Introduction

The prevalence of childhood overweight/obesity in China and other countries is increasing rapidly [1,2,3]. Overweight or obesity in children has been reported to be associated with an increased risk of some chronic diseases such as dyslipidemias, type 2 diabetes, cardiovascular disease and orthopedic problems [4, 5]. Meanwhile, various major risk factors for being overweight or obese have been recognized, including an imbalance between energy intake and expenditure [6, 7], skipping breakfast [8], and poor sleep [9]. Recently, later dinner has been identified as a potential independent risk factor for overweight [10, 11]. Another study reported that Japanese men who went to bed immediately after dinner (less than 2 h) were associated with increased odds for obesity [12]. The dinner-to-bed time concept was proposed by a Japanese researcher, which was found to be associated with some chronic diseases such as colorectal cancer and gastric cancer [13,14,15]. However, there is no study of the effect of dinner-to-bed time on overweight or obesity in children.

At present, the purpose of this cross-sectional study was to examine the relationships between dinner-to-bed time and overweight/obesity after controlling for some potential confounding factors in a study of school-aged children in Ningbo, China.

Methods

Participants

This was a cross-sectional study, which used data from the 2018 Ningbo Youth Risk Behavior Survey (NYRBS). The NYRBS was conducted by the Ningbo CDC to examine the relationships between health-related behaviors and chronic diseases in adolescents. This study was carried out in Ningbo, Zhejiang province, Mainland of China. In brief, a questionnaire survey and a physical examination were conducted in this study. A total of 1739 schoolchildren (8th and 9th grades) aged 14–15 years from 14 middle schools were recruited by using a non-random, purposive sampling in October 2018. We excluded individuals data if they were had unbelievably body mass index (BMI) (< 10 or > 50) or missing body weight or height or meal duration data. Finally, there were 1667 students provided complete information in the presents study. Prior to the study, each participates and their parents or legal guardians must sign a written informed consent. This study was approved by the ethics committee of Ningbo Municipal Center for Disease Control and Prevention.

Measures

Dinner-to-bed time and other behaviors

Data were collected from all participants during one 45-min class period in their classroom through a self-administered questionnaire survey method. The questionnaire information was obtained from each subject: age, sex, parents’ marital status, parents’ education status, physical activity, sleep habits, eating habits, cigarette use, alcohol use, television viewing and computer use. Dinner-to-bed time was defined as the time from finishing dinner to going to bed. The information regarding dinner time and bed time was collected according to the following questions: “When do you usually finish the dinner in the past year?” and “What time do you go to bed on an average evening?” Sleep duration and other details of the questionnaire in this survey had been described in our previous report [16]. Energy intake was measured using a 43-item semi-quantitative food frequency questionnaire (FFQ), and the FFQ was established with the reference to the 2015 China National Nutrition and Health Survey (CNNHS) [17].

Anthropometric measurements

The height, body weight and waist circumference (WC) of all participants were measured by trained research staff in the anthropometric measurement. Some details of the measurement have been described in a previous publication of our group [18]. BMI was calculated using the ratio of weight/height2 (kg/m2). Overweight (including obesity) was defined according to age- and sex-specific cut-off points specified by the International Obesity Task Force (IOTF) [19, 20]. Besides, abnormal obesity was defined as Waist-to-Height Ratio (WHtR) > 0.46 [21].

