Association between Physical Activity, Sedentary Behaviors, Sleep, Diet, and Adiposity among Children and Adolescents in China

Abstract

Introduction

Physical activity (PA), sedentary behaviors (SB), sleep, and diet are related to adiposity among children and adolescents. However, there may be interactions between PA, SB, sleep, and diet, and these lifestyle behaviors may work together to affect body weight. The purpose of this study was to explore the impact of multiple lifestyle behaviors of PA, SB, sleep, and diet on childhood adiposity (body mass index z-score and overweight/obesity), and to investigate the effect of meeting multiple guidelines on adiposity among children and adolescents in China.

Methods

Cross-sectional results were based on 28,048 children aged 6–17 years from the China National Nutrition and Health Surveillance in 2010–2012. Information about PA, SB, and sleep was measured through interview-administered questionnaire. Dietary intake was assessed with food frequency questionnaire. The associations between multiple lifestyle behaviors and BMI z-score and overweight/obese were examined.

Results

The prevalence of overweight/obesity in the participants was 19.2%. The average time of moderate-to-vigorous PA (MVPA), leisure SB, and sleep was 76.7 ± 45.5 min, 2.9 ± 1.4 h, and 8.5 ± 1.1 h per day, respectively. The China Dietary Guidelines Index for Youth (CDGI-Y) score was 62.6 ± 11.0. Sleep duration and diet score were negative associated with BMI z-score (both p < 0.001). MVPA and SB time were positive associated with BMI z-score (p = 0.041, 0.004). Meeting the SB, sleep, and diet guidelines had a lower BMI z-score (all p < 0.01) and lower odds of overweight/obesity (all p < 0.05). There were significant interactions between PA and diet. Compared with meeting no guidelines, those who met multiple guidelines had a lower risk of overweight/obesity (all p < 0.01). The more guidelines the participants met, the lower odds of overweight/obesity (p for trend <0.001).

Conclusions

PA, SB, sleep, and diet are important behaviors associated with adiposity among children and adolescents. Attaining adequate amounts of appropriate multiple behaviors provided an additional benefit. It is important for children to meet recommended behavioral guidelines or recommendations. Interventions that aim to improve awareness of and compliance with these guidelines are needed in future.

Introduction

Many studies have indicated that childhood obesity is a complex but rapidly growing global epidemic [1, 2]. Children and adolescents with obesity are 5 times more likely to be obese when they are adults [2]. Overweight and obesity in childhood are also closely related to cardiovascular and cerebrovascular diseases in adulthood [1, 2]. Many health-protective behaviors contribute to child and adolescent health, including adequate physical activity (PA), enough sleep, and proper dietary intake [3, 4, 5]. Childhood is also an important period to develop a healthy lifestyle. Therefore, identification of child and adolescent lifestyle behaviors that contribute to excess weight gain is essential for promoting their health.

Active PA, less sedentary behaviors (SB), adequate sleep, and good quality diet are reported to be independently related to a lower risk of adiposity [4, 5, 6, 7]. However, there may be interactions between PA, SB, sleep, and diet [8, 9, 10], and these lifestyle behaviors may work together to affect body weight [11]. How combinations of these lifestyle behaviors are associated with adiposity is largely unknown as few studies have examined the influence of PA, SB, sleep, and diet combinations on adiposity among children and adolescents. A subset of studies have considered the combination of PA, SB, and sleep as a whole of 24-h movement and examined its relationship to obesity, but few evaluated the diet quality [3, 12] or the overall dietary patterns [13, 14, 15]. Two studies of small samples on specific age-groups have studied the association between multiple lifestyle behaviors and adiposity in children [16, 17]. Exploring the impact of multiple lifestyle behaviors on adiposity in a large sample can help us study the relationship more convincingly and understand where and how to intervene.

