Association between Dietary Energy Density and Mental Health in Overweight/Obese Women

Objectives : Mental health, sleep quality and dietary intake are interlinked. Impairment of mental health and low sleep quality may contribute to obesity through the consumption of diets high in energy density. Nevertheless, it is not clear whether dietary energy density (DED) influences mental health. This study aimed to examine the association of DED with mental health indices, including depression, anxiety, stress, and sleep quality in overweight/obese women. Results: After adjustment for age, BMI, and physical activity, subjects in the highest quartile of DED had higher systolic and diastolic blood pressure, but lower serum triglyceride, than those in the lowest quartile (p <0.05). DED was significantly associated with increased odds of stress in the crude (OR =2.15, 95%CI: 1.01-4.56, p= 0.04) and adjusted model for age, BMI, and physical activity (OR = 2.56, 95%CI: 1.13-5.79, p=0.02). No significant relationship was observed between DED and depression, anxiety and sleep quality.


Introduction
The prevalence of obesity has progressively risen in all parts of the world during the last decades and has become a main public health concern. It is well-known that the positive energy balance, usually resulted from excessive intake of energy, is the fundamental dietary factor associated with weight gain (1). Assessment of the overall impact of diet is commonly preferred for evaluation of single dietary constituents, such as energy intake. Among diet quality indices, dietary energy density (DED), as a measure of the whole diet, has been at the focus of many recent investigations (2). DED is a comparatively new dietary index that has an important role in body weight control (3), which is defined as the amount of energy per unit weight of a food or beverage, usually reported as kilocalories/100 g (4). High-energy dense diets are rich in fat, because dietary fat provides the greatest amount of energy per gram, but are low in vegetable, fruit and fiber (2,5,6). It has been found that higher DED is associated with the risk of obesity (7) and obesity-related diseases (8), indicating that adopting diets with lower DED are important preventive approaches for obesity-related complications.
Furthermore, obesity is reported to be related to impaired mental health such as anxiety, stress, depression and low quality of sleep (9, 10); and on the other side, mental health, sleep behavior and dietary intake are interlinked (11). There is inadequate evidence in the emerging field of "relationship between dietary intake, mental health, and sleep quality". The majority of studies in this area of research have concentrated on food items, macronutrients, single nutrients or energy intakes (12, 13). Nevertheless, limited research with conflicting results (11, 14-18) exists on the association between DED and indices of sleep quality and mental health. Thus, this study was performed to assed the relation of dietary energy density to mental health and sleep quality in women with overweight and obesity.

Study population
A total of 301 overweight/obese women took part in the present cross-sectional study. All subjects were randomly recruited from individuals referring to health centers in Tehran during 2016-2017.
Inclusion criteria were age 18-56 years, being overweight or obese (body mass index (BMI) ≥25), absence of any acute or chronic infection, no alcohol or drug abuse, no history of hypertension, and not being pregnant. Based on exclusion criteria, prospective subjects with a history of cardiovascular disease, cancer, sustained hypertension, diabetes, thyroid disease, cancer, acute or chronic infections, liver and kidney disease, and smokers were excluded from the study. Enrolment was voluntary, and the participants were informed about the objectives and all the stages of this study.
Written informed consent was obtained from all participants before taking part in the study. The study protocol was approved by the local ethical committee of Tehran University of Medical Sciences (IR. TUMS.VCR.REC. 95-04-161-33893).

Body composition and anthropometric measurement
Body composition of all participants was mustered using a body composition analyzer (InBody770 scanner; InBody, Seoul, Korea) by following the manufacturer's protocol. Weight of the individuals was measured with the use of a digital scale (Seca, Hamburg, Germany) in light clothing and without shoes with precision near to 0.1 kg. Height of participants was evaluated by a seca stadiometer, with exactness close to 0.1 cm in a standing position. BMI was calculated as weight (kg)/hieght 2 (m). Furthermore, Waist (WC) and hip circumference (HC) were measured in the smallest girth and in the largest girth, respectively, with the use of an inelastic tape with accuracy nearest to 0.1 cm.

Evaluation of food intake and DED
To assess dietary intake of participants, a semi-quantitative food frequency questionnaire (sq-FFQ) with 147 Iranian food items, containing a list of foods with standard serving sizes was used. The high reliability and validity and of the FFQ have been confirmed previously (19). All FFQ questionnaires were completed by trained nutritionists. The energy of food consumed was evaluated using Nutritionist 4 software (First Databank Inc., Hearst Corp., San Bruno, CA). DED was calculated by dividing the total dietary energy intake from consumed food (kcal/d) by the total weight (g/d) of consumed foods (excluding beverages) (20).

Measurement of biochemical parameters
Blood samples were obtained from the forearm of all participants in the early morning between 8:00 and 10:00 am after a 10-to 12-hr overnight fasting. The samples were centrifuged at 3,500 rpm for 10 min at −70oC and the separated sera stored at a temperature of −80 °C. Serum concentrations of high-density lipoprotein cholesterol (HDL-C), total cholesterol (TC), triglyceride (TG), and low-density lipoprotein cholesterol (LDL-C) were evaluated by using of enzymatic approaches using related kits (Pars Azemun, Iran) and autoanalyzer system. Insulin level was assessed using an ELISA kit (Human insulin ELISA kit,DRG Pharmaceuticals, GmbH, Germany), and fasting concentration of glucose was measured using glucose oxidase method. Serum high-sensitive C-reactive protein (hs-CRP) was assessed with the use of immunoturbidimetric assay.

