Stress is Associated with Adiposity in Parents of Young Children
Abstract
Objective
This study investigated associations between stress (general stress, parenting distress, and household chaos) and adiposity among parents of young children.
Methods
The sample consisted of 49 mothers and 61 fathers from 70 families with young children living in Ontario, Canada. Linear regression using generalized estimating equations was used to investigate associations between stress measures and BMI, waist circumference (WC), waist to height ratio (WHtR), and percent fat mass.
Results
General stress was significantly associated with BMI (
= 0.54; 95% CI: 0.04-1.03) and WC (
= 1.44; 95% CI: 0.10-2.77). Parenting distress was significantly associated with BMI (
= 0.16; 95% CI: 0.02-0.31), WC (
= 0.39; 95% CI: 0.04-0.75), and WHtR (
= 0.003; 95% CI: 0.001-0.005). Household chaos was significantly associated with all adiposity measures (BMI: 
= 0.20 [95% CI: 0.08-0.33]; WC: 
= 0.48 [95% CI: 0.21-0.75]; WHtR: 
= 0.003 [95% CI: 0.001-0.005]; percent fat mass: 
= 0.29 [95% CI: 0.08-0.49]).
Conclusions
General stress, parenting distress, and household chaos are associated with adiposity among parents of young children. Future research should elucidate mechanisms by which this occurs and elucidate how this risk may be mitigated.
Study Importance
What is already known?
- ► Stress is known to increase the risk of overweight and obesity.
- ► Parenting presents many challenges that could increase the risk of stress.
- ► Little is known about how the stresses of parenting are associated with adiposity.
What does this study add?
- ► Perceived stress is associated with adiposity in parents of young children.
- ► This study included both mothers and fathers, which addresses a gap in much of the family health literature.
- ► This study used more comprehensive adiposity measures than typically seen in the literature.
Introduction
In a trend that has been consistent since 2003, nearly 25% of Canadians (6.7 million) aged 15 and older experience chronic daily stress (1). Stress is the body’s response to any factor (environmental, internal, or external) that may threaten or overwhelm the body’s ability to maintain homeostasis (2); these factors are called stressors. Chronic stress exposure can contribute to maladaptive changes to baseline metabolism, which can negatively impact the body’s systems over time. A growing body of research has implicated chronic stress in increased risk of obesity, type 2 diabetes, cardiovascular disease, immune compromise, cancer, and other disorders (3-5). As such, chronic stress is increasingly recognized as an important public health issue.
Chronic stress is hypothesized to influence adiposity through the disruption of the biological stress system and through the promotion of unhealthful behaviors. The negative effect of stress on diet quality, particularly its propensity to increase the consumption of high-sugar and high-fat foods, has been extensively reported in the literature (6, 7). Despite the well-known antianxiety and antiobesity benefits of physical activity, exercise is inversely associated with stress (8). Sleep patterns are also commonly disrupted by stress (9, 10), which further increases the risk of adiposity.
Whereas a substantial proportion of the Canadian population experiences daily stress, parents of young children may be at special risk for high levels of stress because of the many competing demands on parents’ time (11), as well as the potential stress of adjusting to their role as parents (12). However, limited research has examined how stress is associated with obesity in parents (13). The few studies that have examined this association have included only mothers (13, 14). Previous research has suggested that stress may have a differential impact on health outcomes in men and women (15-17); thus, research is needed that explores the stress-obesity link among both mothers and fathers.
In addition to adversely impacting parents’ own health, research has also suggested that high stress may also impact parenting practices. Other researchers have found that parental stress is positively associated with markers of children’s adiposity (18-20) and negatively associated with healthful child feeding practices (21), restricted television viewing, and active play (22). Thus, reducing stress may allow parents to engage in more positive parenting practices and may produce better child health outcomes. Understanding how stress among parents of young children may influence their obesity risk can inform strategies to help parents develop the skills needed to successfully manage current and future stressors, which is paramount to protecting family health.
Thus, the objective of this study was to investigate the associations between three measures of stress (general stress, parenting distress, and household chaos) and adiposity among a sample of Canadian mothers and fathers of young children. This study also examined whether these results differed between mothers and fathers of young children. It was hypothesized that stress would be positively associated with adiposity, regardless of parent sex.
