Background

In recent years, the prevalence of childhood overweight and obesity is dramatically increasing in China [1, 2]. Childhood overweight or obesity may be associated with an increased risk of future chronic diseases such as cardiovascular diseases, diabetes and orthopedic problems [3, 4]. At the same time, some risk factors for being overweight or obesity have been identified, including high energy intake and low energy expenditure [5], physical inactivity [6], and sleep disorder [7]. Recently, eating speed has been identified as a potential independent risk factor for being obese or overweight, and accumulating evidence indicates that eating quickly may play an important role in development of overweight and obesity. Body mass index (BMI) is an indicator that commonly used for defining overweight or obesity in clinical practice [8], but it does not distinguish lean mass from fat mass. Waist-to-height ratio (WHtR) as a proxy for abdominal fat mass, was independent of BMI associated with the risk of cardiovascular outcomes in adults [9]. Ochiai et al. reported that eating quickly may lead to overweight and excess gains in anthropometric variables (including BMI, % body fat (BF), waist circumference (WC), and WHtR among schoolgirls [10, 11]. Lin et al. showed that eating quickly was independent associated with overweight and obese among pre-school children [12]. However, only one study had examined the association between eating speed and obesity among Chinese school children [13].

Therefore, the aim of this cross-sectional study was to investigate the associations between eating speed and overweight after adjusting for lifestyle factors among school children in China. We hypothesis was that children who eating quickly would have higher BMI, WHtR and serum lipids.

Methods

Participants

Data for this cross-sectional study were obtained from a school-based study, which aimed to examine the relationships between health-related behaviors and overweight/obese in children. This study was carried out in Ningbo city, Zhejiang province, China. In brief, this study contained a self-administered health questionnaire survey and a physical examination. A total of 664 aged 10–12 years schoolchildren (5th and 6th grades) from three primary schools were recruited using a non-random convenience sampling in October 2018. Individuals data was excluded if they were had extreme BMI (˂10 or > 50) or the data (body weight or height or eating speed) were missing in this study. Finally, 629 students were analyzed in the presents study. This study was approved by the ethics committee of Ningbo Center for Disease Control and Prevention. Prior to the study, written informed consents was obtained from all participates and their parents/legal guardians.

Measures

Eating speed and other behaviors

A self-administered questionnaire survey was conducted to assess the students’ eating speed and other health-related behaviors during one 45-min class period in their classroom. The following information was obtained from each participant: sex, age, parents’ marital status, parents’ education status, physical activity, sleep habits, dietary habits, cigarette use, alcohol use, television watching, computer use and eating speed. The information regarding eating speed was collected according to the question: “How fast is your eating speed compared to others when you have an ordinary meal?” The participants chosen from three categories: “slow”, “medium”, and “fast” [12,13,14,15]. Other details of questionnaire of the survey have been published [16].

Measurements

In the general health examination, the height and body weight of all participants were measured by trained research staff. Details of the examination have been described in a previous publication of our group [17]. Body mass index (BMI) was calculated using the ratio of weight/height2 (kg/m2). Overweight (including obesity) were defined according to age- and sex-specific cut-off points specified by the International Obesity Task Force (IOTF) [18, 19]. Central obesity was defined as waist-to-height ratio (WHtR) ≥ 0.48 for boys and WHtR ≥ 0.46 for girls [20]. Fasting blood samples were obtained from all participants, and triglycerides (TG), total cholesterol (TC), high-density lipoprotein cholesterol (HDL-C) and low-density lipoprotein cholesterol (LDL-C) were measured by standard enzymatic methods on a Beckman Coulter AU 640 (Beckman Coulter, Tokyo, Japan). Dyslipidemia concentrations were defined as follows: TG concentration higher than 1.70 mmol/L; TC concentration higher than 5.18 mmol/L; HDL-C concentration lower than 1.03 mmol/L; and LDL-C concentration of 3.37 mmol/L or higher [21].

Statistical analysis

Since 3.8% (24) participants who reported their speed of eating as “slow”, we divided all the participants into two eating-speed groups as follow: not fast (slow and medium) and fast. Continuous variables were summarized as means (standard deviations), and categorical variables were summarized as percentages. Differences in continues variables were examined using the t test, whereas categorical variables were analyzed using a Chi-square test. A multiple linear regression analysis was used to evaluate associations between the eating speed and BMI, WHtR and serum lipids. We used logistic regression analysis to estimate the odds ratios (ORs) and 95% confidence intervals (CIs) of overweight/abdominal obesity by eating speed adjusted for age (year, continuous), sex, parents’ education level on college (none, one and both), watching television ≧ 2 h/day (yes or no), and computer use ≧ 2 h/day (yes or no). A two-tailed p value ˂ 0.05 was considered statistically significant. All analyses were performed using the PASW STATISTICS, version 18.0 (SPSS Inc., Chicago, IL, USA).