For children and adolescents, there are specific guidelines or recommendations for PA, SB, sleep, and dietary intake to promote their health. The PA guidelines for Chinese children and adolescents recommend that children and adolescents should engage in at least 60-min moderate-to-vigorous PA (MVPA) per day, and the daily screen time should be less than 2 h [18]. The National Sleep Foundation has age-specific guidelines for sleep duration in children and adolescents (9–11 h per night for those aged 6–13 years and 8–10 h for those aged 14–17 years) [19]. The 2016 Dietary Guidelines for Chinese recommend balanced dietary patterns and the food amounts for people with different estimated energy requirements and also recommend to restrict unhealthy eating behaviors, such as reducing the intake of sugary drinks and energy-dense nutrient-poor snacks [20]. Research on Chinese children whereby multiple lifestyle behaviors have been examined simultaneously is distinctly lacking, and only one study from Shanghai [21] has examined the effect of sleep, diet, and PA on obesity and overweight elementary school students, in which only the amount of some food consumed was measured, while the overall dietary pattern was not evaluated. Therefore, the purpose of this study is to explore the impact of multiple lifestyle behaviors of PA, SB, sleep, and overall diet on childhood adiposity and to investigate the effect of meeting multiple guidelines or recommendations on adiposity among children and adolescents in China.

Materials and Methods

Study Design and Participants

The data were from the China National Nutrition and Health Surveillance (CNNHS) in 2010–2012, which was a nationally representative cross-sectional study. The survey design and methods have been described in detail previously [22]. In brief, the surveys covered 31 provinces (including autonomous regions and centrally administered municipalities). A multistage stratification and population proportional cluster random sampling method was adopted. For this analysis, we restricted the study sample to participants aged 6–17 years with dietary intake data, which resulted in sample sizes of 28,818 participants. The study was approved by the ethics review committee of the National Institute for Nutrition and Health, Chinese Center for Disease Control and Prevention (No. 2013-018), and all participants' parents or legal guardians signed the informed consent.

PA, SB, and Sleep

A face-to-face interview questionnaire which conducted at children's home or school by trained investigators was used to collect the information about PA, SB, and sleep in the past semester. The children younger than 10 years old finished the questionnaire with the help of their parents. The questionnaire investigated 4 PA domains: school-time PA, transportation PA, leisure-time exercising, and housework. MVPA included school-time MVPA, active transportation PA (walking and cycling), and leisure-time MVPA. Leisure-time SB included watching television, using computers, playing video games, and reading and doing homework in leisure time.

Dietary Intake

Dietary intake over the past year was assessed with a validated food frequency questionnaire (FFQ). The FFQ includes 100 food items. Participants were asked the frequency and amount of each food consumed. Dietary quality was evaluated by the China Dietary Guidelines Index for Youth (CDGI-Y), which is a modified index based on the China Dietary Guidelines Index for Adults (CDGI [2019]-A) [23] and Chinese Children Dietary Index (CCDI-16) [24], which were both based on the Chinese Dietary Guidelines 2016 (CDG-2016) [20]. The CDGI-Y was composed of 12 food-related components and 2 nutrient-related components: (1) other cereals and miscellaneous beans, (2) total vegetables, (3) fruits, (4) milk and dairy products, (5) soybeans, (6) nuts, (7) fish and seafood, (8) poultry and meat, (9) eggs, (10) salt, (11) energy-dense nutrient-poor snacks, (12) sugar-sweetened-beverages, (13) energy balance: ratio of energy intake to estimated energy requirement, and (14) energy supply ratio of carbohydrate. The total score of CDGI-Y was 120 and the criteria for the scoring of each component are shown in online supplementary Table S1 (for all online suppl. material, see www.karger.com/doi/10.1159/000519268).

Anthropometric Measurements

Height and weight were measured in the morning before breakfast. All measurements were conducted by well-trained investigators under standard operation procedure. The height was accurate to 0.1 cm and the weight was accurate to 0.1 kg. Body mass index (BMI) was calculated as weight (kg) divided by height (m) squared (kg/m²). BMI z-scores were computed using age- and sex-specific reference data from the World Health Organization [25]. Overweight and obesity were classified according to age- and sex-specific BMI cutoff points, which developed for Chinese children by Group of China Obesity Task Force Correspondence [26].

Socioeconomic Demographic Measures

Socioeconomic demographic information included the participant's date of birth, sex, region, and household income. The living region of participants was divided into 4 strata: big cities, medium and small cities, ordinary rural areas, and poor rural areas, according to their administrative division, population, and level of economic development [27]. The household income collected in this article is the per capita family annual income, and it was categorized into 4 grades: <20,000 China Yuan (CNY), 20,000 CNY∼39,999 CNY, ≥40,000 CNY, and “Unknown.”