Methods
Study participants
This study used baseline data collected between December 2014 and August 2016 among parents participating in phase 1 and phase 2 of the Guelph Family Health Study, a pilot randomized control trial of a home-based obesity prevention intervention (23). Families were eligible to participate if they had at least one child aged 18 months to 5 years, resided in Wellington County, Ontario, Canada, and had a parent who could respond to questionnaires in English. Of the Guelph Family Health Study cohort of 151 parent participants from 86 families, 41 participants were excluded because of the following: missing data (n = 10), pregnancy or breastfeeding (n = 27), and not disclosing income, which was used as a covariate (n = 4). This yielded a final analytic sample of 110 parent participants (49 mothers and 61 fathers) from 70 families. The study was approved by the University of Guelph Research Ethics Board (REB14AP008). Written consent was obtained from all participants.
Stress measures
We assessed three different types of stress via online (n = 50) or paper (n = 60) surveys. General stress was measured in each participant using the question “Using a scale from 1 to 10, where 1 means ‘no stress’ and 10 means ‘an extreme amount of stress,’ how much stress would you say you have experienced in the last year?” (24). Parenting distress was measured in each participant using the 12-item parental distress (PD) subscale of the Parenting Stress Index (PSI) (25); the standardized Cronbach α was 0.86 in this sample. The PD subscale has been shown to correlate highly (r = 0.92) with the PSI (25) and has demonstrated good internal consistency among both mothers and fathers (26, 27). On a 5-point Likert scale from 1 (strongly disagree) to 5 (strongly agree), participants responded to statements such as “I often have the feeling that I cannot handle things very well,” “I feel trapped by my responsibilities as a parent,” and “Having a child has caused more problems than I expected in my relationship with my spouse (or male/female friend).” For parents who completed the paper survey, the response options included a 4-point Likert scale (i.e., the “neither disagree nor agree” option was not included). To address this discrepancy in response options between the online and paper surveys, we coded the paper survey responses as 1 = strongly disagree, 2 = disagree, 4 = agree, and 5 = strongly agree. When analyzed independently, the online survey data results were consistent with when paper and online survey data were combined. A total score out of 60 was obtained by summing the responses, with higher scores indicating greater parenting distress. Home environment chaos was measured using the 15-item Confusion, Hubbub, and Order Scale (CHAOS) (28), for which the standardized Cronbach α was 0.88. This scale quantifies chaos as the frustration, disorganization, and noisiness of a home. On a 4-point Likert scale from 1 (very much like your own home) to 4 (not at all like your own home), one parent from each family responded to statements such as “There is very little commotion in our home” and “We can usually find things when we need them.” A total score out of 60 was obtained by summing the responses; higher scores indicated greater home chaos. This survey was given only to parent one (defined as the first parent to sign up for the study, of whom 80.6% were mothers), and the same score was applied to the family’s second parent, when applicable.
Adiposity measures
Body composition was examined using the anthropometric measures waist circumference (WC), BMI, and waist to height ratio (WHtR), as well as using air displacement plethysmography (BOD POD, COSMED Inc., Concord, California). Height was measured using a wall-mounted stadiometer with participants barefoot or in socks. Height was measured at the apex of the inhaling breath to the nearest 0.1 cm. Two measurements were taken; if measures differed by more than 0.5 cm, a third measurement was taken. The final data point was the average of the two closest measures. Weight was measured in kilograms using the BOD POD digital scale, in which high reliability meant that only one measurement was needed (29). WC was measured to the nearest 0.1 cm using a Gulick II measuring tape (Country Technology Inc., Gay Mills, Wisconsin). In accordance with Statistics Canada and National Health and Nutrition Examination Survey recommendations (30, 31), the measurement was taken at the top of the iliac crest over bare skin or thin clothing during the pause after expiration and before inhalation. Two measurements were taken; if measures differed by more than 0.5 cm, a third measurement was taken. The final data point was the average of the two closest measurements. One father’s WC value was in excess of 3 SDs above the sample mean; it was removed from WC analyses as an outlier. WC was not measured for one father. Thus, there were 108 WC observations available (60 mothers and 48 fathers).
