Introduction
Duration of sleep time varies among students in different countries [
1]. Recent studies reported that insufficient sleep duration is a key public health issue as an indicator of developing unhealthy dietary habits and obesity in adolescents [
2]. Adolescent development in many countries is accompanied by surfing the internet or watching TV for long periods [
3]. Other factors responsible for the decline in sleep duration include artificial light, caffeine use, and parental attitudes [
4]. Childhood and adolescence are critical life periods because lifelong habits are formed during this period [
5]. Insufficient sleep is associated with increased risks for obesity [
6,
7] via alterations in metabolic regulation and appetite [
8] and other metabolic disorders including impaired insulin resistance, impaired glucose tolerance and type 2 diabetes [
9], and coronary heart disease [
10]. An inverse relationship was reported between sleep duration and hypertension in adults and boys ages 11 to 14 years old [
11] as well as waist circumference (WC) and hypercholesterolemia in Chinese school children and adolescents 6 to 20 years old included in a cross-sectional study [
12].
Obesity, altered metabolic response, and poor nutritional quality are partly mediated by some features of eating behaviors such as late eating time and skipping breakfast [
13]. The cause-effect relation between diet and these sleep-related metabolic phenomena is currently unknown. Emerging evidence has revealed that poor sleep quality and quantity result in increased food intake in both adults and children. Increased wakefulness with consequently higher energy expenditure drives an increased intake of food [
14]. Moreover, individuals with less sleep tend to ingest foods with more fats or refined carbohydrates [
15]. Evaluating the association between dietary nutrients and sleep among women showed that there was a positive relation between sleep acrophase with vitamin D and a reverse association between less sleep and more fat intake [
16]. Other specific dietary nutrients that disrupt sleep quality are inadequate intake of selenium, calcium, and vitamin C [
17]. In addition, lower consumption of certain foods (e.g., fruits and dairy products) can affect quantity and quality of life negatively [
18]. The underlying mechanisms by which insufficient sleep may increase food intake are not fully understood. However, a review of the literature suggests that both homeostatic and non-homeostatic factors, namely cognitive, emotional and behavioral, may influence food intake as mediators [
14].
As mentioned earlier, there is a relation between disrupted sleep, emotional factors such negative effects and emotional stress, and food intake. Sleep regulates emotions, modulates affective neural systems, and reprocesses recent emotional involvements [
19]. Thus, insufficient sleep leads to dysregulation of emotions in both adult [
20] and pediatric populations [
21]. In many cases, eating especially sweet or energy-dense foods increases calmness and feeling of satisfaction [
22]. Higher scores of emotional and external eating behavior were observed in healthy women with inadequate sleep using the Dutch Eating Behavior Questionnaire [
23]. However, the exact cause-effect relation between sleep disorders, emotional eating behavior, and food intake has not been clarified.
Discussion
Considering that poor sleep is associated with impairments in emotion, motivation, and cognitive functioning in addition to greater risk for medical conditions like obesity, diabetes, and cardiovascular disease, which lead to financial and non-financial costs in community, sleep problems have become of global public health concern [
32]. Unfortunately, many adolescents experience poor sleep due to biological changes in the sleep/wake cycle as well as social, recreational, and academic pressures. Overall, the consequences of insufficient sleep may include an adverse effect on the control of emotions, attention, and behavior that can be associated with health-risk behaviors [
33]. As adolescence is a life period in which the future adult health pattern is established, a focus on adolescence is vital to the success of various public health agendas [
34].
The purpose of this study was to investigate the relations among sleep quality, emotional eating behavior, and energy/macronutrients intake in female school-age students and to explore the mediating role of emotional eating underlying the relation between sleep quality and energy/macronutrients intake. Males and females sleep differently because of sex differences, largely physiological and biological differences [
24]. We aimed to include only females in the current study because of their higher prevalence of sleep difficulties and emotional eating.