Data Handling

Meeting multiple guidelines or recommendations were defined as active PA: ≥60 min/day of MVPA [18]; low SB: ≤2.0 h/day of leisure SB [28]; enough sleep: ≥9 h/day of sleep for children aged 6–13 years; and ≥8 h/day for those aged 14–17 years [19]. Those not meeting these guidelines or recommendations were classified as “inactive,” “high SB,” and “short sleep,” respectively. Similar to other studies evaluating dietary quality [17], the top quintile of sample CDGI-Y scores was deemed as meeting dietary guidelines. In the analytic sample, the top quintile had a CDGI-Y score above 72.0. The number of meeting guidelines or recommendations was the sum of PA, SB, sleep, and dietary guidelines or recommendations achieved (possible score 0–4).

Statistical Analysis

Central tendencies of descriptive characteristics were calculated. The associations between PA, SB, sleep, and diet and BMI z-score were examined using multiple linear regressions. In model 1, relationships were assessed adjusting for potential covariates. In model 2, all 4 behaviors and covariates were included simultaneously to obtain independent associations between each behavior and BMI z-score. BMI z-score was also compared between groups (e.g., active vs. inactive) using multiple linear regressions. In model 1, all categorical variables were included, adjusting for potential covariates and other behaviors. In model 2, all 6 possible 2-way interactions between behavioral groups were added to model 1, and any that were nonsignificant were deleted in a backward stepwise manner until only significant interactions were left (p < 0.05). The least squares means of significant interactions were computed, and post hoc tests using the Bonferroni correction were conducted to identify which were significantly different. The associations between each behavioral group, number of meeting guidelines, and overweight/obesity were examined using binary logistic regression. A sensitivity analysis was done to see if the missing values would affect the results. All data were analyzed using the Statistical Analysis System (SAS) 9.4 software (SAS Institute Inc., Cary, NC, USA).

Results

Characteristics of Participants

There were 28,818 participants completed the PA questionnaire, FFQ, and the anthropometric measurements. Further excluding participants with missing data for the measured activities resulted in an analytic sample of 28,048 participants. No significant differences were observed between those included or excluded for any of the exposure or outcome measures.

Descriptive statistics for the analytic sample are displayed in Table 1. The mean age of participants was 12.0 (±3.3) years, 50.2% of participants were boys and 19.2% were classified as overweight/obese. Approximately 60% of participants met the PA guidelines, 40.1% and 63.4% met the SB and sleep recommendations, respectively. The top quintile (20%) of CDGI-Y was preselected as meeting the diet guideline. Only 3.2% of participants met all 4 guidelines or recommendations. In those who met 3 (20.4%, n = 5,734), the majority met PA, sleep, and SB recommendations (12.5%, n = 3,493). In those who met 2 (39.6%, n = 11,118), the majority met PA and sleep guidelines (17.3%, n = 4,847).

Table 1.

Descriptive characteristics of sample (n = 28,048)

	Mean (SD)
	total sample	boys	girls
Continuous variables
Age, years	12.0 (3.3)	12.1 (3.3)	12.0 (3.3)
MVPA, min/day	76.7 (45.5)	78.8 (47.5)	74.6 (43.2)
Leisure SB, h/day	2.9 (1.4)	2.9 (1.4)	2.9 (1.4)
Sleep, h/night	8.5 (1.1)	8.5 (1.1)	8.5 (1.1)
CDGI-Y score	62.6 (11.0)	61.8 (11.0)	63.4 (10.9)
BMI z-score	0.04 (1.2)	0.16 (1.3)	−0.08 (1.1)
Categorical variables, N (%)
Sex	28,048 (100.0)	14,070 (50.2)	13,978 (49.8)
Region
Big cities	6,736 (24.0)	3,359 (23.9)	3,377 (24.2)
Medium and small cities	8,201 (29.2)	4,124 (29.3)	4,077 (29.2)
Ordinary rural areas	8,256 (29.4)	4,178 (29.7)	4,078 (29.2)
Poor rural areas	4,855 (17.3)	2,409 (17.1)	2,446 (17.5)
Household income
<20,000 CNY	16,483 (58.8)	8,306 (59.0)	8,177 (58.5)
20,000–39,999 CNY	3,576 (12.7)	1,828 (13.0)	1,748 (12.5)
≥40,000 CNY	929 (3.3)	479 (3.4)	450 (3.2)
Unknown	7,060 (25.2)	3,457 (24.6)	3,603 (25.8)
BMI status (% overweight/obese)	5,379 (19.2)	3,186 (22.6)	2,193 (15.7)
MVPA (% active)	16,600 (59.2)	8,560 (60.8)	8,040 (57.5)
Leisure SB (% low SB)	11,254 (40.1)	5,670 (40.3)	5,584 (39.9)
Sleep (% enough sleep)	17,771 (63.4)	9,005 (64.0)	8,766 (62.7)
Diet (% meeting dietary guidelines)	5,609 (20.0)	2,558 (18.2)	3,051 (21.8)
Guidelines met#
0	2,091 (7.5)	961 (6.8)	1,130 (8.1)
1	8,208 (29.3)	4,128 (29.3)	4,080 (29.2)
2	11,118 (39.6)	5,705 (40.5)	5,413 (38.7)
3	5,734 (20.4)	2,849 (20.2)	2,885 (20.6)
4	897 (3.2)	427 (3.0)	470 (3.4)

