The Effects of Food Intake and Eating Behavior on Sleep Quality in Adolescents

Background: Food and eating behaviors can be related to nutrition as well as sleep


INTRODUCTION
There are many factors in the emergence of nutrition-related diseases such as obesity. High stimulant food and beverages, unhealthy nutritional behaviors and night eating behavior affect sleeping related to obesity [1][2][3] . In recent years, there have been serious changes in the body weights and eating behaviors of children and adolescents in Turkey as well as all over the world 4,5 . In the past few decades children and adolescents have been sleeping less because of many social, demographic, and other numerous factors as food and drinks (e.g., energy drinks, high caffeinated drinks) [6][7][8][9][10] . Recommended nighttime sleeping duration is nine hours for adolescents 11,12 . Many cross-sectional studies emphasized that shorter sleeping duration or sleeping problems were related to higher BMI, obesity and other unhealthy eating behaviors or eating disorders (e.g., anorexia nervosa) [13][14][15][16] . The effect of short sleep duration depends on nutrition, and it can be explained by several factors, including increased sympathetic activity, elevated cortisol and ghrelin levels, decreased leptin and/or impaired glucose tolerance on obesity [17][18][19][20] . In studies [21][22][23] among adolescents according to middle and high school education short sleep duration time associated with overweight and obesity and study results have shown that an obesity risk of 6.5% may occur with a shortening of 1 hour of sleep time. All these results once again reveal the importance of sleep quality in adolescents.
When the relationship between sleep duration and quality and nutrition is examined, it is seen that various macro and micronutrients affect the sleep mechanism 24,25 . Glucose, tryptophan and tyrosine, and essential fatty acids affect sleeping patterns and they should not be forgotten that caffeine is a substance responsible for it [26][27][28] . Consumption of highcarbohydrate foods leads to an increase in postprandial sleep time 29 . Tryptophan, which is found in all proteincontaining foods and has the feature of being the precursor of serotonin, also has the effect of increasing the tendency to sleep 30 . It has been shown that milk, which is one of the foods with the most tryptophan, is more effective than drug treatment in the treatment of sleep disorders. Protein-rich foods make it easier to fall asleep.
Eicosapentaenoic acid (EPA) and docosahexaenoic acid (DHA) synthesized from omega-3 fatty acids, which have a significant effect on the nervous system, are also very important fatty acids in the sleep process 31,32 . Caffeine, which has a sleepdelaying effect, has a moderate stimulating effect on the central nervous system depending on the amount of consumption.
Sleep quality is closely related to healthy eating behaviors, and adolescents who get insufficient sleep tend to consume more high-energy foods obtained from fat, sugar, and refined carbohydrates [33][34][35] . Also, caffeinated drinks or caffeine-rich foods (e.g., coffee, energy drink) negatively affect sleep, delay the process of falling asleep and reduce the sleep duration, thus causing decreases in sleep quality. Caffeinated food consumption use 1 hour before night also reduces falling asleep, sleep duration and restful sleep 36,37 . It has been seen that adolescents frequently consume caffeinated drinks to consciously benefit from this effect on sleep and to stay awake at times, such as during the exam period. The aim of the study is to determine the eating behaviors and sleep quality of adolescents, and the daily energy and nutrient intake.

Study Population and Data Collection
The research was designed as a survey model to examine the sleep quality, daily energy expenditure and nutrient intake and eating behaviors of 14-18-yearold adolescents. The population of the research consists of eight high schools selected by simple random sampling method in 137 high schools, 90 of which are public and 47 private high schools in city center of Konya, Turkey. Population was calculated as one independent means (one groups) applied by power analysis with G*Power 3.1.9.2 software program 38 . The parameters were entered in G*Power as 0.15 for effect length, 0.05 for error (α) and 0.95 for power (1-β) and the sample size was found 580. Using the stratified sampling method, the number of adolescent individuals to be sampled from the three central districts was determined. The layer weight of Selcuklu, Meram and Karatay district is 0.52 (n=300), 0.25 (n=147) and 0.23 (n=133), respectively 39 . The data were collected with questionnaire by using face-to-face interview with adolescents in their schools.

Measurements
The questionnaire form was performed by researchers, and it was prepared with the help of three experts. The draft was pre-applied to 15 adolescents and the necessary corrections were made to the form by researchers and then the finalized form was performed to participants. Questionnaire consisted of four parts and in the first part were demographic variables and anthropometric measurements (e.g., height, weight), the second part was 24-hour recalled food consumption record, the third part used the Pittsburgh Sleep Quality Index (PSQI) scale to determine sleep quality and the last part, Eating Behavior Scale (EBS), was used to determine eating behaviors. Anthropometric measures calculated Body Mass Index (BMI) (kg/m 2 ) and categorized with percentiles according to WHO Adolescents' Growth Standards 40 and Turkey Dietary Guideline (TDG) 41 . BMI percentile values of the adolescents participating in the study were determined by <3 the percentile is "weak"; 3-15 percentile range of "weakness risk"; 15-85 percentile range is "normal;" 85-97. percentile is "overweight;" and >97 percentile is stated as "obese." 40,41 The 24-hour recalled food consumption record method was used and the foods consumed in the last 24 hours were questioned. The PSQI consists of nine questions in total, and it was to evaluate the sleep of adolescents in the last month, and the EBS prepared in the form of a metric scale was applied to determine the eating behaviors by 58 items. Informed consent form was given to the adolescents, and none of them was forced to participate in the study.
Agargun et al. 42  Evaluation for each sub-dimension is made between 0-3 points. In all cases, a score of "0" indicates 'no difficulty', and while a score of "3" was 'difficulty'. Global score between 0-21 points and when the data are evaluated, a score less than 5 is considered as "good," and a score of 5 points or above is as "poor" sleep quality. EBS, which was developed by Ozdogan 43 is a metric scale,thatd indicatesreactions to each behavioral expression by placing a mark on a line 0 to 10 points extending from toward the most negative to the positive. Total EBS score between 0-580 points and at the end of the evaluation four attributes as ≤145 points are considered "poor," 146-290 points are "moderate," 291-435 points are "good," and ≥436 points are "very good." The Cronbach α of the EBS is 0.72.

