Adiposity Mediates the Association of Objectively Measured Physical Activity with Cardiorespiratory Fitness in Children

Abstract

Background:

The relationship of physical activity (PA) to cardiorespiratory fitness (CRF) is well established in children. However, the extent to which adiposity affects this association remains unclear.

Objective:

The study aimed to explore whether the relationships of different PA intensities to CRF are explained by adiposity.

Methods:

Cross-sectional data were collected from 374 children (51.87% boys) aged 7–12 years. The time spent (min/day) in vigorous PA (VPA), moderate PA (MPA), light PA (LPA), and sedentary behavior was objectively measured using triaxial accelerometry. Height, weight, and waist circumference (WC) were objectively measured, from which the BMI was derived. The 20-meter shuttle run test was conducted to estimate maximal oxygen consumption [VO_2max, mL/(kg·min)]. Linear mixed models and mediation analysis with bootstrapping were used to analyze data.

Results:

VO_2max was positively associated with VPA [β = 0.143, 95% confidence interval (95% CI): 0.091 to 0.194], MPA (β = 0.051, 95% CI: 0.016 to 0.086), and moderate-to-vigorous PA (MVPA) (β = 0.052, 95% CI: 0.029 to 0.075), but not LPA or sedentary time. Both BMI and WC partially mediated the relationships of VPA and MVPA to VO_2max, with the percentage of the total effect mediated by adiposity ranging from 18.0% to 19.6%. Similar results were also observed among boys but not among girls.

Conclusions:

Only moderate or vigorous intensity of PA is favorably correlated with CRF in children. BMI and WC play a pivotal mediating role in these associations, especially in boys, suggesting that promoting higher intensity of PA might benefit children's CRF through reducing adiposity.

Introduction

Cardiorespiratory fitness (CRF), which was defined as the integrated ability to deliver oxygen to the mitochondria to support muscle activity during physical work,¹ has emerged as a holistic health indicator in children² and affects cardiovascular risk factors later in life.³ However, a systematic review demonstrated that only two-thirds of boys and half of girls had healthy CRF,⁴ and poor CRF in childhood has been found to stem well into adulthood.⁵ CRF was genetically determined, but also could be significantly influenced by nongenetic factors. Among these factors, physical activity (PA) and sedentary time (ST) have evidenced the most vital modifiable ones.⁶

Existing epidemiological literature has repeatedly recommended PA promotion and ST reduction to better develop CRF, and improvement in CRF produced by PA interventions has been deemed promising among children and adolescents.^7–9 Remarkably, a systematic review⁸ examining characteristics of PA interventions on CRF demonstrated that interventions were more effective in children with normal weight than in those with overweight/obesity. In addition, another study¹⁰ proposed that even physically active persons who are overweight could not achieve better than average fitness levels compared with their normal-weight peers, because of an unfavorable influence of adiposity on CRF in adolescents. These findings, therefore, conferred a great interest in disentangling which extent to adiposity affects the association between PA and CRF.

Previous evidence has recognized that PA is a common determinant of both fatness and CRF since increased PA decreases adiposity and benefits CRF development.¹¹ Moreover, adiposity is appraised as a robust influential factor for CRF.¹² Owing to excess body mass from abnormal fat accumulation, participants with overweight/obesity are burdened by the metabolic cost with higher oxygen demand during an unloaded task, and thus, predict poorer performance during sustained exercise (i.e., 20-meter shuttle run test).¹³ Therefore, one potential mechanism proposed to link reduced PA and prolonged ST with a low level of CRF was excess adiposity.

However, existing studies on whether adiposity acts as a mediator remained inconclusive. A previous review concluded that there is a lack of distinction of excessive body weight as cause or effect of low levels of PA and fitness in adolescents,¹⁴ while another subsequent study demonstrated that over a third of the association between total PA assessed with pedometers and CRF was mediated by the percentage of body fat (PBF) among adolescents.¹⁵ Furthermore, existing studies mainly focused on adolescents and limited to total PA rather than its intensity-specific influence. As primary school children are in a particularly critical period to promote a healthy lifestyle,¹⁶ a detailed characterization of the role of adiposity would be crucial to help achieve better CRF promotion for children.

