Central adiposity confers increased risk of mortality from cardiovascular disease, cancer, and all causes among women (31
). This cross-sectional study provides preliminary evidence that school-related stress and CAR interact to play a role in the determination of body fat distribution among Hispanic female youth. The combination of high CAR and school stress was related to unhealthy fat distribution profiles characterized by higher volumes of VAT and SAT, and reduced peripheral fat stores. This is consistent with known effects of cortisol to redistribute fat from the periphery to the abdominal region (5
Our hypothesis was that overall stress would interact with CAR in predicting current abdominal fat, and the use of domain-specific stressful life events was exploratory. Stress has strongly and consistently been related to abdominal fat accumulation in well-controlled animal studies, including nonhuman primates (32
) and rodents (33
) consuming a palatable high-fat diet; however, evidence from human studies is not as clear. For example, a large, prospective study of adults found chronic work stress to be associated with increases in central fat stores over a 19-year period (34
). However, a 5-year prospective study of adolescents (aged 11–15) found perceived stress to be associated with waist circumference in cross-sectional analyses, but not in longitudinal analyses (35
). These discrepancies in human studies may be due to differences in measuring stress. Animal studies typically study objective stress (e.g., tail pinching, restraint), whereas human studies use subjective measures of stress (e.g., perceived stress). Stress is generally defined as the inability to meet or cope with environmental demands (36
). Life events represent more objective measures of stress because they measure events that are generally considered to be stressful, whereas perceived stress measures the subjective evaluation of an event as being stressful (36
Interestingly, of the four domains of stressful life events examined, school-related life events subscale was the only significant predictor of abdominal fat in multivariate models, and this was moderated by CAR. It is notable that the relationship between school-related stress, CAR, and abdominal fat parallels research in the adult literature, where work-related stress has been associated with central adiposity (34
) and CAR (14
). A considerable amount of time is spent at work and school, for adults and children respectively; thus, strain in these domains is likely to have a considerable degree of chronicity. Increased CAR has been noted on weekdays compared to weekends, and appears to be greater in those experiencing stress at work (38
Anticipating stress at school or work may result in higher CAR over time or a greater number of days of high CAR, especially for those with polymorphisms in genes associated with both the CAR and stress (e.g., 5-HTT gene variants) (39
). Thus, gene–environment interactions are a plausible explanation for the interaction of school-related life events, the CAR, and the accumulation of visceral and subcutaneous abdominal fat. In the current study, the interaction of school stress and the CAR revealed that for children with high CAR, but not for children with low CAR, school stress was related to increased VAT and SAT, and decreased percent body fat, after controlling for abdominal fat. A potential explanation is that the overnight stay was stressful for the participants and that participants predisposed to stress-induced increases in the CAR would have elevated CAR during the study. Therefore, girls that experience increased CAR associated with stress may have chronically high CAR when they are consistently experiencing school stress, leading to increases in abdominal fat. Girls that experience increased CAR associated with stress, but do not have chronic stress at school, may not have consistently high CAR and therefore would not have increases in abdominal fat. It would be useful for future research to examine the frequency of each school-related event in order to assess the chronicity of each stressor, and the participant’s appraisal of each stressor. It would also be useful to measure the CAR at home over several school days to determine whether a dose–response relationship exists between school-related stress, the CAR, and central adiposity.
Children reporting greater numbers of stressful school events may be experiencing chronic stress due to academic or social difficulty in school. One of the most frequently reported items on the school-related life events measure was, “Had to study for a big test.” In previous research, exam stress has been associated with total cortisol and weight gain among female nursing students (40
), increases in weight and waist circumference among female medical students (41
), and increases in CAR for women, but not men. Thus, similar to previous studies examining adult women, academic stress may play a role in increased central adiposity among female children with high CAR.
This study was strengthened by the use of precise, well-validated, measures of central adiposity, total body fat, and the CAR, as well as the use of a previously validated subjective measure of stress for urban minority youth. Despite these strengths, there are notable limitations concerning causality, sample size, and study population. This was a cross-sectional analysis, and therefore ambiguity of causal direction is a concern; however, longitudinal evidence in highly controlled animal studies indicates that cortisol and stress lead to subsequent increases in VAT (32
). The most important limitation of this study was the small sample size. Maxwell cautions, “… the tendency to conduct underpowered studies will tend to produce an inconsistent body of literature (42
).” Because the nature of the data collection (i.e., minority children required to stay overnight and provide blood samples) make it difficult to obtain large samples, we used two strategies to increase power and the reliability of our estimates: (i) We used precise measurement techniques (multislice magnetic resonance imaging and BodPod) and well controlled study variables (monitored cortisol collection), and (ii) reported 95% confidence intervals around regression coefficients. Although we recognize that Maxwell recommends the use of confidence intervals around standardized regression coefficients (42
), the standardized coefficients are improper for an interaction term due to centering. Therefore, the interaction was created using the cross-product of the z
-scores, which gives the proper standardized regression coefficients and corresponding confidence intervals in the non-standardized/raw solution (30
The study population was urban Hispanic peripubertal females, either overweight/obese or at potential risk for developing overweight. Racial/ethnic differences in stressful life events were not reported in previous research using this negative life events inventory among urban youth (28
); however, stress may vary by race/ethnicity and location. For example, in schools where educators are not experienced in teaching children speaking another language in the home, or where racial/ethnic diversity is not common, school may be more stressful for children. It would be helpful to examine stressors specific to Hispanic youth in academic settings to understand whether differences in stress contribute to disparities in metabolic health.
We encourage similar studies with longitudinal methods and larger, more diverse samples, in order to delineate the independent and interactive effects of stressors and HPAA activity in the development and proliferation of abdominal fat and nonabdominal subcutaneous fat. These preliminary data are not strong enough to recommend large scale interventions; however, studies utilizing stress-reduction interventions (e.g., stress-reduction guided imagery, reduction in school stress) may be helpful in understanding the malleability of CAR and school stress, and how these mechanisms may lead to changes in adiposity and fat distribution over time.
According to the US Census Bureau (43
), it is projected that the majority of youth in the United States will be Hispanic by 2050 and it is estimated that over half of Hispanic female youth will develop diabetes in their lifetime (21
). Thus, it is important to understand the psychological and physiological mechanisms that lead to diabetes in this important population. This study provides evidence that for Hispanic female youth, school stress may play a significant role in the accumulation visceral fat, a well-known precursor to diabetes.