BMI was calculated by dividing weight (kilograms) by height (meters squared). One father’s and one mother’s BMI were identified as outliers (exceeded the sample mean by more than 3 SDs) and were removed from BMI analyses. Data were not available for one mother (scale calibration error) and one father (declined to be weighed), and so there were 106 BMI observations (58 mothers and 48 fathers). WHtR was calculated by dividing WC (centimeters) by height (centimeters). WHtR is complementary to WC because it is adjusted for height and improves the accuracy in predicting disease risk (32-34). WHtR was not available for one father because his WC was not measured, and so the WC analytic sample consisted of 109 observations (60 mothers and 49 fathers).
Body composition was assessed in 108 participants using a BOD POD, an air displacement plethysmograph. The BOD POD measures body volume, calculates body density, and uses the Siri equation (35) to convert body density to fat mass. Raw body volume was measured with participants wearing tight-fitting clothing, such as a bathing suit and a bathing cap, to minimize air trapping, with jewelry and glasses (all that which is not “self”) removed. Participants were instructed to sit quietly, limit movement, and breathe normally while in the test chamber. Thoracic gas volume was either measured (n = 79) using a single-use tube connected to the rear of the test chamber or, for those participants unable to complete thoracic gas volume testing, calculated (n = 29); there were no significant differences in mean volume by method. The entire BOD POD test took approximately 15 minutes per participant. Our reliability testing of the percent fat mass (%FM) measurement showed an intraindividual day-to-day variation of 2.1% and an intraoperator variation of 2.6%, consistent with reference values of 1.7% to 4.5% and 2.7%, respectively (29). The instrument was calibrated twice in the morning of each data collection day, once with the test chamber empty and once with a 49.980-L calibration cylinder in the chamber. No %FM outliers were identified; however, one mother’s data could not be used because of a BOD POD calibration error, and one father declined to undergo BOD POD assessment, and so %FM data were available for 108 participants (59 mothers and 49 fathers).
Analytic technique
Statistical analyses were performed using SAS University Edition version 3.6 (SAS Institute Inc., Cary, North Carolina) (36). Linear regression coefficient estimates (
) and 95% CI were calculated using generalized estimating equations. The generalized estimating equation approach was used to account for the potential correlation between cohabitating participants (37). Independently, each body composition measure was regressed onto each of the stress measures. Models were adjusted for age, sex, family size (total number of people living in the home, in categories of < 4, 4, and > 4), and annual household income (< $10,000-$150,000+). Models initially included parent sex as a potential moderating variable; however, interaction terms were not found to be significant. P ≤ 0.05 was considered statistically significant.
Results
Descriptive data
The average age of mothers and fathers was 35.6 years and 36.7 years, respectively. Most participants were white (83.6 %), married (87.3%), and highly educated (51.2% with a university education or more), and they had a relatively high annual household income (64.3% of families earned $80,000+ per year). Including all adults and children living in the home, 17.1% of families had three members, 38.6% had four members, and 44.3% had five or more members.
As shown in Table 1, parents reported moderately high levels of stress on all three measures; specifically for the PD subscale, parents in this sample scored in the 67th percentile relative to all of the parents examined during the development of the PSI (25). Paired t tests of mother-father pairs revealed no differences in mothers’ versus fathers’ reporting of general stress (n = 38) or parenting distress (n = 39). The mean BMI for both mothers and fathers would be categorized as “overweight” based on the World Health Organization recommendations (38). Also, 77.1% of mothers’ and 51.7% of fathers’ WC measurements exceeded recommendations of 80 and 94 cm, respectively (39). The WHtR for 51.0% of mothers and 64.4% fathers was above 0.5, which indicates greater health risk (33, 34). The mean %FM was 26.3% for fathers and 34.5% for mothers. Although no specific guidelines exist for determining which level of body fat increases health risk, it is generally accepted that %FM of 10% to 22% of body weight for men and 20% to 32% of body weight for women is within healthy ranges (40, 41); 64.4% of mothers and 75.4% of fathers had %FM above these ranges.