Given a substantial decline in CRF levels among children and adolescents worldwide,¹⁷ particularly girls' CRF begins to progressively decline in mid-childhood,¹⁸ an understanding of its modifiable behaviors (i.e., PA and ST) and potential correlates (i.e., adiposity) may contribute to providing valuable implications for advancing specific intervention strategies. Also, a previous animal study revealed that there was intensity-specific effect of exercise intervention on CRF. However, in human, especially in children, narrow focus has limited to the benefit of top activity spectrum [i.e., moderate-to-vigorous PA (MVPA)],¹⁹ causing a substantial loss of information and limited understanding of its intensity-specific influence. To our knowledge, the mediating role of adiposity in combination with PA intensity-specific influence has not been previously investigated among children. Thus, this study was performed to (1) investigate the relationships of objectively measured PA intensity levels and ST with CRF in children aged 7–12 years, and (2) explore the mediating role of adiposity in these associations. We hypothesized that children with higher intensity of PA and less ST had better CRF, which might be explained by their lower level of adiposity.

Materials and Methods

Study Design and Participants

This cross-sectional study used baseline data from a school-based prospective cohort study, which was conducted between March and December in 2017 in Guangzhou, China (Clinical Trial Registration No. NCT03582709). The project had been approved by the Ethical Review Committee for Biomedical Research, School of Public Health, Sun Yat-sen University (L2016-010), and all parents or guardians of participants signed an informed written consent form voluntarily before the study. The study design and sampling procedure had been previously described elsewhere.²⁰ In brief, we invited 8324 students from 5 randomly selected primary schools to participate in the study, of whom 4991 students with their guardians agreed and completed a validated questionnaire.²¹ This questionnaire aimed to collect general information, and preliminarily investigate PA and ST level using the international PA questionnaire short form. Subsequently, according to the information on PA (MVPA ≥60 min/day or <150 min/week) and ST (gender-, age-specific ST ≥75% or <25% percentile), 637 participants were asked to wear an accelerometer and access CRF additionally.

After excluding individuals with no valid accelerometer data (n = 244; 38.3% of total), or those without CRF test (n = 32; 5.0% of total), a total of 374 subjects were enrolled eventually into the analysis. The mean ± standard deviation (SD) age of the participants was 9.17 ± 1.60 years (range: 7–12 years), and 51.87% were boys. In general, participants in our analysis had no significant differences in demographic characteristics compared with the sampling or agree-to-participating population (data not shown).

Measurements

Physical activity and sedentary time

PA and ST were assessed using a triaxial accelerometer GT3X (Actigraph, Pensacola, FL) worn on the right hip during waking hours for seven consecutive days, except for water-based activities. Data, sampled at 30 Hz, were collected starting at 6:00 am and ending at 11:59 pm using the unit of counts per minute (cpm). The accelerometer data files were reintegrated to 30-second epochs, and nonwear periods were identified (and excluded from further analysis) by scanning the data array for periods of ∼60 minutes of consecutive zeros (allowing for 2 minutes of nonzero interruptions)²² converted into gravity-corrected vector magnitude units. The inclusion criteria for accelerometer data were a valid wear time of a minimum of 10 h/day for ∼4 days (included three weekdays and one weekend). Data download, cleaning, and analyses were conducted using manufacture software ActiLife Version 6.13.3 (Actigraph). Application of the wear time inclusion criteria resulted in an analytical sample of 393 children (61.7% of consenting children), with no difference in descriptive characteristics from those excluded. Prediction equations of Evenson et al.²³ were used to identify cut-points for classifying activity into ST (≤100 cpm), light PA (LPA) (100–2295 cpm), moderate PA (MPA) (2296–4011 cpm), and vigorous PA (VPA) (≥4012 cpm). MVPA was calculated as the sum of the time spent on MPA and VPA. All values of PA intensity levels were expressed in min/day.

Assessment of anthropometry

Professionally trained doctors performed the anthropometric examinations, according to standardized procedures. Height was measured to the nearest millimeter with a portable stadiometer, with the children standing straight against the wall without shoes. Weight was measured to the nearest 0.1 kg with a lever-type weight scale when the children were in light clothing and without shoes. BMI was derived as weight in kilograms divided by the square of the height in meters (kg/m²), and BMI z-score was calculated according to the reference of the World Health Organization (WHO).²⁴ Waist circumference (WC) was measured to the nearest millimeter and located at the 1 cm above umbilicus with a measuring tape. According to percentile by age- and sex-specific cutoffs from WHO,²⁴ BMI ≥85th percentile was defined as overweight (included overweight and obesity), and BMI <85th percentile was defined as nonoverweight (included normal weight and underweight).

