00	.06	0	−.01	.06	.01	.07	.09*	–
20. Physical activity T1	4.53 (1.48)	.02	−.07	.01	−.05	−.08*	.10*	−.21**	.21**	.00	.10*	.05	−.03	−.05	−.04	.11**	.04	−.08	−.11**	−.07	–
21. Physical activity T2	3.04 (1.36)	.04	−.06	.04	−.01	.02	.04	−.12**	.19**	−.09*	.06	−.04	−.03	.01	.02	.07	.091*	.01	.00	−.02	.36**	–
N total = 630

BP: blood pressure; BMI: body mass index; SD: standard deviation; WFC: work–family conflict; T1: Time 1 (Year 1995); T2: Time 2 (Year 2003).

p < .05; **p < .001.

Biomarker data.

Biomarker data were collected from three General Clinical Research Centers (at the University of California, Los Angeles (UCLA), the University of Wisconsin, and Georgetown University) by trained staff members or clinicians. Participants’ height, weight, and blood pressure were measured as part of a physical exam. BMI was assessed directly using height (cm) and weight (kg) measurements. Triglycerides, glucose levels, and blood pressure were assessed from lipid assay panels collected during the clinical encounter. Full details about protocol and procedures have been previously detailed (Dienberg Love et al., 2010).

Covariates

Control variables were age (years), sex, educational level (scores range from 1 = no school/some grade school to 12 = professional degree), T1 financial situation, T1 marital status, T1 use of blood pressure medication, heart conditions at T1 and T2, smoking and drinking history, as well as physical activity at T1 and T2. T1 blood pressure medication was measured by a question asking participants to indicate whether they were taking prescribed medication for high blood pressure (1 = Yes, 0 = No). Heart trouble was identified by asking the participants “Have you ever had heart trouble suspected or confirmed by a doctor?” (1 = Yes, 0 = No) at both T1 and T2.

Data analysis

Analyses involved two steps to address primary hypotheses. First, T1 indicators were used to predict T2 biomarker outcomes (H1). A series of logistic regression models were constructed with covariates and T1 baseline levels of spillover, predicting systolic blood pressure, BMI, blood glucose, and triglycerides at T2. In a second analysis, covariates and baseline levels of negative spillover (T1) were entered with degree of change in negative spillover between T1 and T2 modeled against outcomes of interest (H2). In both models, outcomes were entered according to “high-risk” cutoffs of clinical significance (Alberti et al., 2006; Grundy et al., 2004). Since the metabolic syndrome confers increased risk of disease dependent upon higher levels of indicators, outcome variables were dichotomized to reflect increased risk using cutoff levels consistent with conventions from the American Heart Association, such as levels of BMI over 25 to confer overweight status and fasting plasma glucose levels higher than 110 to indicate pre-diabetes (1 = higher than 110 mg/dL, 0 = otherwise).

Change scores were used in the second step, as they often are, when only two time points are available (Peter et al., 1993; Rogosa, 1995; Turiano et al., 2012). Change scores were calculated using negative spillover at T1 subtracted from T2 scores. Individuals with positive change scores were those for whom the T2 score was higher than the T1 score (i.e. spillover increased over time). Conversely, negative change scores indicate a T2 score lower than the T1 evaluation of spillover (i.e. spillover decreased over time).

Results

Overall, there was a positive change (increase over time) in spillover between T1 and T2 in this study. Correlations, means, and standard deviations for key study variables are noted in Table 1.

Systolic blood pressure at T2

After adjusting for covariates, negative spillover at T1 did not significantly predict higher blood pressure at T2, nor did increases in spillover predict elevated blood pressure (see Table 2).

Table 2.

Predicting biomarkers at T2 from negative spillover (negWF) at T1 and change in negative spillover from T1 to T2 (negWF change).