Results

The analyzed sample was composed of 629 Chinese school children. The mean of age was 10.26 (SD = 0.45) years old, 45.0% were girls. Overall, 26.2% participants reported eating fast, 93.5% participants parents’ marital were married, 33.7% of participants parents had no college degree. Among all the students, the prevalence of overweight (including obesity) was 22.9%, and the mean of sleep duration was 9.69 (SD = 0.63) hours (Table 1). Table 1 also shows the association between lifestyle factors and eating speed. Subjects who ate faster were more likely to be boys, with higher BMI and WHtR, higher TG and TC. Compared with not eating fast, participants who were inclined to eat faster were also more likely to be overweight or abdominal obese, to skip breakfast, to smoking, and to have more time spent on television and computer. However, no significant correlation was found between eating speed and dyslipidemia among children in the results of this study.

Table 1 Characteristics of participants by eating speed in the study
Full size table

A multiple linear regression analysis was applied to assess the associations of BMI, WHtR, TG, TC, HDL-C and LDL-C with eating speed is presented in Table 2. According to these findings, slower eating speed were independently associated with lower BMI (B = − 0.70, 95% CI − 1.26 to − 0.14) and TG (B = − 0.16, 95% CI − 0.28 to − 0.04), after controlling for sex, age, physical activity, sleep duration, breakfast skipping, watching television, computer use, cigarette use and alcohol use.

Table 2 Regression analyses among the two eating speed groups and anthropometric profile (n = 629)
Full size table

Table 3 shows the results of multivariable logistic regression analysis of eating speed and overweight. Compared to eating not fast, participants who ate fast were more likely to be overweight (OR 1.81, 95% CI 1.19–2.75) after adjusting for sex, age, physical activity, sleep duration, breakfast skipping, television watching, computer use, cigarette use and alcohol use. No significant correlation was found between eating speed and abdominal obesity.

Table 3 Multivariable-adjusted OR and 95% CI for the association of eating speed with overweight or abdominal obesity (n = 629)
Full size table

Discussion

The association between eating speed and overweight among school-aged children has received relatively little attention in previous researches. Some studies have suggested that eating quickly has a role in obesity not only in adults [20, 22] but also in adolescents [10, 23, 24]. However, Leong et al. suggested that once women have reached mid-life, eating quickly does not increase the risk of obesity [25]. So the relationship between eating speed and overweight need more evidence.

Consistent with prior researches [10, 23, 24], our study showed that eating quickly was significantly associated with a higher BMI, a higher TG and a higher risk of overweight among children, even after adjusting for sex, age, physical activity, sleep duration, breakfast skipping, television watching, computer use, cigarette use and alcohol use. However, the small sample size in our study may affect the statistical efficiency, no significant relationships were found between eating fast and dyslipidemia, and abdominal obesity among children in our study. To the best of our knowledge, this present study is the first one to investigate the association between eating speed and BMI, serum lipids, overweight, and abdominal obesity among school children in China.

Precise mechanism linking eating speed and overweight is not well understood. However, one possible mechanism can partly explain this association. Some studies show that eating quickly may cause more energy intake [26, 27], while eating slowly may help to reduce energy intake [28]. This may be because more energy is ingested by fast eaters before the brain recognizes the signal of satiety [29]. As a result, people who eat quickly could lead to insulin resistance, obesity and metabolic syndrome [30, 31].

The present study also has some limitations. First, all participants were recruited from primary school. This may limit the ability to extrapolate these results to the general population of children. Second, the cross-section study nature does not determine the temporal relationship between eating speed and overweight. Third, self-reported eating speed may have introduced measurement error. Forth, the energy consumption was not controlled in the present study. Finally, we did not measure other aspects of eating behaviors, such as overeating [32], selective eating [33], eat timing [34], daily eating frequency [35], eating out [36] and trait impulsivity [37], which may also be related to the risk of overweight. Thus, further studies including more details about eating behaviors with prospective design are needed.

Conclusions

In summary, our data demonstrated that eating speed affected BMI and TG, and eating fast was associated with overweight among school children. The present study suggested that modifying eating speed from quick to a relatively slow may help prevent overweight in children.

What is already known on this subject?

Recently, eating speed has been identified as a potential independent risk factor for obesity or overweight, and a few studies indicated that eating quickly may play an important role in development of overweight and obesity. However, there were limited studies focused on the effect of eating speed on overweight or obesity in children.

What does this study add?

Our findings revealed that eating speed was independently associated with BMI and TG, and eating fast was associated with increased risk of overweight/abdominal obesity among children. Our results suggested that eating slowly may help prevent overweight in children.