Statistical analysis

Participants were divided into four groups, according to categories of dinner-to-bed. All continuous variables were summarized as means (standard deviations), and categorical variables were summarized as percentages. A t test was used to compare the difference in continuous variables, and a chi-square test was used to compare the difference in categorical variables. A multivariable linear regression analysis was used to evaluate the relationship between dinner-to-bed time and BMI. Multivariable logistic regression analysis was performed to estimate the odds ratios (ORs) and 95% confidence intervals (CIs) of overweight by dinner-to-bed time adjustment of age (year, continuous), sex, moderate physical activity ≧ 1 day/week (yes or no), watching television ≧ 2 h/day (yes or no), computer use ≧ 2 h/day (yes or no), breakfast skipping (yes or no), sleep duration (hours, continuous), dinner time (hh: mm, continuous), alcohol use(yes or no), eating frequency (times/day, continuous) and energy intake (kcal/day, continuous). Analyses were conducted with the PASW STATISTICS, version 18.0 (SPSS Inc., Chicago, IL, USA), and a two-tailed p value < 0.05 was considered statistically significant. Restricted cubic spline regression was drawn by the ggplot2 and the rms packages of R software 3.6.3 to evaluate the shape of the relation between dinner-bed-time and the odds of overweight by assigning 4.5 h to the reference value of dinner-bed-time and 2.78, 4.00, 4.50 and 5.70 h (5th, 35th, 65th, and 95th percentiles), respectively, to the three knots. Two-tailed P value of < 0.05 was considered to be statistically significant.

Results

The final analyzed sample size was composed of 1667 Chinese school-aged children. The mean age was 14.26 (SD = 0.51) years, 49.46% were girls. Overall, 90.5% of participants parents’ marital were married, 29.2% of participants parents had a college degree. Among all the students, the prevalence of overweight was 17.6%, and the mean of dinner-to-bed time was 4.26 (0.93) h. Characteristics of the study participants according to the categories of dinner-bed-time are shown in Table 1. There were significant differences between the 4 dinner-to-bed time groups for age, sex, moderate physical activity, alcohol use, sleep duration, breakfast skipping, watching TV, computer using, BMI, WHtR, abnormal obesity and dinner time (Ps < 0.05). However, no significant association were observed between the four dinner-to-bed time groups for eating frequency, and energy consumption per day (P > 0.05, respectively).

Table 1 Participants’ characteristics according to the categories of dinner-to-bed time (N = 1667)
Full size table

A multivariable linear regression analysis was applied to assess the associations of BMI with dinner-to-bed time are presented in Table 2. According to these findings, after adjustment for covariates, dinner-to-bed time was negatively associated with BMI and WHtR (B = − 0.47, P < 0.05; B = − 0.02, P < 0.05, respectively).

Table 2 Multivariable linear regression between BMI, WHtR and dinner-to-bed time
Full size table

Results of multivariable logistic regression analysis of dinner-to-bed time and overweight and abdominal obesity (WHtR ≧ 0.46) are shown in Table 3. After adjusting for sex, age, moderate physical activity, watching television, computer using, sleep duration, dinner time, alcohol use, breakfast skipping, eating frequency and energy intake, those with dinner-to-bed time less than 3 h or 3.01 ~ ≦ 4.00 h were more likely to be overweight (OR = 1.94, 95% CI 1.10–3.42; OR = 1.65, 95% CI 1.03–2.65, respectively) or abdominal obesity (OR = 3.03, 95% CI 1.86–4.95; OR = 2.61, 95% CI 1.73–3.92, respectively) compared with dinner-to-bed time more than 5 h. In addition, long dinner-to-bed time was associated with lower risks of overweight (OR = 0.83, 95% CI 0.70–0.97) and abdominal obesity (OR = 0.63, 95% CI 0.54–0.73).

Table 3 Associations of dinner-to-bed time and overweight or WHtR≧ 0.46
Full size table

Cubic spline regression analyses are shown in Figs. 1 and 2. The association between the dinner-to-bed time and overweight or abdominal obesity seemed to be a linear (P > 0.05). The odds of overweight or abdominal obesity steadily decreased with the increasing dinner-to-bed time (Figs. 1 and 2).