	Systolic blood pressure			BMI			Triglycerides			Glucose
	B	SE	e^B	B	SE	e^B	B	SE	e^B	B	SE	e^B
Model 1	Nagelkerke R² = .15	Nagelkerke R² = .10	Nagelkerke R² = .19	Nagelkerke R² = .13
Age	0.05**	0.01	1.05	−0.02	0.01	0.98	−0.01	0.02	0.99	0	0.02	1
Male	0.37	0.25	1.44	0.42^†	0.25	1.52	1.60*	0.42	4.97	0.63	0.38	1.88
Separated/divorced/widowed T1	0.08	0.34	1.08	0.24	0.33	1.27	−1.23^†	0.66	0.29	−0.08	0.48	0.93
Not married T2	0.38	0.34	1.47	−0.15	0.34	0.86	0.22	0.47	1.24	−1.04	0.65	0.35
Education T1	−0.15*	0.05	0.86	−0.03	0.05	0.97	−0.24*	0.08	0.79	0.02	0.07	1.02
Financial T1	0	0.06	1.00	−0.12*	0.06	0.88	0	0.09	1.00	0.09	0.09	1.09
Smoking history T1	−0.28	0.56	0.76	−0.55	0.57	0.58	0.67	0.81	1.96	0.5	0.83	1.65
Smoking history T2	−0.09	0.55	0.92	0.62	0.57	1.85	−0.24	0.82	0.79	0.25	0.82	1.28
Drinking history T1	0.22	0.25	1.24	0.19	0.24	1.21	0.02	0.34	1.02	0.17	0.34	1.18
Drinking history T2	−0.26	0.27	0.77	−0.62*	0.26	0.54	0.09	0.37	1.10	−0.51	0.36	0.6
Heart trouble T1	1.55*	0.60	4.72	0.42	0.49	1.52	0.95	0.57	2.58	−0.44	0.65	0.64
Heart trouble T2	−0.06	0.39	0.94	0.36	0.38	1.43	0.06	0.49	1.06	1.18*	0.44	3.26
BP medication T1	0.65	0.45	1.92	0.74^†	0.42	2.09	−0.93	0.73	0.39	0.15	0.54	1.17
Physical activity T1	0.10	0.09	1.11	−0.10	.008	.090	−0.08	0.12	0.92	0.01	0.12	1.01
Physical activity T2	−0.03	0.09	0.97	−0.05	0.09	0.95	−0.17	0.14	0.84	−0.01	0.13	0.99
Model 2	Nagelkerke R² = .17			Nagelkerke R² = .14			Nagelkerke R² = .21			Nagelkerke R² = .15
Age	0.05**	0.05	1.05	−0.01	0.01	0.99	−0.01	0.02	0.99	0.01	0.02	1.01
Male	0.34	0.34	1.40	0.46^†	0.26	1.58	1.58*	0.43	4.85	0.68^†	0.39	1.97
Separated/divorced/widowed T1	0.02	0.02	1.02	0.32	0.34	1.38	−1.19^†	0.67	0.30	0.04	0.49	1.04
Not married T2	0.40	0.40	1.50	−0.15	0.34	0.86	0.19	0.47	1.21	−1.06	0.65	0.35
Education T1	−0.16 **	−0.16	0.86	−0.04	0.05	0.96	−0.25*	0.08	0.78	0.02	0.07	1.02
Financial T1	0.01	0.01	1.01	−0.09	0.06	0.92	0.03	0.09	1.03	0.1	0.1	1.1
Smoking history T1	−0.31	−0.31	0.73	−0.63	0.59	0.53	0.66	0.81	1.94	0.5	0.83	1.65
Smoking history T2	−0.07	−0.07	0.94	0.77	0.59	2.17	−0.17	0.81	0.84	0.3	0.83	1.35
Drinking history T1	0.24	0.24	1.28	0.10	0.24	1.10	−0.06	0.35	0.94	0.1	0.35	1.1
Drinking history T2	−0.25	−0.25	0.78	−0.70*	0.27	0.50	0.08	0.37	1.08	−0.55	0.36	0.58
Heart trouble T1	1.61*	1.61	5.03	0.26	0.50	1.30	0.82	0.59	2.27	−0.59	0.67	0.56
Heart trouble T2	−0.04	−0.04	0.96	0.19	0.38	1.21	−0.08	0.50	0.92	1.14*	0.45	3.13
BP medication T1	0.61	0.61	1.84	0.95*	0.43	2.58	−0.76	0.75	0.47	0.33	0.55	1.39
Physical activity T1	0.11	0.11	1.11	−0.07	0.09	0.93	−0.05	0.13	0.96	0.02	0.12	1.02
Physical activity T2	−0.03	−0.03	0.97	−0.08	0.09	0.92	−0.19	0.14	0.83	−0.02	0.13	0.98
WFC T1	0.02	−0.02	0.98	0.19**	0.06	1.21	0.15^†	0.08	1.16	0.10	0.08	1.1
WFC change	0.09^†	−0.09	0.92	0.12*	0.05	1.13	0.09	0.07	1.09	0.14*	0.07	1.16