Statistical analysis
Analyses of continuous variables to assess differences among quartile of DED were performed using one-way analysis of variance (ANOVA). Analysis of Covariance (ANCOVA) was then used in order to find the difference between the means of investigated variables across quartiles of DED adjusted for age, physical activity and BMI. The logistic regression analysis was applied to find the relation of DED to sleep quality, stress, anxiety, and depression; this model was then adjusted for age, physical activity and BMI. The level of significance was set at a probability of ≤ 0.05 for all tests. All statistical analyses were conducted using a statistical Package for Social Science (Version 22.0; SPSS Inc., Chicago IL, USA).

Results
A total of 301 women, aged 18-56 y, participated in this study and 293 subjects completed measurements. General characteristics of the study subjects are reported in Table 1. The mean age and BMI of participants were 36.39±8.41 and 30.77±3.79, respectively. There was a decreasing trend in serum TG across quartiles of DED (from Q1 to Q4) in the crude analysis and also after adjustment for age, BMI, and physical activity (p≤0.05). Moreover, after adjustment for age, BMI, and physical activity, subjects in the highest quartile of DED had higher systolic (p=0.01) and diastolic blood pressure (p=0.04) than those in the lowest quartile. No significant differences were observed in age, total cholesterol, HDL, LDL, hs-CRP, insulin, HOMA-IR, FBS, depression, anxiety, stress, sleep quality, body composition, and anthropometric indices across the categories of DED (Table 2).

Discussion
This study aimed to investigate the association between dietary energy density, a marker of dietary quality, with mental health of overweight/obese women. After adjustment for potential covariates, higher DED was significantly related to 2.56-fold increased odds of stress. Although, no significant relationship was found between DED with anxiety, depression and sleep quality.
In line with our finding, Heath et al. (23) found that higher levels of stress are associated with a higher energy intake. Under stressful conditions, humans desire palatable foods that are energy dense (24), specially elevated eating of high-sugar, high-fat foods and processed foods (25-28) and a decrease in consumption of main meals, fruits and vegetables (26, 28, 29). Moreover, chronic stress was reported to be related to empty calories (added solid fat and sugar) consumption and evening intake of added sugars (30). The preference of highly palatable, energy-dense food items is attributed to hormones secreted in stress response, such as cortisol (31). Behaviorally, it has been suggested that during stress people have less energy and time to devote to the preparation of foods; thus, they have an elevated dependence on pre-processed convenience food items, which are frequently rich in energy (32, 33). The 'comfort food hypothesis' proposes that chronic stress could endorse a coping strategy resulting in higher intake of macronutrient and preference towards food comprising more carbohydrates and saturated fats (29), which have higher energy density.
The western, high fat-high sugar, and sweet dietary patterns have been reported to be related to higher odds for depression (34)(35)(36)(37); though, some studies did not reveal such a relationship (38). In the study by Grossniklaus et al. (39), increased depressive symptoms, independently predicted increased food and beverage energy density, and explained variance above that explained by male gender, younger age, and reporting adequate caloric intake. There is only one study exploring the relation of DED to depression, which consistent with our study, found no association between DED and depression (40). For anxiety, prior studies found that the saturated fat and added sugars dietary pattern is significantly related to higher anxiety, but in agreement with our results, no association between anxiety level and energy intake was detected (41). Other recent investigations showed that diets high in sugars and fats are associated with higher anxiety level via changes of protein, glucose, and energy homeostasis, and increases in corticosterone and inflammatory cytokines (42,43).
Some studies, in agreement with the current study, found no association between DED and sleep duration (15), but the majority of studies shave identified that because of the elevated feeling of appetite and hunger, people with low sleep quality or short sleep duration have a higher DED or total daily energy intake and preferred food items with high content of carbohydrates and fats, compared with individuals with adequate sleep duration (11, 16). Mechanistically, sleep modulates the pattern of secretion of two key hormones involved in appetite and energy regulation: ghrelin and leptin (44).
Specifically, partial sleep deprivation appears to lead to increased serum ghrelin and reduced serum leptin, both of which result in elevated appetite (45). One likely justification for the lack of link between sleep quality and DED in the present study may be that compared with other study populations, our participants differed in some characteristics. For example, our samples were restricted to women and BMI, as an important factor affecting sleep quality, differed significantly between previous studies and our study.

Conclusion
In conclusion, this study provides the first evidence that DED is significantly associated with stress.
Additional large well-designed studies should be conducted to elucidate the relation of dietary energy density to mental health to reach to a robust conclusion.

Limitation
Some limitations of the current study should be considered. Briefly, this observational investigation might have potential recall bias because of its cross-sectional design, and consequently, causal inferences could not be extracted. There may be some factors, such as socioeconomic or education level of subjects, affecting the DED and mental health that were not measured in the present study.
Demographic and mental health data were self-reported, and these data might be subject to social desirability bias and under or over reporting. Finally, the study population were restricted to women, which this limits generalizability of findings to men; therefore, replication of our results with use of larger samples in both sexes is essential.