Based on the results of the study, more than half of the participants were poor sleepers. The study demonstrated that even though PSQI could predict energy intake directly, sleep quality, emotional eating, and energy intake were not interrelated and that emotional eating did not act as a mediator between sleep quality and energy intake in this study group. Notably, in spite of the direct effect of poor sleep quality on fat intake, a mediating effect of emotional eating related to sleep quality and macronutrients intake was not established.
The results showed that 75.3% of the students had a PSQI score above the cutoff point (> 5). It seems that some degree of this problem exists in several countries [
35,
36]. We used the PSQI global score and its seven individual components to examine the sleep pattern of adolescents. This instrument
’s validity was examined in different age groups [
29,
37,
38]. PSQI measures sleep quality subjectively, which leads to the question of whether subjective methods are valid and reliable as objective methods. Determination of the correlation between objective and subjective sleep evaluation has shown varying results [
39,
40] . Overestimation of total sleep time by children and adolescents has been reported when measured subjectively compared to objective measurement of sleep [
41]. As a whole, however, while subjective measurement of sleep quality may not detect changes in details, it does discriminate between poor and good sleepers and detects changes in sleep quality that are clinically important [
42].
The EEQ was used to assess to what extent emotions affect emotional eating. In the present study, a significant difference in emotional eating behavior between the two groups of poor and good sleepers was observed. A significant relation between poor sleeping and emotional eating confirmed the observed result (β = 0.26,
p = 0.001). This association means that by increasing the PSQI score, the score of emotional eating elevated as well. This observation is in accordance with Dweck’s study, which found that high scores of emotional and external eating were associated with poor sleep quality [
23]. Thus, disturbed sleep could cause problems in emotional regulation, which could result in a desire to eat more food [
14]. However, in the present study emotional eating did not act as a mediator between poor sleep quality and food intake. Another study indicated that poor sleep quality was associated with an increase in hunger, disinhibition, dietary restraint, and emotional susceptibility to disinhibition whereas sleep duration was not [
43]. Thus, poor sleep quality can be an indicator or marker of problematic eating behaviors.
As mentioned earlier, emotional eating occurs in response to some psychological conditions like depression, anxiety, and loneliness. As a result, subjects with a high emotional eating score will likely suffer from poor sleep quality due to a correlation between emotional stress and sleep [
44]. Previous studies confirm this conclusion. In one study on healthy women in the United States, the emotional and external eating scores were significantly higher in poor sleepers, and subjects with higher scores of emotional eating behavior were more likely to sleep for shorter times and have higher food intake [
23]. However, Lotfi et al. found a trend close to significance in eating behaviors between groups of poor and good sleepers; a positive relation between the score of different eating behaviors and the sleep score was observed [
18]. As a conclusion, it is suggested that there is a reciprocal linkage between sleep quality and emotional eating, which can affect food intake, healthy behavior, and body weight. In our study, although the overall effect of sleep quality on energy intake was statistically significant, there was not a causal link between the sleep quality, emotional eating, and energy intake of female students, which might be because poor sleep quality displays as a stressor or results in disinhibition of eating and directly causes the emotional eating as well as increased energy intake.
Eating patterns may be different along with altered sleep quality in which the characteristics of the subject play a role. In this study, poor sleep quality correlated with fat intake (β = 0.18.
p = 0.013). Inadequate sleep for several days under laboratory conditions shifted dietary patterns to unhealthy choices via physiologic alterations in leptin and ghrelin levels and increased the subjective rate of appetite, with morning cravings for high fat, high carbohydrate foods [
45,
46]. Our data are consistent with experimental findings with regard to higher fat intake for adolescents who had higher PSQI score. Although it was assumed that eating patterns are influenced by sleep duration, the relation between sleep duration and eating patterns is likely to be bidirectional [
8]. This means that high fat intake might alter sleep duration through reduced endogenous lipid synthesis, which delays the phosphorylation of e1F2-α. It has been suggested that P- e1F2-α is a signal for sleep. As a result, it was assumed that eating patterns are influenced by sleep duration, the relation between sleep duration and eating patterns is likely to be bidirectional [
8].