Fig. 1
figure 1

Restricted cubic spline regression for the association between the dinner-to-bed time and Overweight. The solid line represents the OR and the shadow represents the 95% CI. Knots were placed at the 5th, 35th, 65th, and 95th percentiles (2.78, 4.00, 4.50 and 5.70 h, respectively) of overweight. The reference value was 4.5 h. The model was adjusted for age (year, continuous), sex, moderate physical activity ≧ 1 day/week (yes or no), watching television ≧ 2 h/day (yes or no), computer use ≧ 2 h/day (yes or no), breakfast skipping (yes or no), sleep duration (hours, continuous), dinner time (hh: mm, continuous), alcohol use(yes or no), eating frequency (times/day, continuous) and energy intake (kcal/day, continuous)

Full size image
Fig. 2
figure 2

Restricted cubic spline regression for the association between the dinner-to-bed time and Abdominal Obesity. The solid line represents the OR and the shadow represents the 95% CI. Knots were placed at the 5th, 35th, 65th, and 95th percentiles (2.78, 4.00, 4.50 and 5.70 h, respectively) of overweight. The reference value was 4.5 h. The model was adjusted for age (year, continuous), sex, moderate physical activity ≧ 1 day/week (yes or no), watching television≧ 2 h/day (yes or no), computer use ≧ 2 h/day (yes or no), breakfast skipping (yes or no), sleep duration (hours, continuous), dinner time (hh: mm, continuous), alcohol use(yes or no), eating frequency (times/day, continuous) and energy intake (kcal/day, continuous)

Full size image

Discussion

This study demonstrated that dinner-to-bed time was independently associated with overweight and abdominal obesity among Chinese school-aged children. Some studies have reported that eating a late dinner or eating late at night was associated with increased risk of overweight and obesity [12, 22, 23]. One study suggested that adults went to bed immediately after dinner had higher odds for obesity [12]. To the best of our knowledge, this is the first study to examine the relationship between dinner-to-bed time and overweight or abdominal obesity among school-aged students. Therefore, more evidence on the relationship between dinner-to-bed time and overweight/obesity is needed.

One of the possible mechanisms can partly explain the association between dinner-to-bed time and overweight/obesity. A short dinner-to-bed time usually means eating a later dinner [24]. In addition, a later dinner may be associated with more energy intake and less energy expenditure [25,26,27]. Moreover, having dinner later might be associated with a decreased plasma leptin level [28]. As a result, people who had short dinner-to-bed time could lead to overweight or obese.

The major strength of the present study was that we used a relatively large school-based sample of students. However, our study also has some limitations. First, the cross-section study nature does not determine the temporal relationship between dinner-to-bed time and overweight/obesity. Second, self-reported dinner and sleep time may introduce measurement error. Finally, we did not measure other aspects of eating behaviors, such as overeating [29], selective eating [30], diet diary [31], eating out [32],eating until full [33], and snacks, especially snacks after dinner [12], which may also be related to the risk of overweight/obesity. Thus, above these potential confounding factors should be considered in future studies.

The results of the present study showed that short dinner-to-bed time was significantly associated with a higher BMI and a higher risk of overweight/ abdominal obesity among children, even after adjusting for sex, age, moderate physical activity, watching television, using computer, sleep duration, dinner time, alcohol use, breakfast skipping, eating frequency and energy intake.

Furthermore, previous studies did not provide the shape of the relationship between dinner-to-bed time and overweight/abdominal obesity. In the analysis using a restricted cubic spline regression model, we found a linear relationship between dinner-to-bed time and overweight/abdominal. For data on dinner-to-bed time, a monotonically decreasing relationship was observed for long dinner-to-bed time. Findings from this study indicate that the shorter dinner-to-bed time, the higher risk of being overweight.

Conclusions

In conclusion, findings from this study demonstrated that dinner-to-bed time was negatively correlated with BMI, and short dinner-to-bed time was associated with increased risk of overweight/abdominal obesity among children. This study suggested that modifying the dinnertime later to early in a day may help prevent overweight in children.

What is already known on this subject?

Recently, later dinner has been identified as a potential independent risk factor for overweight. A recent study reported that men who went to bed immediately after dinner were associated with increased odds for obesity. However, there is no epidemiology study focus on the effect of dinner-to-bed time on overweight or obesity in children.

What does this study add?

Our findings revealed that dinner-to-bed time was negatively correlated with BMI, and short dinner-to-bed time was associated with increased risk of overweight/abdominal obesity among children. Our results suggested that eating dinner early in a day may help prevent overweight in children.