In this cross-sectional, clinic-based cohort of youth with overweight and obesity, there were expected increases in % total body fat with increasing classes of obesity. While the individuals with severe obesity were slightly older, the differences were not statistically significant. Similarly, youth with severe obesity had a higher prevalence of co-morbidities and including, asthma, OSA, depression, presence of multiple cardiometabolic risk factors, low HDL, elevated HOMA and elevated CRP. Psychosocial and behavioral phenotypic differences were most compelling and possibly interrelated, including higher rates of emotional eating, eating when bored and perceived stress with increasing severity of obesity, and lower participation in organized sports as well as less maternal education. Understanding the underlying biology and interrelationships between these psychosocial and behavioral factors may help us to develop more effective prevention and treatment interventions to improve the health of youth with severe obesity.
The adverse levels of HDL-C, CRP and HOMA in individuals with severe obesity are consistent with findings from other studies. Lower levels of HDL-C have been identified in youth with severe obesity from population-based NHANES data at younger ages prior to identifiable differences in other lipid parameters.5 Elevated levels of CRP, considered a marker of inflammation that has been associated with obesity in adults,21 and is thought to herald the onset of cardiometabolic consequences, such as coronary heart disease,22 hypertension,23,24 metabolic syndrome25,26 and diabetes27 was significantly higher in the youth with severe obesity compared to those without severe obesity. While such levels have been previously identified in adults, a similar association has not been well defined in children.21 The presence of elevated CRP and HOMA in youth with severe obesity adds to the growing body of evidence that supports an urgent call for intervention to prevent the potential oncoming wave of cardiometabolic disease in future generations.
Similar to other studies, there was a higher proportion of asthma, depression and obstructive sleep apnea in individuals with severe obesity. In a systematic review, Papoustakis et al noted the association of asthma in 30 out of the 31 cross-sectional studies and 12 out of the 13 prospective studies of individuals with obesity.28 While the exact mechanism of this association remains to be elucidated, proposed mechanisms include systemic inflammation and mechanical effect, both due to expansion of the adipose tissue. There may also be a contribution of insulin resistance and a role of intestinal microbial dysbiosis causing higher lipopolysaccharides or other inflammatory agents.29,30 The association of depression with severe obesity has also been described, albeit without a clear directionality. One representative study of adolescents with obesity found that after adjusting for demographics and emotional eating, the odds of having severe obesity versus obesity were 3.5 times higher for patients with depression (as measured by PHQ-9; a score ³ 11 was considered depression) compared with those without (OR = 3.5; 95% CI 1.2,20.9, p = 0.030).31 Similarly, the odds of having severe obesity were also higher with anxiety (Generalized Anxiety Disorder Scale ³ 10), OR = 4.93, 95% CI = 1.17,20.85, p = 0.030). In this study, there was no association of either depression or anxiety with emotional eating, but other studies have found emotional eating as a mediator for obesity in individuals with depression. 32,33 A high level of psychosocial dysregulation along with problematic eating patterns and behaviors was noted by Gowen et al in their study of 54 young adults and adolescents.34 Similarly, in a comparable group of adolescents prior to bariatric surgery, higher BMI, depressive symptoms and the number of medical comorbidities were significant predictors of the impaired weight related quality of life and closely related to dysregulated eating behaviors.35 It is also postulated that the presence of depression elicits a response from the hypothalamic adrenal axis that leads to inflammatory milieu leading to obesity.36
Notably, this study found that emotional eating and eating when bored was increased in individuals with severe obesity. A body of literature demonstrates the association of emotional dysregulation with weight gain and obesity. 12,31,37,38 It has been suggested in adults, that a poor emotional regulation may entail the use of maladaptive strategies to manage emotions and stress, for example, by using highly palatable and energy dense food to suppress emotions.39 Studies in children have demonstrated that emotional eating is often followed by negative emotions, 40–43 and is likely to be a learned behavior. 44,45 College students endorsed boredom as the emotion that most commonly triggers eating 46 that was replicated in the laboratory where the completion of a boring task was associated with snacking desire.47 Assessment of subtypes of emotional eating in a cohort of 189 adults showed eating in response to depression (EE-D), anxiety/anger (EE-A) and boredom (EE-B) related to poorer psychological well-being, greater symptoms of eating disorder and more difficulties with emotional regulation.48 It is also possible that the association of depression and severe obesity is mediated by emotional eating as demonstrated in college students in Mexico City 32 and Netherlands.33 Findings highlight the importance of building skills for emotional regulation for children and adolescents with severe obesity. 49 Findings for emotional eating could also be explained by experiences of weight stigma (e.g. weight related teasing) that is more likely to be experienced by youth with severe obesity.50