Studies that have focused on adolescents have revealed different results. Our result is in agreement with Weiss et al., who found that adolescents sleeping less than 8 h had a higher intake of fats and a lower intake of carbohydrates [
2], which may be indicative of choosing snacks rather than healthy foods by students with poor sleep quality. This finding is supported by Rangan et al. [
47], but not by two other studies in which girls with poor sleep quality tended to consume carbohydrate-rich foods [
47,
48]. It seems that negative emotion may be responsible for the selection of sweet and high fat foods with more palatability [
49,
50]. The present study could not confirm emotional eating as a mediator between poor sleep quality and more energy intake from fat, which is in contrast with Dweck et al.’s study that reported short sleep duration resulted in more food intake in subjects who were emotional eaters [
23]. Limited studies have examined the causes of these discrepancies. Disruption of the circadian clock [
51], which dysregulates the leptin level [
52], could be one main reason. Another factor might be using different tools to evaluate sleep quality, emotional eating, and food intake.
As a whole, there are several reasons for overeating among individuals with poor sleep quality or quantity other than emotional factors, which did not serve as a mediator in this study. Time and type of eating is different in insufficient sleepers and normative sleepers [
23]. Subjects who go to sleep later at night have more time to eat and, because of an obesogenic environment, they increase the amount of snacking in comparison to eating during routine meal times [
4]. Other underlying factors that link disturbed sleep to food intake are homeostatic (imbalance in leptin and ghrelin levels), cognitive (disruption of reward sensitivity and inhibitory control), and behavioral (difficulty in control of impulsive behavior) [
14].
To our knowledge, the present study is the first to focus on the mediating effect of emotional eating on the relation between sleep quality and food intake in Iranian female adolescents. However, previous studies have explored sleep quality among students [
1,
53] and the relation with food intake [
18,
48]. The current study has several limitations. The first is that our study sample consisted of female students only and that the source of enrolment was limited to one area. Therefore, the results of this study are not generalizable to other populations and further studies are needed on males and other age groups, as well as in different socioeconomic areas with more participants, since a small sample size can influence the power of the study and result in nonsignificant data. Secondly, we used SFFQ to measure food intake. This instrument may be flawed because of its dependency on memory. Even though it measured food intake, it does not assess cravings and hunger, which are likely to have an association with sleep deprivation. In addition, an assessment of stress over time may clarify the association between emotional eating and food intake. This cross-sectional study was carried out in autumn, a period of time in which students are going to school, which could affect the results. It would be better to analyze the sleep quality during both summer holidays and school periods. Moreover, we did not measure the sleep quality objectively using a measurement such as actigraphy. The subjectivity of the questionnaires may result in under/overestimation of data by participants, consciously or unconsciously. However, PSQI and EEQ are more valid and reliable compared to other questionnaires [
27,
54] for determining sleep quality and eating behavior, respectively.
In conclusion, poor sleep quality, emotional eating, and energy/macronutrients intake were not interrelated in female students. The findings confirmed that poor sleep quality had a direct effect on energy intake and the proportion of calorie intake from fat without a mediating effect of emotional eating. However, there was a correlation between poor sleep quality and emotional eating. These findings suggest that poor sleep quality may be related to increased energy consumption and unhealthy eating behavior. Sleep hygiene should be promoted in adolescence to prevent obesity in the future. Some suggestions include regular sleep-wake schedules, use of practical instructions to change the incorrect sleeping habits, and regular physical activity and exercise. Due to conflicting findings when sleep is assessed objectively or subjectively, simultaneous measurements are needed. Examining the environmental risk factors of sleep disturbances will help to discover a more detailed relation between sleep and associated problems.