Introduction

In recent studies, the importance of sleep health in chronic disease development and management has been emphasized [1, 2]. There is growing evidence that the duration and/or quality of sleep and the development and management of many chronic diseases, such as hypertension, diabetes, coronary heart disease, kidney disease, and obesity, are related [1,2,3]. Individual and environmental factors as well as reduced sleep quality are reported to be the major factors causing weight gain. A large number of studies indicate insufficient sleep duration and quality as important risk factors for obesity worldwide [4,5,6]. Various meta-analyses and review studies have reported that there is a negative correlation between sleep duration and body weight. However, in a critical review of the epidemiological evidence it is unclear from the available adult epidemiological literature whether either short and/or long sleep is associated with obesity or weight gain [7].

Assessing the association between sleep quality and food intake is important because while dietary factors that affect sleep quality can be easily improved, altering sleep quality autonomously [8]. However, there exist few studies evaluating the relationship between sleep quality and dietary intake [8, 9].

Our primary aim was to determine the relationship of specific macro/micro nutrients and food groups with sleep duration and quality in adults.

Methods

Study population

This cross-sectional study was conducted between July and September 2017 in 20–64 year-old 3262 adults, who were randomly selected using a multi-stage stratified quota sampling method in Turkey.

The study data were collected via a questionnaire including general socio-demographic characteristics (age, gender, income status, education, occupation, smoking, alcohol), physical activity status, sleep characteristics and 24-h dietary recall collected by the researchers with face-to-face interview method. Volunteers were interviewed face-to-face and a signed written consent form was obtained from each participant.

Individuals who met the following criteria were excluded from the present study: life-threatening illness, neurological disease, diagnosed phsychological disorder or sleep disorders, use of medication with known effects on sleep, younger than 20 years and older than 65 years, physical and mental disabilities and refuse to participate in the survey. Participants who did not provide informed consent or did not complete the study questionnaire were excluded from the study.

In this study, a total of 3262 participants have available data of sleep duration. We excluded 188 subjects because of missing or implausible information on PSQI and those 2636 subjects reported a daily dietary intake. Of those we excluded 127 subjects because of missing BMI or waist–hip circumference as well as 63 subjects less than 750 kcal (n = 55) or above 4500 kcal (n = 8). Finally, data from 2446 participants were included in this study (the overall participation rate in the study was 75%). The study was approved by Kırıkkale University Social Sciences and Humanities Research Ethics Committee (dated 18.07.2017, approval # 7).

Sleep duration and quality

The Pittsburgh Sleep Quality Index (PSQI) was used to assess sleep duration and quality. This questionnaire evaluates the individual’s sleep quality for the previous 1 month. The total PSQI score is between 0 and 21, and the sleep quality is “good” if the total score is ≤ 5 and the sleep quality is considered “poor” if > 5 as defined by Buysse [10]. Reliability and validity of the Turkish version has been verified in terms of its validity and reliability by Ağargün and Cronbach’s alpha internal consistency coefficient was determined as 0.804 [11].

In the evaluation of the sleep period, the question “How many hours of night sleep did you get in the last month? (which may differ from the time you spent time in bed)” was used. There are differences in the literature regarding the classification of sleep duration. In the present study, responses to sleep duration were classified in three groups: short: ≤ 6 h, normal: > 6–≤ 8 h, and long: > 8 h) according to previous studies [12, 13].

Assesment of dietary intake

Dietary intake was measured using 24-h dietary recalls (24-h DR). Recalls were performed via face-to-face interviews. A 24-h dietary recall (24 h) is a structured interview intended to capture detailed information about all foods and beverages. A photographic atlas was used to record the type and portion size of the food and meal. The dietary data analysis was analyzed using BeBİS programme. Nutrition Information Systems (Beslenme Bilgi Sistemi-BeBiS), which is a food software program in compliance with Turkish food, was used for assessment of nutrients, food, and food groups (BEBİS 2004). [14].

Anthropometric measurements

Anthropometric measurements (body weight, height, waist, and hip circumferences) of the participants were taken in accordance with proper techniques [15]. Height was measured with a 0.1 cm accuracy while having no shoes on, in a standing position, looking straight ahead and putting the shoulders and back of the feet in one direction. Weight was measured using a scale while having no shoes on, wearing minimum clothing, and after excretion.

Body mass index (BMI) was calculated as weight in kilogram divided by the squared height in meters [16]. The BMI values of the participants were grouped as underweight (BMI < 18.5 kg/m2), normal weight (18.5–24.9 kg/m2), overweight (25.0 ≤ BMI kg/m2) and obese (BMI ≥ 30.0 kg/m2) [17].