MVPA, moderate-to-vigorous physical activity; SB, sedentary behaviors; CDGI-Y, China Dietary Guidelines Index for Youth; BMI, body mass index.

^#Guidelines or recommendations met: the number of meeting (1) active PA, MVPA ≥60 min/day; (2) low SB, leisure SB time ≤2.0 h/day; (3) enough sleep, sleep time >9 and 8 h per night for children aged 6–13 and 14–17 years, respectively; (4) meeting dietary guidelines, top quintile of sample scores of CDGI-Y.

PA, SB, Sleep, Diet, and BMI z-Score

As shown in Table 2, more MVPA was associated with a higher BMI z-score (p = 0.041). Each additional hour of SB was associated with a higher BMI z-score by 0.015 units (p = 0.004). Every additional hour of sleep was associated with a lower BMI z-score by 0.077 units (p < 0.001). Each additional 10 points of CDGI-Y score was associated with a lower BMI z-score by 0.047 units (p < 0.001).

Table 2.

The relationships between PA, SB, sleep, diet, and BMI z-score: β coefficients and SE

	Model 1*			Model 2#
	β	SE	p value	β	SE	p value
MVPA time (h/day)	0.020	0.009	0.031	0.019	0.009	0.041
Leisure SB time (h/day)	0.021	0.005	<0.001	0.015	0.005	0.004
Sleep duration (h/night)	−0.080	0.008	<0.001	−0.077	0.008	<0.001
Diet score (CDGI-Y score§)	−0.047	0.007	<0.001	−0.047	0.007	<0.001

BMI, body mass index; MVPA, moderate-to-vigorous physical activity; SB, sedentary behaviors; CDGI-Y, China Dietary Guidelines Index for Youth.

^*Model 1: assessed using linear regression with adjustment for age, sex, region, and household income.

^#Model 2: assessed using linear regression with adjustment for age, sex, region, and household income, and other behaviors.

^§β is expressed per 10 points increase in CDGI-Y total score.

Table 3 shows the relationships between lifestyle behavioral groups and BMI z-score. No significant relationship was found in PA groups. Participants who met the SB, sleep, and diet guidelines had a BMI z-score by 0.046, 0.129, and 0.09 units lower than those who did not.

Table 3.

The relationships between lifestyle behavior groups and BMI z-score: β coefficients and SE

	Model 1*			Model 2#
	β	SE	p value	β	SE	p value
Active PA§	0.012	0.014	0.407	−0.005	0.016	0.778
Low SB§	−0.046	0.015	0.002	−0.047	0.015	0.002
Enough sleep§	−0.129	0.015	<0.001	−0.129	0.015	<0.001
Meeting dietary guidelines§	−0.090	0.018	<0.001	−0.137	0.027	<0.001
Active PA × meeting dietary guidelines§§				0.081	0.036	0.024

BMI, body mass index; PA, physical activity; SB, sedentary behaviors.

^*Model 1: assessed using linear regression with adjustment for age, sex, region, and household income, and other behaviors.

^#Model 2: model 1 + significant interactions: all 6 possible 2-way interactions between lifestyle behavior groups were added to model 1, and any that were nonsignificant were deleted in a backward stepwise manner until only significant interactions were left (p < 0.05).