T1: Time 1 (Year 1995); T2: Time 2 (Year 2003); e^B: exponentiated B; BMI: body mass index; SE: standard error; WFC: work–family conflict.

†

.05 < p < .08; *p < .05; **p < .001.

BMI at T2

In this model, negative spillover at T1 predicted higher BMI at T2, B (SE) = 0.19 (0.06), p < .001, adjusting for covariates. Increase in spillover (indicated by positive change score) from T1 to T2 also predicted higher BMI at T2 (p < .05).

Triglycerides at T2

Negative spillover at T1 marginally increased the likelihood for raised triglycerides at T2. Increased spillover over time was not significantly associated with higher triglyceride levels at T2.

Blood glucose at T2

Negative spillover at T1 was not significantly associated with higher blood glucose levels at T2; however, increase in spillover did predict higher blood glucose at T2, B (SE) = 0.14 (0.07), p < .05.

Taken together, baseline levels of negative spillover at T1 predicted higher levels of BMI and were marginally associated with elevated triglycerides at T2 while increases in negative spillover between T1 and T2 predicted higher levels of BMI and blood glucose levels at T2.

Discussion

In a national study of midlife working adults, we examined the relationship between negative WFS and metabolic syndrome biomarkers at a 9-year follow-up. Results from this study extend research on WFC and aid in better understanding the impact of negative spillover over time on indicators of health. Higher levels of spillover at baseline significantly predict increased BMI nearly a decade later, and increases in spillover predict increased BMI and high blood glucose levels at follow-up. Interestingly, negative spillover did have a significant effect on blood pressure. Overall, this study indicates the potential of negative spillover from work to family life to have an impact on metabolic risk factors years later, particularly for BMI. This research has the potential to broaden the scope of investigations of negative spillover and chronic disease biomarkers.

Although modest effects were observed, they should be expected given that weaker associations between self-report measures (e.g. perceived negative spillover) and objective measures have been well documented (Herbert and Cohen, 1993). Compared with self-reported measures, objective measures are more strongly associated with physiological and anthropometric biomarkers (Atienza et al., 2011; Lynch et al., 2010). Therefore, the ability to track these associations using subjective perceptions of negative spillover 9 years apart suggests that the relationships between perceived work–family spillover and physiological outcomes are moderately strong. Regarding the clinical significance of our findings, even modest changes in BMI and blood glucose levels may confer increased risk of diabetes and vascular-related disease in later life, especially when considered with other risk factors (Alberti et al., 2006; Brotman et al., 2007; Everson et al., 1996). Additional longitudinal investigations on the effects of negative spillover are needed to clarify the role of occupational stressors in the etiology of disease risk.

Considerable research has been devoted to the study of work spillover effects, yet little of this research has examined the effect of spillover over time. The effects of WFS are likely to develop over years, often decades, and cross-sectional studies can be limited in discerning latent risk and lag effects. More importantly, this study suggests lasting effects of WFS in that baseline measures are related to key indices of health in a national sample nearly a decade later. This research supports similar findings from smaller samples conducted in occupation-specific studies (Jacobsen et al., 2014). Overall, our results emphasize the importance of exploring more complicated dimensions of the work-family interface, including how it impacts multiple indicators of health over time and to what extent these effects might be mediated by other protective or risk factors.