Statistical analysis

All statistical analyses were done using a statistical Package for Social Science (Version 22.0; SPSS Inc., Chicago IL, USA) Pearson Chi-square test was used to compare categorical variables. Continuous variables were compared with parametric tests for normal distribution (Independent sample t-test and ANOVA test). In the case of a group difference, the tukey’s test was used to determine which group the difference originated from. Statistical significance level was set as p < 0.05.

Results

The mean age and BMI of the participants were 38.7 ± 12.70 years and 26.4 ± 4.93 kg/m2, respectively. There was a significant difference in age, gender, education, socioeconomic status (for these parameters p < 0.001), smoking (p = 0.008), alcohol use (p = 0.017), PSQI and sub-scales (p < 0.001, p = 0.008 for use of sleep medication). In terms of sleep quality, there is a difference in gender (p = 0.002), education level (p < 0.001), SES (p < 0.001), alcohol use (p = 0.033), WC (p = 0.043), and WC/HC (p = 0.039), sleep latency (p < 0.001), sleep duration (p < 0.001), sleep disturbance (p < 0.001), use of sleep medication (p < 0.001) and day-time dysfunction (p < 0.001).

When energy and macro- and micro-nutrient intake of the individuals were compared, it was observed that long sleepers had lower energy-from protein (%) (p = 0.015) compared to normal sleepers. Saturated fat and retinol intake of short sleepers was significantly higher (p = 0.018, p = 0.029), and PUFA, vitamin E, and thiamine intake was significantly lower compared to normal sleepers (p = 0.024, p = 0.020, and p = 0.036, respectively). It was found that zinc intake was significantly higher in normal sleepers compared to long sleepers (p = 0.019) (Table 2). Consumption of fiber (p < 0.001), beta-carotene (p = 0.003), thiamine (p = 0.024), vitamin B6 (p = 0.023), total folate (p = 0.002), vitamin C (p = 0.002), calcium (p = 0.037), magnesium (p = 0.037), potassium (p = 0.002), and iron (p = 0.017) were found to be significantly higher in individuals with good sleep quality compared to those with poor sleep quality (Table 2).

Consumption of milk and dairy products, meat and processed meat products was found to be lower (p = 0.014, p = 0.026) in long sleepers compared to short sleepers when the consumption patterns of food groups were examined. Fruit consumption was higher in individuals with good sleep quality (193.3 ± 202.01 g/day) compared to those with poor sleep quality (171.4 ± 185.36 g/day) (p = 0.006).

Discussion

To the best of our knowledge, this is the first study involving a large population in which both sleep duration and quality along with the status of individuals’ consumption of food groups as well as specific macro- and micro-nutrient intake are addressed together among Turkey. In the study, some nutrients were found to have a relationship with sleep duration and quality without any significant differences in BMI, hip circumference and regular physical activity status.

Sleep habits are among the probable factors contributing to obesity. The obesity prevalence has been increasing day-by-day due to the increase of working hours and sleep deprivation. Although there are studies indicating that there is an inverse relationship between the duration of short sleep and BMI [18,19,20], there are other studies reporting no relationship [21, 22] or a U-shaped relationship [23, 24] between these two variables. In line with previous studies [9, 21, 22], we did not identify associations between sleep duration and BMI (Table 1). On the other hand, there was no difference between the BMIs of individuals in terms of sleep quality, which agrees with the literature [25, 26]. Differences in eating habits, reduced diet quality, irregular eating habits, and increased energy intake may be considered as possible mechanisms in elucidating this inconsistency [13, 27].

Table 1 Distribution of some main characteristics according to sleep duration categories and sleep quality of the study population (frequency and proportion or mean ± SD)
Full size table

In previous studies, despite the methodological differences, it is indicated that there is a relationship between short sleep duration or irregular sleep and unhealthy eating habits [28]. Although there exist studies indicating that there is a relationship between intake of energy and sleep duration [6, 19, 29,30,31,32,33] there also exist studies reporting that there is no relationship between the two, as is the case in the present study [9, 12, 33, 34].