This study highlights the role of stress experienced by youth with obesity, especially in the older age group, that is increasingly being recognized to have a complex, perhaps bidirectional, relationship.51 Studies of hair cortisol concentration, thought to be deposited over longer periods of time, both in adults52,53 and in children54 have been shown to have high associations with obesity. Whether this relationship is linked to biological factors such as higher levels of circulating glucocorticoids or higher response to circulating glucocorticoids that predispose individuals to weight gain, or changes in internal milieu, such as imbalances in glucose homeostasis, is not clear. On the other hand, the higher levels of weight stigma carried by individuals with obesity, mental health issues (e.g. depression, anxiety etc), or physical disorders (e.g. asthma, OSA) can lead to chronic stress and/or higher circulating corticosteroids, making it a vicious cycle. The presence of higher levels of co-morbidities that can lead to increased levels of biological stress as well as the higher perceived levels of stress in this study, emphasize the role of both recognition and management of stress in youth with obesity. The differences in the PEDS-QL in children with severe obesity, as have been noted in prior studies of children with obesity55 were more prominent in the younger age group in this study, perhaps indicating that older children and adolescents may perceive the psychosocial pressures induced by obesity as stress.
Participants with severe obesity exhibited a reduced number of organized physical activities. It is not known whether the inactivity preceded the severe obesity or is a consequence of potential reduction in the agility of the body by excessive weight. However, there is evidence in the literature that lack of physical activity may be linked to the state of emotional dysregulation known to be associated with obesity.8,48 Emerging evidence emphasizing the positive role of team sport participation on longer term mental health, especially in children who have experienced adverse childhood events56 makes this an important consideration in youth such as those in this study. This study found no statistically significant differences in the dietary intake of foods shown to be related to risk of obesity including the servings of water, sugar sweetened beverages, fruits and vegetables, as well as the patterns of meals eaten at home and fast food intake. The lack of differences may be due to the source of the sample, as participants were attending weight management programs and following treatment recommendations provided those programs.
This study is limited by its cross-sectional design as well as recruitment in tertiary care weight management programs that may limit the generalizability of some of the findings. While a thorough assessment of body composition and cardiometabolic risk factors was done, the measurement of dietary intake was limited to self-reported frequency of selected food groups and physical activities. Parent report on surveys for younger children limits our ability to discern the extent to which this represents children’s perceptions on various measures such as quality of life and stress. Further, the small sample size limits the comparison across gender and race/ethnicity groups. However, the comprehensive assessment in the study has provided the opportunity to identify important psychosocial, behavioral and clinical correlates that may play a role in the more effective management of obesity in such a population.
This study is one of the few studies that has tried to comprehensively characterize differences in physiological, psychosocial and behavioral phenotypes in adolescents with severe obesity. The higher prevalence of psychological and behavioral phenotypes in this sample suggests that there are modifiable and possibly related targets of intervention. It also re-emphasizes the need for a multi-disciplinary team including clinician, behavioral psychologist, social worker and exercise physiologist to identify and address multitude of complex problems in these patients. Future larger and longitudinal studies are needed to tease out the causal pathways and interactions between these important modifiable targets of intervention.