Several limitations warrant mention. Of primary concern is the lack of biomarker assessments at MIDUS T1, precluding testing of longitudinal relationships in physiological outcomes. Therefore, causality cannot be determined. Second, given the selected sampling of participants with both longitudinal and biomarker data, there was limited representation of individuals from racial and ethnic minority groups. Third, self-report measures were used for assessing WFS. While it is important to consider potential bias associated with using self-report measures, this limitation represents a broad challenge within the work–family literature (Allen, 2013). Furthermore, this investigation was focused on understanding the perceptions of spillover and therefore self-report is likely appropriate in distilling the effect of perceived stress from work and family. Despite some limitations, the observed associations support models of work-related strain, contributing to theoretical frames that link occupational stressors to health.

Future directions

Given these findings, one direction for future research is a continuation of studies linking occupational stressors, particularly stress related to interferences between work and family, to health and risk factors for health. There is an especially critical need to highlight potential mechanisms through which WFC is consequential for health, expanding the depth of existing theoretical frameworks. Therefore, future studies may benefit from attending to both physiological and psychological manifestations of WFS over time and exploring psychosocial processes.

In particular, both baseline levels and increases in negative spillover over time predict elevated BMI, indicating that weight fluctuations may be associated with work–life balance, as recent research suggests (Lacey et al., 2017). Additional considerations, such as age of transition into the labor force, organizational and family support, consistency of ties to employment, and behavioral or stress-related pathways associated with these transitions may be important. Integrating this research requires a contextual approach that frames what has been described as “conflict,” “spillover,” and “interference” as a cumulative stressor, all of which are likely to indicate a chronic burden yielding wide-ranging effects (Edwards and Rothbard, 2000). Linking occupational stress with biomedical and psychological research represents an important step in advancing the implications for occupational research on health generally.

At the same time, additional attention is needed to understand preventive pathways that may stem the effects of WFC. In recent years, researchers have developed interventions focused on training supervisors in family support strategies, yielding benefits for employee physical health, job satisfaction, and reduced turnover (Hammer et al., 2011; Odle-Dusseau et al., 2016). Other studies have found that alternative work schedule initiatives and in-house childcare programs were beneficial to workers’ job quality and overall well-being (Kelly et al., 2014; Moen et al., 2011a, 2011b). Therefore, in spite of its complex nature, negative WFS can be intervened upon. In addition to framing organizational change as a benefit for employers and to some extent, job quality, a primary aim should measure how these changes impact employee health (Hwang and Hong, 2012; Kivimäki and Kawachi, 2015). To what extent modifications to workplace policies, degree of autonomy (e.g. schedule control), social support, or integration between home and family results in reducing conflict long term will be important considerations in future research.

Footnotes

Appendix

Authors used the publicly available MIDUS database for this study. A description of the MIDUS study can be found at http://midus.wisc.edu/. A bibliography of journal articles, working papers, conference presentations, and dissertations using these data is available at http://midus.wisc.edu/findings/index.php. All variables and relationships examined in this article have not been examined in any previous or current articles, or to the best of our knowledge, in any papers that will be under review soon.

Declaration of conflicting interests

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

Funding

The authors received no financial support for the research, authorship, and/or publication of this article.

ORCID iD

H Shellae Versey

References

Alberti

Zimmet

Shaw

(2006) Metabolic syndrome—A new world-wide definition: A consensus statement from the international diabetes federation. Diabetic Medicine 23(5): 469–480.

Allen

(2013). The work-family role interface: A synthesis of the research from industrial and organizational psychology. In Schmitt

N. W.

Highhouse

Weiner

I. B.

(Eds), Handbook of psychology: Industrial and organizational psychology (pp. 698-718). Hoboken, NJ: John Wiley & Sons Inc.

Allen

Martin

(2017) The work-family interface: A retrospective look at 20 years of research in JOHP. Journal of Occupational Health Psychology 22(3): 259–272.

Allen

Herst

DEL

Bruck

, et al. (2000) Consequences associated with work-to-family conflict: A review and agenda for future research. Journal of Occupational Health Psychology 5: 278–308.

Atienza

Moser

Perna

, et al. (2011) Self-reported and objectively measured activity related to biomarkers using NHANES. Medicine & Science in Sports & Exercise 43: 815–821.

Barnett

Baruch

(1985) Women’s involvement in multiple roles and psychological distress. Journal of Personality and Social Psychology 49: 135–145.