^§Reference categories were inactive, high SB, short sleep, and not meeting the dietary guidelines, respectively.

^§§Estimates refer to the specified groups. All other possible group combinations act as the referent group.

Significant interactions were found between PA and diet. As shown in Figure 1, participants who met the dietary guidelines but were inactive had a significantly lower BMI z-score than those who did not meet the dietary guidelines, whether they were active (p < 0.001, g = −0.143) or inactive (p < 0.001, g = −0.149). For the participants who met the dietary guidelines, the inactive ones had a significantly lower BMI z-score than the active ones (p = 0.045, g = −0.086).

Fig. 1.

Significant interaction among behavioral groups and BMI z-score: PA × diet interaction. BMI, body mass index; PA, physical activity.

PA, SB, Sleep, Diet, and Overweight/Obesity

Low SB, enough sleep, and meeting dietary guidelines were associated with lower odds of overweight and obesity in both model 1 and model 2. No significant relationship was observed for active PA (Table 4).

Table 4.

Associations of lifestyle behavior groups with overweight/obesity in children and adolescents

	Model 1*			Model 2#
	ORs and 95% CIs	p value	ORs and 95% CIs	p value
Active PA§	1.015 (0.954–1.080)	0.636	1.010 (0.949–1.074)	0.755
Low SB§	0.912 (0.855–0.972)	0.004	0.922 (0.865–0.982)	0.012
Enough sleep§	0.816 (0.766–0.869)	<0.001	0.821 (0.771–0.875)	<0.001
Meeting dietary guidelines§	0.855 (0.791–0.923)	<0.001	0.858 (0.794–0.926)	<0.001

PA, physical activity; SB, sedentary behaviors.

^*Model 1: adjusting for age, sex, region, and household income.

^#Model 2: adjusting for age, sex, region, and household income, and other lifestyle behaviors.

^§Reference categories were inactive, high SB, short sleep, and not meeting the dietary guidelines, respectively.

Number of Meeting Guidelines or Recommendations and Overweight/Obesity

Participants who met one or more guidelines or recommendations had lower odds of overweight and obesity than those who met zero one (OR = 0.822, 0.738, 0.689, and 0.662, respectively, for meeting 1, 2, 3, and 4 guidelines/recommendations). The more guidelines the participants met, the lower odds of overweight/obesity (overall p for trend <0.001) (shown in Fig. 2).

Fig. 2.

Associations of number of meeting guidelines or recommendations^# with overweight/obesity in children and adolescents: ORs and 95% CIs, adjusting for age, sex, region, and household income. ^#The number of meeting (i) active PA, MVPA ≥60 min/day; (ii) low SB, leisure SB time ≤2.0 h/day; (iii) enough sleep, sleep time ≥9 and 8 h per night for children aged 6–13 and 14–17 years, respectively; (iv) meeting dietary guidelines, top quintile of sample scores of CDGI-Y (≥72.0). MVPA, moderate-to-vigorous physical activity; SB, sedentary behaviors; CDGI-Y, China Dietary Guidelines Index for Youth; BMI, body mass index; PA, physical activity.

Discussion

This study examined the relationships between multiple lifestyle behaviors of PA, SB, sleep, diet, and adiposity among children in a large sample. The results showed SB time, sleep duration, and diet score to be independently associated with BMI z-score, which were similar to previous studies [16, 17]. This study also explored the impact of meeting behavioral guidelines or recommendations on BMI z-score and overweight/obesity, meeting guidelines for SB, sleep, and diet was significantly associated with a lower BMI z-score, and one significant interaction (PA × diet) was observed. Few studies explored the interactions of lifestyle behaviors including PA, sleep, and diet on children adiposity, but the available evidence from a UK study on 9- to 11-year-old children [16] is consistent with our findings. In our study, the probability of being overweight/obese when meeting the 4 guidelines or recommendations was the lowest compared to meeting none, 1, a combination of 2 or 3 recommendations; similar findings have been reported [13, 17].