Cardiorespiratory fitness

CRF was estimated by predicted maximal oxygen consumption (VO_2max) based on the 20-meter shuttle run (20MSR) test using Léger's protocol.²⁵ A previous meta-analysis²⁶ showed that the criterion-related validity for estimating VO_2max was moderate to high [r_p = 0.78, 95% confidence interval (95% CI): 0.72 to 0.85] between 20MSR test (Léger's protocol) and standardized laboratory-based incremental test. In brief, participants were required to run between two lines 20 m apart while keeping pace with audio signals emitted from a prerecorded compact disk. The initial speed was 8.5 km/h and then increased by 0.5 km/h every minute. Participants were encouraged to keep running until they failed to reach the end lines concurrent with the audio signals on two consecutive occasions or stopped because of fatigue. We recorded the total laps (TL) and maximum run speed (MRS) completed as an indicator of his or her CRF. Estimations of VO_2max were obtained using Léger's equation²⁵: (VO_2max = 31.025 + 3.238 × MRS −3.248 × age + 0.1536 × MRS × age).

Covariates

Parents reported their monthly household income [0-4999 Renminbi (RMB, also named Chinese Yuan) (equivalent to 0–712 USD), 5000–7999 RMB (equivalent to 713–1140 USD), 8000–11,999 RMB (equivalent to 1141–1710 USD), ≥12,000 RMB (equivalent to ≥1711 USD), or refuse to disclose] and education level by the questionnaire. Parental education level was defined as the highest level of education completed by either parent (high school or below, college, or university or above). With the assistance of parents, children reported their food intakes, including the frequency and amount of fruit, vegetable, grain, fish, meat, fried food, milk, soybean, and sugar-sweetened beverage intake in the last 7 days. The average daily intake of these foods was calculated with the following formula: average daily intake = [days × (amount in each of those days)]/7. The questionnaires had been piloted and validated, and were found to have acceptable reliability and validity (data not shown). Data on menarche and spermatorrhea (yes vs. no) were collected by professionally trained doctors upon physical examination to estimate the initiation of puberty, and preliminarily reflect puberty status in children. In addition, accelerometer wear time was included as a covariate.

Statistical Analysis

Descriptive data are presented as mean ± SD for continuous variables or percentages for categorical variables. An independent t-test or chi-square test was used to compare descriptive data between boys and girls. Partial correlations were performed to evaluate the relationships among adiposity variables (BMI and WC), PA intensities (VPA, MPA, MVPA, and LPA), ST, and CRF (VO_2max, TL), adjusting for age, gender, parental education level, monthly household income, puberty initiation, food intake, and accelerometer wear time. Furthermore, the associations of PA intensities and ST with CRF were investigated with linear mixed-effects models, with schools fitted as random effects. Each PA intensity or ST was assessed using a separate model owing to high collinearity. Three models with increasing covariate adjustment were conducted. Model 1 controlled for the above covariates. In model 2 and model 3, we further adjusted for BMI or WC, respectively. As significant differences existed in PA, ST, and CRF between boys and girls, we evaluated effect modification by stratified analyses according to gender.

To evaluate whether the association between PA and CRF was mediated by adiposity, mediation analysis was conducted using the PROCESS macro version 3.1 presented by Preacher and Hayes.²⁷ In the unstandardized regression equation (ordinary least-squares regression), one of PA intensities was modeled as the predictor, CRF was modeled as the outcome, BMI or WC was modeled as the mediator, and the covariates mentioned above were modeled as covariates. The goal of mediation analysis was to determine the total (c) and direct effect (c′) of an independent variable on the dependent variable in each model. The coefficients a and b demonstrated the relationship of an independent variable to mediator and a mediator to outcome variable, respectively. The indirect effect (IE) was obtained from the products of coefficient (a × b). A significant IE through the mediator between the PA variable on CRF was determined if the 95% CI did not overlap zero. In this study, a 95% CI for each a × b product was obtained with 10,000 bootstraps resample. As we only found significant associations of VPA, MPA, and MVPA with VO_2max, mediation analyses were performed only for these relationships.

In addition, a sensitivity analysis was performed by replacing VO_2max with TL on the 20MSR test to reflect CRF as they are less subject to biases.⁹ To address the potential bias by the anthropometric measurements, we also performed a sensitivity analysis by using PBF to reflect adiposity, which was measured by electrical bioimpedance (Inbody 230) in a subsample of population (n = 137). All statistical analyses were conducted using SPSS software (SPSS 21; IBM), and tests were two sided and considered statistically significant at p < 0.05.

Results

Sample Characteristics

The general characteristics of the subjects are presented in Table 1. Boys had a significantly higher mean BMI, BMI z-score, WC, more time spent in all intensities of PA, and better performance in the 20MSR test, but less ST than girls. There is no significant difference in socioeconomic characteristics between boys and girls.

Table 1.