Beehr

Newman

(1978) Job stress, employee health, and organizational effectiveness: A facet analysis, model, and literature review. Personnel Psychology 31: 665–699.

Berkman

Liu

Hammer

, et al. (2015) Work-family conflict, cardiometabolic risk, and sleep duration in nursing employees. Journal of Occupational Health Psychology 20(4): 420–433.

Boylan

Ryff

(2015) Psychological well-being and metabolic syndrome: Findings from the Midlife in the United States national sample. Psychosomatic Medicine 77(5): 548–558.

10.

Brim

Ryff

Kessler

(2004) How Healthy Are We? A National Study of Well-Being at Midlife. Chicago, IL: The University of Chicago Press.

11.

Brotman

Golden

Wittstein

(2007) The cardiovascular toll of stress. The Lancet 370: 1089–1100.

12.

Casper

Eby

Bordeaux

, et al. (2007) A review of research methods in IO/OB work-family research. Journal of Applied Psychology 92: 28.

13.

Clark

(2000) Work/family border theory: A new theory of work/family balance. Human Relations 53: 747–770.

14.

Dettmers

Vahle-Hinz

Bamberg

, et al. (2016) Extended work availability and its relation with start-of-day mood and cortisol. Journal of Occupational Health Psychology 21: 105–118.

15.

Dich

Lange

Head

, et al. (2015) Work stress, caregiving and allostatic load: Prospective results from Whitehall II cohort study. Psychosomatic Medicine 77(5): 539–547.

16.

Dienberg Love

Seeman

Weinstein

, et al. (2010) Bioindicators in the MIDUS national study: Protocol, measures, sample, and comparative context. Journal of Aging and Health 22: 1059–1080.

17.

Edwards

Rothbard

(2000) Mechanisms linking work and family: Clarifying the relationship between work and family constructs. Academy of Management Review 25: 178–199.

18.

Everson

Kaplan

Goldberg

, et al. (1996) Anticipatory blood pressure response to exercise predicts future high blood pressure in middle-aged men. Hypertension 27: 1059–1064.

19.

Fransson

Nyberg

Heikkilä

, et al. (2015) Job strain and the risk of stroke. Stroke 46: 557–559.

20.

Frone

(2000) Interpersonal conflict at work and psychological outcomes: Testing a model among young workers. Journal of Occupational Health Psychology 5: 246–255.

21.

Frone

Russell

Cooper

(1992) Antecedents and outcomes of work-family conflict: Testing a model of the work-family interface. Journal of Applied Psychology 77: 65–78.

22.

Frone

Russell

Cooper

(1997) Relation of work–family conflict to health outcomes: A four-year longitudinal study of employed parents. Journal of Occupational and Organizational Psychology 70: 325–335.

23.

Ganster

Rosen

(2013) Work stress and employee health: A multidisciplinary review. Journal of Management 39: 1085–1122.

24.

Grandey

Cropanzano

(1999) The conservation of resources model applied to work–family conflict and strain. Journal of Vocational Behavior 54: 350–370.

25.

Greenhaus

Allen

Spector

(2006) Health consequences of work–family conflict: The dark side of the work–family interface. In: Perrewé

Ganster

(eds) Research in Occupational Stress and Well Being: Employee Health, Coping and Methodologies, vol. 5. Oxford: Elsevier, pp. 61–98.

26.

Greenhaus

Allen

(2011) Work–family balance: A review and extension of the literature. In: Quick

Tetrick

(eds) Handbook of Occupational Health Psychology. Washington, DC: American Psychological Association, pp. 165–183.

27.

Greenhaus

Beutell

(1985) Sources of conflict between work and family roles. Academy of Management Review 10: 76–88.

28.

Grundy

Brewer

Cleeman

, et al. (2004) Definition of metabolic syndrome: Report of the National heart, lung, and blood institute/American heart association conference on scientific issues related to definition. Circulation 109(3): 433–438.

29.

Grzywacz

(2000) Work-family spillover and health during midlife: Is managing conflict everything? American Journal of Health Promotion 14: 236–243.

30.