| Combined | Fathers | Mothers | |
|---|---|---|---|
| Body composition | |||
| BMI, mean (SD) | 27.7 (5.7), n = 106 | 28.3 (5.3), n = 58 | 26.9 (6.0), n = 48 |
| BMI categories, n (%) | |||
| Normal weight (18.5-24.9) | 42 (39.6) | 18 (31.0) | 24 (50.0) |
| Overweight (25.0-29.9) | 36 (34.0) | 23 (39.7) | 13 (27.1) |
| Obesity (≥ 30.0) | 28 (26.4) | 17 (29.3) | 11 (22.9) |
| WC, mean (SD), cm | 96.4 (14.5), n = 108 | 100.4 (14.1), n = 60 | 91.5 (13.5), n = 48 |
| WHtR, mean (SD) | 0.56 (0.08), n = 109 | 0.57 (0.08), n = 61 | 0.55 (0.08), n = 48 |
| %FM, mean (SD) | 30.0 (10.3), n = 108 | 26.3 (9.4), n = 59 | 34.5 (9.6), n = 49 |
| Stress scores, mean (SD) | |||
| General stress a | 6.6 (2.0), n = 106 | 6.6 (1.9), n = 58 | 6.7 (2.1), n = 48 |
| Parenting distress b | 30.3 (7.3), n = 107 | 30.2 (6.2), n = 58 | 30.4 (8.5), n = 49 |
| Home chaos (family level)c | 32.4 (8.3), n = 67 | N/A | N/A |
- a General stress scale ranges from 1 to 10.
- b Parenting distress scale ranges from 12 to 60.
- c Home environment chaos ranges from 15 to 60.
- N/A, not applicable.
Linear regression of adiposity on stress
Associations between measures of stress and body composition are presented in Table 2. General stress was positively associated with BMI (
= 0.54; 95% CI: 0.04-1.03) and WC (
= 1.44; 95% CI: 0.10-2.77). Parenting distress was positively associated with BMI (
= 0.16; 95% CI: 0.02-0.31), WC (
= 0.39; 95% CI: 0.04-0.75), and WHtR (
= 0.003; 95% CI: 0.001-0.005). Household chaos was positively associated with all measures of adiposity (BMI: 
= 0.20 [95% CI: 0.08-0.33]; WC: 
= 0.48 [95% CI: 0.21-0.75]; WHtR: 
= 0.003 [95% CI: 0.001-0.005]; %FM: 
= 0.29 [95% CI: 0.08-0.49]).
General stress,  (95% CI) |
Parenting distress,  (95% CI) |
Household chaos,  (95% CI) |
|
|---|---|---|---|
| BMI | 0.54 (0.04 to 1.03),* n = 103 | 0.16 (0.02 to 0.31),* n = 104 | 0.20 (0.08 to 0.33),** n = 102 |
| WC | 1.44 (0.10 to 2.77),* n = 104 | 0.39 (0.04 to 0.75),* n = 105 | 0.48 (0.21 to 0.75),** n = 104 |
| WHtR | 0.007 (−0.001 to 0.014), n = 105 | 0.003 (0.001 to 0.005),** n = 106 | 0.003 (0.001 to 0.005),** n = 104 |
| %FM | 0.30 (−0.61 to 1.21), n = 105 | 0.22 (−0.01 to 0.45), n = 106 | 0.29 (0.08 to 0.49),** n = 103 |
- a Models adjusted for sex, age, household income, and family size.
- * Significant at P < 0.05 level.
- ** Significant at P < 0.01 level.
Discussion
This study is one of few in the literature to investigate associations between stress and adiposity among both mothers and fathers of young children and the first to do so among Canadian families. As hypothesized, we found that stress was positively associated with adiposity, independently of parent sex. Our results suggest that multiple dimensions of stress may be important determinants of adiposity among both mothers and fathers of young children. The moderately high level of stress reported by these parents underscores the importance of examining the influence of stress on health among parents of young children. The consistency of our results with multiple measures of stress suggests that multiple dimensions of family life may lead to increased obesity risk among parents.