In this study, 59.2% of the participants met the PA guideline, and the average time of MVPA was 76.7 min/day. The prevalence of meeting PA guidelines was higher than that reported in PAFCTYS (Physical Activity and Fitness in China − The Youth Study, another national survey, conducted in China) [29, 30], and the possible reason was that different survey tools were used, although both tools have been tested for reliability and validity [31, 32]. The association between PA levels and obesity has been widely investigated [6, 33]. Reports of the relationship between PA and weight are inconsistent. Most studies have reported that the risk of overweight/obesity is inversely correlated with PA, while some report no significant link [34] or positive correlation [35]. We observed that children engaging in greater PA had a higher BMI z-score than others. One reason for this finding is that overweight/obesity children may have more MVPA in an attempt to lose weight [36]. Alternatively, overweight/obese children may be more likely to exaggerate the time and intensity of PA they engaged in. In addition to the reported inaccurate level of PA, the effect of PA may be offset by other factors or habits and may not imply that more hours of MVPA will lower BMI [35]. Similar to other studies [34, 37], there was no expected effect when considering PA alone, but there were significant ones when combined with other behaviors. In any case, PA is good for health of children and adolescents [38]. It is important to provide young people opportunities and encouragement to participate in PA that is appropriate for their age [18, 39].

Sleep duration was inversely associated with adiposity in the present study, which is consistent with other studies [40, 41]. Insufficient sleep as a possible cause of weight gain and obesity has received a considerable attention over the past decade [42]. The potential mechanism is the short sleep-obesity association by a change of metabolism and weight-related behaviors [10, 43]. Although the United States [19] and Canada [39] recommend children's sleep time as a range, recent studies concluded that there was no health risk associated with long sleep [44]. Surveys in most countries suggested that long sleepers were rare [40, 45]; and the sleep duration of children and adolescents showed a declined tendency [46]. More attention should be paid to childhood sleep insufficiency with the increase of school tasks and extracurricular classes. It is important to let people understand that sleep is not a waste of time and sleep health is an important factor of preventing obesity [47].

Strong evidence demonstrates a significant relationship between greater time spent in SB and higher all-cause mortality rates in adults, but the evidence on children and adolescents was limited [38, 48]. In our study, additional time of leisure SB was associated with a higher BMI z-score; and there was a significant inverse correlation between SB and adiposity. Similar findings have been reported [13, 14, 16]. Related to higher weight status or adiposity in children and adolescents, evidence in television viewing or screen time was stronger than SB [38]. Our study provides a large sample of evidence for the relationship between leisure SB and adiposity. Considering the health risk and the popularity of using cell phones, tablets, and other devices for playing video games, more countries recommended the maximum screen time of children in 2 h/day [18, 39]. More researches are needed to propose the recommendation of maximum SB time.

Dietary indexes are potentially useful methods for dietary assessment because they offer valuable information on overall dietary patterns [49]. There were limited studies using dietary scores to evaluate the dietary quality of children and adolescents in China. This is the first study in China, showing the dietary index-health associations after adjusted for demographic and behavioral confounders in a national sample. Our results displayed a lower BMI z-score with the CDGI-Y score increased; the odds of overweight/obesity reduced by 14% among children with the top quintile score of CDGI-Y. The study on British children aged 9–10 years [50] and another on Irish children aged 9 years [51] were consistent with our findings that after multiple adjustments of PA and demographic covariates, higher dietary quality scores were associated with improved weight status. Many foods or nutrients may be involved in the promotion or protection against adiposity. When only one food or nutrient is studied in relation to obesity, the findings are often inconsistent across studies [52]. There were several dietary indexes used in studies on childhood obesity such as Healthy Eating Index [17], Diet Quality Index and the Health Diet Indicator [50], diet quality score [51], and Mediterranean diet KIDMED index [53]. These indexes are mostly based on the national dietary guidelines. The CDGI-Y used in our study was based on Chinese Dietary Guidelines for children, reflected the overall dietary patterns, and may be used more in future studies in China.

This study showed that the more lifestyle behavioral guidelines or recommendations an individual met, the lower the risk of adiposity. Other studies have similar results [13, 16, 17, 37]. A 12-country study [13] showed that the probability of being obese when meeting the 3 recommendations of PA, SB, and sleep was the lowest compared to meeting none, 1, or a combination of 2 recommendations; and another study in US adolescents [17] showed that adolescents who met multiple guidelines of PA, sleep, and diet had a lower risk of adiposity; and the ISCOLE study (International Study of Childhood Obesity, Lifestyle and the Environment) from the UK [16] showed that participants who did not meet the guidelines for either behavior of PA, SB, or sleep had the highest BMI z-score. Pereira et al. [37] identified a combination of sedentary and poorer diet quality to be associated with a higher prevalence of overweight/obesity in Portuguese children.