Grzywacz

Marks

(2000) Reconceptualizing the work–family interface: An ecological perspective on the correlates of positive and negative spillover between work and family. Journal of Occupational Health Psychology 5: 111–126.

31.

Haffner

(2006) The metabolic syndrome: Inflammation, diabetes mellitus, and cardiovascular disease. American Journal of Cardiology 97: 3–11.

32.

Halbesleben

Neveu

J-P

Paustian-Underdahl

, et al. (2014) Getting to the “COR” understanding the role of resources in conservation of resources theory. Journal of Management 40: 1334–1364.

33.

Hamer

Malan

(2010) Psychophysiological risk markers of cardiovascular disease. Neuroscience and Biobehavioral Reviews 35: 76–83.

34.

Hammer

Kossek

Anger

, et al. (2011) Clarifying work–family intervention processes: The roles of work–family conflict and family-supportive supervisor behaviors. Journal of Applied Psychology 96: 134–150.

35.

Herbert

Cohen

(1993) Stress and immunity in humans: A meta-analytic review. Psychosomatic Medicine 55: 364–379.

36.

Hwang

Hong

(2012) Work-related cardiovascular disease risk factors using a socioecological approach: Implications for practice and research. European Journal of Cardiovascular Nursing 11: 114–126.

37.

Jacobsen

Reme

Sembajwe

, et al. (2014) Work-family conflict, psychological distress, and sleep deficiency among patient care workers. Workplace Health & Safety 62: 282–291.

38.

Kahn

Wolfe

Quinn

, et al. (1964) Organizational Stress: Studies in Role Conflict and Ambiguity. Oxford: John Wiley & Sons.

39.

Kanter

(1977) Some effects of proportions on group life: Skewed sex ratios and responses to token women. American Journal of Sociology 82(5): 965–990.

40.

Kaur

(2014) A comprehensive review on metabolic syndrome. Cardiology Research and Practice. Epub ahead of print 11 March. DOI: 10.1155/2014/943162.

41.

Kelly

Moen

Oakes

, et al. (2014) Changing work and work-family conflict: Evidence from the work, family, and health network. American Sociological Review 79: 485–516.

42.

Kivimäki

Kawachi

(2015) Work stress as a risk factor for cardiovascular disease. Current Cardiology Reports 17: 74.

43.

Kossek

Ozeki

(1998) Work-family conflict, policies, and the job-life satisfaction relationship: A review and directions for organizational behavior-human research. Journal of Applied Psychology 83(2): 139–149.

44.

Krieger

(1994) Epidemiology and the web of causation: Has anyone seen the spider? Social Science & Medicine 39: 887–903.

45.

Krieger

(2001) Theories for social epidemiology in the 21st century: An ecosocial perspective. International Journal of Epidemiology 30: 668–677.

46.

Lacey

Sacker

Bell

, et al. (2017) Work-family life courses and BMI trajectories in three British birth cohorts. International Journal of Obesity 41: 332–339.

47.

Low

Salomon

Matthews

(2009) Chronic life stress, cardiovascular reactivity, and subclinical cardiovascular disease in adolescents. Psychosomatic Medicine 71: 927–931.

48.

Lynch

Dunstan

Healy

, et al. (2010) Objectively measured physical activity and sedentary time of breast cancer survivors, and associations with adiposity: Findings from NHANES (2003–2006). Cancer Causes & Control 21: 283–288.

49.

McEwen

(2012) Brain on stress: How the social environment gets under the skin. Proceedings of the National Academy of Sciences 109: 17180–17185.

50.

McEwen

Seeman

(1999) Protective and damaging effects of mediators of stress: Elaborating and testing the concepts of allostasis and allostatic load. Annals of the New York Academy of Sciences 896: 30–47.

51.

McMunn

Lacey

Worts

, et al. (2015) De-standardization and gender convergence in work-family life courses in Great Britain: A multi-channel sequence analysis. Advanced Life Course Research 26: 60–75.

52.

Martire

Stephens

MAP

Townsend

(2000) Centrality of women’s multiple roles: Beneficial and detrimental consequences for psychological well-being. Psychology and Aging 15: 148–156.

53.