Our findings that general life stress was positively associated with markers of adiposity are consistent with other research. Among a sample of Canadian adults, Sampasa-Kanyinga and Chaput (2) found that perceived life stress and BMI were positively associated and that these associations were independent of sleep and physical activity. Similarly, a review by Björntorp (4) presented substantial literature linking chronic stress, assessed by cortisol, with several adiposity outcomes, including BMI, waist-hip circumference ratio, and sagittal abdominal diameter, among adults. However, less research exists that focuses specifically on parents of young children. The few studies that have examined parents’ stress and adiposity have found positive associations between mothers’ perceived stress and BMI (13, 14). Our research extends this work by including fathers and by exploring associations with additional, more comprehensive measures of adiposity.
Our PD subscale results show that, on average, participants feel uneasy in their role as parents and that this stress is associated with increased adiposity, as assessed via BMI, WC, and WHtR. Most of the existing research that explores the PSI scale or other measures of parenting distress has focused on mothers only and investigated families in clinical populations, such as children with autism spectrum disorder, attention deficit hyperactivity disorder, diabetes, and cystic fibrosis. (42, 43). This study helps to fill these gaps and provides insight into the need for parenting distress reduction supports among nonclinical samples, as evidenced by the mean PD scores of our community-based sample of parents of young children; in addition, this study brings attention to the connections between parenting distress and adiposity in both mothers and fathers.
In addition to the mean CHAOS scores indicating moderate household disorganization, the associations with all four markers of adiposity investigated here suggest that the home environment plays a substantial role in parental health. Whereas several studies have explored the association between chaos in the home and child weight outcomes, few studies have focused on parents’ health. In a study of family dynamics and BMI in families with children aged 12 to 17, Cyril et al. (44) found that family functioning, which is a measure of the quality of interactions between members of a family, was inversely associated with child BMI, but no associations were found with parent BMI. Similarly, a mother-child study by Payas et al. (45) also found no associations between family functioning and mothers’ BMI. This discrepancy between our findings and those of studies exploring family functioning suggests that CHAOS is measuring a distinct construct, and future research should explore the mechanisms by which CHAOS in the household influences obesity risk.
Although this study has many strengths, some limitations should be considered when interpreting our results. This study used a single-item assessment of general stress; the use of a more comprehensive measure of general stress likely would provide a more accurate reflection of general stress in this population. The level of household chaos was reported only by the first parent to register for the study (88.9% mothers). Although it is possible that fathers may have perceived household chaos differently than mothers, the fact that fathers and mothers reported similar levels of general stress and parenting distress suggests that overall stress levels are similar between mothers and fathers. These measures of stress were all based on self-report and therefore subject to biases in ways that a physiological measure of stress, such as cortisol levels, would not be. In addition, this sample was relatively small and contained a large proportion of white individuals of high socioeconomic status, which may limit the generalizability of our findings. As with all cross-sectional research, disentangling the causal component in these associations was not possible.
Despite these limitations, this study extends previous findings reported in the literature. The inclusion of both mothers and fathers addresses a gap in the existing literature (12-14). Another strength of this study is the complementing of BMI with other measures such as WC, WHtR, and %FM from an air displacement plethysmograph. This comprehensive approach allowed for different measures of obesity to be investigated without relying on BMI, a common proxy measure of obesity.
Conclusion
Overall, our study found that stress is associated with adiposity in parents of young children. General stress, parenting distress, and stress in the home environment may be important determinants of parental weight outcomes. Future prospective research with larger and more diverse samples are needed to investigate the mechanisms by which stress leads to obesity. Elucidation of these associations is needed to inform stress reduction strategies for reducing both parental stress and adiposity.
Clinical trial registration
ClinicalTrials.gov identifier NCT02223234.
Funding agencies
This research was funded by the Health for Life Initiative at the University of Guelph; however, it had no role in the project design, data collection, analyses, interpretation of data, or writing of the manuscript.
Disclosure
The authors declared no conflict of interest.
Author contributions
VH conceived of the analyses, conducted and interpreted all analyses, and wrote the manuscript. TA assisted in data collection and compilation. GD was the statistical advisor for this project and assisted with reviewing and revising the manuscript. ACB and JH served as coadvisors for this project, including in interpretation of the data and in reviewing and revising the manuscript. DWLM and JH are the codirectors of the Guelph Family Health Study and oversaw this project. All authors read and approved the final manuscript.