In comparison to the studies mentioned above, this study examined simultaneously the impact of the 4 lifestyle behaviors of PA, SB, sleep, and diet on childhood adiposity in a much larger sample. Overall, multiple health behaviors of PA, SB, sleep, and diet provided added benefit on child and adolescent health.

It is concerning to see very low levels of adherence to the combined recommendations among Chinese children and adolescents. The results were similar to those found in the 12-country study [13] where only 7% of children met all 3 recommendations across the 12 countries and only 1.5% of children in China. In the study by Kracht et al. [17], there were only 0.4% of the participants met all 3 guidelines. Lifestyle behaviors are developed and formed during childhood, and without effective intervention, unhealthy lifestyle behaviors often extend into adulthood [54]. This study illustrated the importance of childhood behavior development, which future research should pay more attention to.

In addition to behavioral factors, obesogenic environmental factors, including socioeconomic status, have emerged as also contributing to the obesity problem. This study also analyzed the impact of living region and household income on childhood obesity, which were not shown in the results, but were included as adjustment factors. Urban areas with higher household income were positively associated with childhood obesity in this study, which is consistent with other studies in developing countries [55, 56]. Besides, the number of siblings, parents' occupational and educational level, and parental role models are also important factors that affect childhood obesity, which were not collected in this study. More research focused on socioeconomic status should be carried out in the future.

Major strengths of this study include the large sample size with national representativeness, the rigorous standardization of measurement and data collection, the combined analysis of PA, SB, sleep, diet, and health, and evaluation of the overall dietary patterns. The limitations are the cross-sectional study design, which does not allow a cause-effect interpretation, the self-reported measures which may have recall and social desirability bias, and factors that may affect the results but are not taken into account (i.e., biological age/maturity, family size, parents' occupational and educational level, parental role models, the role of screen time separate from SB, and the role of lifestyle behaviors across school and weekend days separately). And also, the data were collected in 2010–2012, and the model of children's lifestyle behaviors may have changed, especially with the rapid changes in media behavior. So the characteristics of children's PA, SB, sleep, and diet in the current society should be considered when referring to this paper. Future research using objective assessment measures and in longitudinal analysis may provide more convincing evidence.

Conclusion

PA, SB, sleep, and diet are important behaviors associated with adiposity among children and adolescents. Attaining adequate amounts of appropriate multiple behaviors provided additional benefit. Meeting all 4 behavior guidelines or recommendations resulted in the lowest odds ratios for obesity, while meeting 3 ones was better than meeting 2, meeting 2 was better than meeting 1, and meeting 1 was better than meeting none. It is important for children to meet recommended behavioral guidelines or recommendations. Interventions that aim to improve awareness of and compliance with these guidelines are needed in future.

Statement of Ethics

All participants signed written informed consent. Ethical approval for this study was provided by Ethics Committee of National Institute for Nutrition and Health, Chinese Center for Disease Control and Prevention (ethical approval No. 2013-018).

Conflict of Interest Statement

The authors have no conflicts of interest to declare.

Funding Sources

This research was funded by Study of Diet and Nutrition Assessment and Intervention Technology from Active Health and Aging Technologic Solutions Major Project of National Key R&D Program (grant No. 2020YFC2006300) and Major program for health care reform from National Health Commission of the People's Republic of China (grant No. 20120212).

Author Contributions

C.D. participated in the data check and analysis, and drafted this manuscript; A.L. conceptualized and designed the study, revising it for important intellectual content. J.F., G.F., F.Y., and W.G. reviewed and revised the manuscript. C.S., Y.M., and Z.C. participated in the data check. All authors read and approved the final manuscript.

Data Availability Statement

The dataset supporting the conclusions of this article is not publicly available but can be available upon request and after approval by National Institute for Nutrition and Health, Chinese Center for Disease Control and Prevention, and the National Health Commission of China. Further enquiries can be directed to the corresponding author.

Supplementary Material

Supplementary data