SHRM Foundation Research
Not a Member?  Become One Today!

SHRM Foundation Research


Unhealthy Balance?  The Consequences of Work and Family Demands and Resources on Employees' Health and Health Behaviors

Funded: June 2010     Completed: September 2013 

Theresa Glomb, Ph.D. , Jean Abraham, Ph.D. , Erin Kelly, Ph.D. , University of Minnesota

Executive Summary
Given current demographic and economic trends, there is growing interest in understanding the effects of the demands of work (e.g., work hours, job demands), family (e.g., children, marriage), and the resulting inter-role conflicts on workers’ health. Understanding these issues is valuable for employers given the large financial implications associated with health insurance provision and illness-related productivity losses.

Although prior reviews suggest the adverse effect of work and family demands on psychological and physical health, they also suggest existing research is narrow in scope and has important methodological limitations. Specifically, research typically (a) is based on small samples from single organizations that lack diversity in demographic and family structural characteristics, (b) examines health outcomes using narrowly defined conditions or self-reports of overall health, (c) does not examine health-related behaviors such as regular exercise and (d) primarily uses cross-sectional data.

To address these limitations, we integrate the literatures from organizational sciences, with its detailed attention to work-family conflict, and health economics, with its detailed examination of medical conditions and expenditures, to investigate the health-related toll of work and family demands on employees. We examine the relationship of work and family demands (captured using structural measures of these concepts) on workers’ health measured by the presence of stress-related conditions including (1) hypertension, (2) gastrointestinal disorders, and (3) anxiety or mood disorders, and workers’ health behaviors, measured by (4) lack of regular exercise and (5) smoking. We also examine health care utilization associated with these conditions in the form of office based visits and prescriptions.


Although many demands did not evidence strong and consistent effects on conditions, overall our results suggest that work demands are generally detrimental to one’s health. But, surprisingly, we find that family structural demands such as the presence of children tended to be generally beneficial for one’s health.

1. Overall there were few relationships between emotional and cognitive job demands and health outcomes. Higher emotional labor demands were associated with a higher chance of reporting hypertension among married males and married females, suggesting the physiologically taxing nature of managing interpersonal interactions in the workplace. However, contrary to hypotheses, emotional labor demands were also associated with more regular exercise among males (non-married) and females (married); perhaps this is a stress relief mechanism sought by those who work in jobs with high interactions. Cognitive demands increased the likelihood of smoking among married men and reduced the likelihood among married females. Cognitive demands also reduced the likelihood of hypertension among married females.

2. Contrary to our predictions, higher physical demands evidence some beneficial health effects particularly for males. Specifically, working in an occupation with greater physical demands was associated with a lower chance of reporting gastrointestinal problems and anxiety or mood disorders among non-married males. The effects for women were mixed, with non-married females experiencing lower hypertension with physical demands and married females experiencing higher hypertension in the panel analyses. Physical demands also resulted in more regular physical activity among both married and non-married males and females. However, it would be premature to speculate that work demands might be a source of healthy behaviors; importantly, we find that physical demands are associated with greater likelihood of smoking among married males and unmarried females.

3. Work hours showed few relations to our medical conditions, but were related to health behaviors; specifically, work hours were associated with greater likelihood of smoking among males and married females.

4. In terms of family demands, we largely found the flip side of our predictions for the health conditions. Among married men, children were clearly beneficial; the more children male respondents had, the less likely that they reported hypertension, gastrointestinal problems, anxiety or mood disorders, and actively smoking. For females, the effects of children were minimal; only hypertension (among non-married) and gastrointestinal issues (among married) were reduced. Despite the beneficial effects of children, there is one outcome where they were harmful—regular physical activity. Our results suggest that children are associated with a lack of regular physical activity for both males and females in most model specifications. Due to our stratification by marital status, we did not examine marital status directly, however examination of the pattern of results suggests that marriage might also have protective effects on health.

5. Overall, we found few effects of the role of work resources on health conditions and behaviors. Longer job tenure was associated with a higher chance of reporting hypertension among males and non-married females. This is opposite of our prediction, but given that tenure may be related to position in the organization it may be indicative of job demands not captured in our indices. Tenure is also related to a reduced likelihood of smoking among men. The findings for the resource of employer-provided benefits were mostly negligible; provision of benefits was associated with a reduced likelihood of smoking (for males) and increased chance of physical activity (married males). However, benefits may increase the likelihood of utilization and condition identification which would work against our hypotheses. 

6. The results for income suggest it is an important family resource and related to improved health; household income was associated with lower hypertension among non-married females, lower anxiety/mood among non-married males, and more physical activity for married females. Income was associated with a lower likelihood of smoking for males and females.

7. Analyses examined whether work and family demands had an effect on health care utilization, defined as having and office based doctor visit or a prescription, once one of the three conditions (anxiety/mood disorder, hypertension, GI) was present. Overall, there were few significant effects of our work and family demands on utilization once controls were included.

Study Methods
Our study uses the nationally representative Medical Expenditure Panel Survey (MEPS) for 1997-2007 a nationally representative annual survey using an overlapping panel design of approximately 30,000 individuals conducted by the Agency of Healthcare Research and Quality (AHRQ). The MEPS provides a comprehensive assessment of conditions, with surveyors collecting information from respondents for five rounds over a two-year period who detail associated medical conditions (coded by trained coders into five-digit International Classification of Diseases, 9th Revision (ICD-9) codes) and medical care utilization associated with each condition. The MEPS includes employees in a diverse set of occupations and industries as well as under-studied groups such as single parents. Our sample includes adults 19 to 64 years of age who were employed at least 30 hours per week, who reported being continuously insured, and who were recorded as the primary respondent to the survey.

Download the Full Report. 

View the full list of SHRM Foundation funded research.