Overall, the sleep duration was associated with energy from protein (%) intake, sleep duration/quality, but not significant difference in carbohydrate and total fat (Table 2). The relationship between the intake of macronutrients and duration/quality of sleep in adults has been reported in a number of studies [8, 9, 12, 13, 31, 34, 35] but results are largely inconsistent. Some studies were found where there was significant relationship between sleep duration and intake of carbohydrate [31, 34] and protein [12, 31, 34]. In Bavarian adults, there was no significant difference between sleep duration, sleep quality, and sleep mid-point and carbohydrate, protein consumption [9]. As seen in the literature, the relationship between carbohydrate and protein intake and sleep duration is complex. Therefore, there are studies pointing out that not only the amount of carbohydrate but also the type of carbohydrate and the glycemic index may affect sleep quality [8, 36].

Table 2 Comparison of mean energy, macro- and micro-nutrient intake/day according to sleep duration categories and sleep quality of the study population
Full size table

In the literature, there is no clear result about the relationship between fat intake and sleep duration and quality [9, 12, 13, 29, 37,38,39]. Our findings were similar with some previous studies suggesting sleep duration or quality was not associated with percentage of energy from fat intake [9, 12] but some studies found relation between fat intake and sleep duration [13, 29, 37,38,39]. However, when the fatty acids were examined, it was found that saturated fat intake of short sleepers was significantly higher, and their PUFA intake was lower than long sleepers. In another study a positive correlation was found between saturated fat and sleep duration [38]. In a review of many studies, similar to the results in this study, a significant relationship was reported between short sleep duration and increased SFA intake [35]. Short sleepers may have more tendency towards unhealthy food choices and an imbalance in their nutritional pattern may explain the higher SFA and the lower PUFA intake.

Available information on micronutrient intake after sleep restriction is controversial. This study indicated significant relationship between sleep duration and the dietary intake of certain micronutrients and there was significant relation between sleep quality and most of micronutrients (Table 3). However, short and long sleepers’ vitamin E, B1, and zinc intakes were found lower than normal sleepers, but retinol intake was found higher in short sleepers. In line with this study, Grandner et al. [13] found that B1 and E levels were lower in short sleepers because B vitamins act in the regulation of the release of melatonin, which affects sleep [28]. When the micronutrient intake was evaluated in terms of sleep quality beta-carotene, vitamin E, total folic acid, vitamin C, potassium, and iron intake were found to be significantly higher in the subjects with good sleep quality than those with poor sleep quality. In this study, high levels of fruit consumption in individuals with good sleep quality may account for higher beta-carotene, vitamin C, total folate, and potassium levels because whole grains, vegetables, and fruit are rich in fiber [40]. Also, vegetables and fruits are rich in vitamin C and potassium [41]. There are studies in the literature focusing on the relationship between sleep quality and macronutrient intake and food groups [8, 9], but there exists no study evaluating the relationship between sleep quality and micronutrient intake. Peuhkuri et al. [42] also emphasized that the relationship between nutrient intake and sleep duration and quality is unclear in the literature. The mechanisms by which the effects of micronutrients on sleep are not clear [27]. For this reason, it is considered that this study will make a significant contribution to the literature.

Table 3 Comparison of mean daily food groups and beverages according to sleep duration categories and sleep quality of the study population
Full size table

As a result, possibly there are various mechanisms that stimulate the effects of nutrients during sleep. In this study, the complex relationship between dietary intake and sleep was assessed. Relationships between some macro- and micro-nutrients of the diet and sleep duration and quality were found. Although there seem to be minor differences in vitamin- and mineral content in terms of sleep duration and quality, these small quantities are important in protecting against diseases. It is difficult to say that consumed foods directly affect sleep because it is unclear via what mechanisms nutrients and nutritional elements regulate sleep. Development of various nutrition strategies may be an important solution for the prevention of sleep disorders. For this reason, considering all these factors, providing adequate and balanced nutrition training to the individuals by nutritionists will contribute to the improvement of sleep quality.

Limitations

Studying over large sample sizes brings some limitations. The most important one is that the result of a hypothesis test may demonstrate a significant difference. In this case, it is important to report effect sizes which is independent of sample size besides p values. For this reason, we added effect sizes of all results according to Cohen’s guidelines.”

In the present study, self-reports were taken into account when assessing sleep duration and dietary intake. In terms of dietary intake, a single administration of 24-h is unable to account for day-to-day variation, two or more non-consecutive recalls could better reflect to estimate usual dietary intake distributions.

Objective assessment methods such as polysomnography and actigraphy were not utilized in this study to measure sleep duration. Another limitation is that leptin and ghrelin, which are hormones of appetite that are important factors in sleep duration, were not evaluated in this study. The evaluation of these hormones may make an important contribution to future studies.