Matthews

Woodall

Allen

(1993) Cardiovascular reactivity to stress predicts future blood pressure status. Hypertension 22: 479–485.

54.

Matthews

Wayne

Ford

(2014) A work–family conflict/subjective well-being process model: A test of competing theories of longitudinal effects. Journal of Applied Psychology 99: 1173–1187.

55.

Michel

Kotrba

Mitchelson

, et al. (2011) Antecedents of work–family conflict: A meta-analytic review. Journal of Organizational Behavior 32: 689–725.

56.

Moen

Kelly

Hill

(2011a) Does enhancing work-time control and flexibility reduce turnover? A naturally occurring experiment. Social Problems 58: 69–98.

57.

Moen

Kelly

Tranby

, et al. (2011b) Changing work, changing health: Can real work-time flexibility promote health behaviors and well-being? Journal of Health and Social Behavior 52: 404–429.

58.

Mottillo

Filion

Genest

, et al. (2010) The metabolic syndrome and cardiovascular risk: A systematic review and meta-analysis. Journal of the American College of Cardiology 56(14): 1113–1132.

59.

Nyberg

Fransson

Heikkilä

, et al. (2014) Job strain as a risk factor for type 2 diabetes: A pooled analysis of 124,808 men and women. Diabetes Care 37: 2268–2275.

60.

Odle-Dusseau

Hammer

Crain

, et al. (2016) The influence of family-supportive supervisor training on employee job performance and attitudes: An organizational work–family intervention. Journal of Occupational Health Psychology 21: 296–308.

61.

Panisoara

Serban

(2013) Marital status and work-life balance. Procedia-social and Behavioral Sciences 78: 21–25.

62.

Peter

Churchill

Jr Brown

(1993) Caution in the use of difference scores in consumer research. Journal of Consumer Research 19: 655–662.

63.

Radler

Ryff

(2010) Who participates? Accounting for longitudinal retention in the MIDUS national study of health and well-being. Journal of Aging and Health 22: 307–331.

64.

Rantanen

Kinnunen

Pulkkinen

, et al. (2012) Developmental trajectories of work–family conflict for Finnish workers in midlife. Journal of Occupational Health Psychology 17: 290–303.

65.

Rogosa

(1995) Myths and methods: “Myths about longitudinal research” plus supplemental questions. In: Gottman

(ed.) The Analysis of Change. Hove: Psychology Press, pp. 3–66.

66.

Rosenthal

Alter

(2012) Occupational stress and hypertension. Journal of the American Society of Hypertension 6: 2–22.

67.

Rozanski

Blumenthal

Kaplan

(1999) Impact of psychological factors on the pathogenesis of cardiovascular disease and implications for therapy. Circulation 99: 2192–2217.

68.

Shockley

Allen

(2013) Episodic work–family conflict, cardiovascular indicators, and social support: An experience sampling approach. Journal of Occupational Health Psychology 18: 262–275.

69.

Spruill

(2010) Chronic psychosocial stress and hypertension. Current Hypertension Reports 12: 10–16.

70.

Terrill

Garofalo

Soliday

, et al. (2012) Multiple roles and stress burden in women: A conceptual model of heart disease risk. Journal of Applied Biobehavioral Research 17: 4–22.

71.

Treiber

Kamarck

Schneiderman

, et al. (2003) Cardiovascular reactivity and development of preclinical and clinical disease states. Psychosomatic Medicine 65: 46–62.

72.

Turiano

Pitzer

Armour

, et al. (2012) Personality trait level and change as predictors of health outcomes: Findings from a national study of Americans (MIDUS). Journals of Gerontology Series B: Psychological Sciences and Social Sciences 67: 4–12.

73.

Van Hedel

Mejía-Guevara

Avendaño

, et al. (2016) Work–family trajectories and the higher cardiovascular risk of American women relative to women in 13 European countries. American Journal of Public Health 106: 1449–1456.

74.

Voydanoff

(2004) The effects of work demands and resources on work-to-family conflict and facilitation. Journal of Marriage and Family 66(2): 398–412.

75.

Woods-Giscombé

(2010) Superwoman schema: African American women’s views on stress, strength, and health. Qualitative Health Research 20: 668–683.