Statistical Analysis
The data was performed using SPSS (version 22.0) software program. Descriptive statistics like frequency (n), ratio (%), mean (x), standard error of mean (SE) was computed for all one group items. Parametric (e.g., F test) and non-parametric statistic tests (e.g., Chi-Square) were used to examine the differences between predictor variables. The p value was accepted as <0.05 in all statistical analyses. BEBIS (Nutrition Information Systems) software program was used to evaluate the 24-hour recalled food consumption data to calculate the energy and nutrients consumed by adolescents. Only the dinner, snack and night food consumption data were used and calculated of the 24hour recalled food consumption record. The adequacy of daily energy and nutrients consumption was calculated according to age and gender 41 .

Demographic Characteristics and BMI
The study participants included 580 adolescents (mean age 15 Among the adolescents, the rate of those who stated that regular main pattern was found 36.0% in girls and 64.0% in boys and significant difference between the gender (p<0.05). Another finding is that 33.4% of the adolescents skip breakfast, 21.2% lunch and 3.6% skip the evening meal. Breakfast was the most skipped meal in girls (61.8%) and in boys (52.0%). In many studies 50-53 , skipping meals especially breakfast of adolescents ranged from 5% to 55% both sexes. Adolescents' night eating was found to be 16.0%, and boys' night eating levels were higher than girls (p>0.05). When comparing the consumption of caffeinated drinks, boys have similarly higher consumption levels than girls (p>0.05). Especially in recent years, adolescents are turning to caffeinated drinks, e.g., energy drink, coffee for different purposes such as stay awake, work at night. However, a significant difference appears between the groups according to their regular meal pattern (p<0.05). On the other hand, always, sometimes, and never regular meal pattern behavior was found 41.8%, 43.8% and 14.4%, respectively. Girls have higher meal pattern than boys. Hernandez et al. 54 implied that 21% of adolescents were night eaters. Other studies 55,56 represented similar results to explain how eating behaviors were affected by the night eating.  Table 3, as a result of the evaluation of the PSQI scores of the adolescents according to the seven sub-dimensions of the scale, sleep duration, sleep disturbances, daytime dysfunction, and global scores were found significant differences (p<0.05), and due to the night eating situations, sleep disturbances, daytime dysfunction and global score were differed by gender (p<0.05). Raniti et al. 57 highlighted important covariation between poor sleep duration, efficiency, and latency. Şimşek and Tekgül 58 stated 82.0% of adolescents had poor sleep quality. According to the Tetik and Kar Şen's 59 study, 50.3% of adolescents had poor sleep quality. When the EBS scores were examined, it was determined that the boys were higher than the girls (p<0.05), and the scores of the adolescents with night eating behavior were similar (p=0.000). Table 4. Evaluation of PSQI scores of age, BMI, and night eating behavior.

Good (< -5)
Poor ( Table 4 that the highest age was 15 in the good ones and that the similarity was also obtained in the poor. It was determined that those with normal BMI had higher good (63.8%) and poor (66.2%) sleep quality than the others, but no difference was found between them (p>0.05). Overweight (16.0%) and weakness risk (11.6%) groups were determined as the higher after the normal BMI group in both groups. Conversely, it was determined that those who did not have night eating behavior were at the highest level compared to both good (38.1%) and poor (45.9%) sleep quality groups (p<0.05). Cleator et al. 60 reported that poor sleep quality is common is obese group, but night eating is rare.   Table 5, when EBS scores are examined, due to the BMI groups no difference is found (p>0.05), but good and moderate level of EBS is detected higher in normal than others. The same as BMI, in night eating EBS score was differed (p<0.05) and moderate by a long way from poor and good. Unlike PSQI, night eating behavior was determined to be similarly high, and no difference could be determined according to gender (p>0.05). Hunot-Alexander et al. 61 reported girls' higher scores of emotional over-eating, satiety responsiveness, and slowness in eating than boys. When the daily nutrient intakes of adolescents were evaluated, total energy, protein, carbohydrate, B1, B6, niacin, folic acid, B12, Zn, and Fe differed by gender (p<0.05). Protein, carbohydrate, niacin, B12, and Na intakes were higher than the DRI values of Turkey. On the other hand, vitamin D, Ca, K levels were found to be lower than those of Turkey's DRI in both adolescent groups. 41 A study 54 stated that daily energy intake at night is 25% of adolescents. Bartel et al. 62 emphasized an adolescents' study showing high consumption of caffeine shortens sleep durations, delays the start of sleep, increases the number of times an individual awakes during the night, and causes sleepiness during the day. Similar findings were obtained in Tetik and Kar Şen's 59 study. Weiss et al. 63 found short sleeping duration associated with snack consumption and more energy intake at night and similarly our findings overlapped energy, carbohydrates, and lipids. Diethelm et al. 64 carried out a study on European adolescents in eight cities in Europe and found short fatty acid and salt intake was too high and vitamin D, folate, I and F were less than %55 of recommendations and nutritional quality index score was found moderate. Moreover, our nutrient intake findings are in line with the results other European studies [65][66][67] .

CONCLUSIONS
Adequate and balanced diet is essential to be healthy in adolescence. Although nutrition is important in every age group, the importance of nutrition