Characteristics of the Participants and Group Differences by Gender

Variables	Total (n = 374)	Boys (n = 194)	Girls (n = 180)	p
Age, years, %				0.765
7	19.95	20.00	19.89
8	18.62	20.00	17.13
9	19.68	18.46	20.99
10	16.76	18.46	14.92
11	15.96	13.85	18.23
12	9.04	9.23	8.84
Parental education level, %				0.660
High school or below	15.86	17.19	14.44
College	22.85	21.35	24.44
University or above	61.29	61.46	61.11
Monthly household income, %				0.699
≤5000 RMB	22.89	25.52	20.00
5000–8000 RMB	26.16	26.56	25.71
8000–12,000 RMB	19.35	17.71	21.14
12,000–15,000 RMB	20.16	19.79	21.57
Refuse to disclose	11.44	10.42	12.57
Anthropometrics
Height, cm	139.25 ± 11.90	138.97 ± 12.19	139.55 ± 11.61	0.645
Body mass, kg	33.84 ± 10.26	34.84 ± 10.63	32.76 ± 9.76	0.050
BMI, kg/m²	16.98 ± 2.95	17.59 ± 3.05	16.32 ± 2.70	<0.001
BMI z-score^a	0.04 ± 1.34	0.38 ± 1.32	−0.33 ± 1.27	<0.001
BMI categories,^a %				0.003
Nonoverweight	74.03	67.55	81.03
Overweight	25.97	32.45	18.97
WC, cm	59.17 ± 8.41	60.49 ± 9.11	57.66 ± 7.32	0.001
Physical activity
VPA, min/day	10.74 ± 6.82	12.99 ± 6.65	8.31 ± 6.15	<0.001
MPA, min/day	31.33 ± 11.47	36.13 ± 11.11	26.16 ± 9.45	<0.001
MVPA, min/day	42.07 ± 16.84	49.12 ± 16.23	34.47 ± 13.95	<0.001
LPA, min/day	261.88 ± 57.79	274.77 ± 57.82	247.99 ± 54.60	<0.001
ST, min/day	523.99 ± 71.87	502.55 ± 66.51	547.09 ± 70.43	<0.001
MVPA ≥60 min/day, %	15.69	25.64	4.97	<0.001
Accelerometer wear time, min/day	827.93 ± 56.10	826.44 ± 58.64	829.54 ± 53.35	0.594
Cardiorespiratory fitness
TL	25.77 ± 11.87	27.86 ± 13.83	23.53 ± 8.80	<0.001
MRS, km/h	9.87 ± 0.68	9.98 ± 0.77	9.74 ± 0.54	0.001
VO_2max,^b mL/(kg·min)	46.85 ± 3.35	47.46 ± 3.46	46.20 ± 3.10	<0.001

Data are presented as mean ± SD or percentage, and p-value denotes an independent t-test, or chi-square test calculated the difference between three groups.

Age- and sex-specific BMI cutoffs proposed by WHO inference 2007.

VO_2max values obtained with Leger's equation (VO_2max = 31.025 + 3.238 × maximum run speed −3.248 × age +0.1536 × maximum run speed × age).

LPA, light physical activity; MPA, moderate physical activity; MRS, maximum run speed in 20 meter shuttle run test; MVPA, moderate-to-vigorous physical activity; SD, standard deviation; ST, sedentary time; TL, total laps in 20 meter shuttle run test; VO_2max, maximal oxygen consumption; VPA, vigorous physical activity; WC, waist circumference; WHO, World Health Organization.

Associations of PA, ST, and Adiposity with CRF in Children

Results of partial correlations showed that VPA, MPA, and MVPA were positively, while BMI and WC were negatively, associated with VO_2max (all p < 0.05). There was no significant association of LPA or ST with VO_2max (See Supplementary Table S1). As linear mixed models depicted in Table 2, in general, VO_2max was positively related to VPA (p < 0.001), MPA (p = 0.004), and MVPA (p < 0.001), but not LPA or ST, when adjusted for covariates aforementioned in model 1. After additionally controlled for BMI (model 2) or WC (model 3), the magnitudes of these associations were consistently attenuated but remained significant. Stratified analysis by gender demonstrated similar changes in boys, except that the correlation between MPA and VO_2max was not significant after adjusting for WC. While in girls, neither PA nor ST was correlated with VO_2max when accounting for adiposity indices. However, there was no significant interaction between gender and any PA parameter (VPA/MVPA/MPA/LPA/ST) on CRF (all p_interaction > 0.05) (See Supplementary Table S2).

Table 2.

Mixed Linear Regression Coefficients (β) of Physical Activity Variables and Adiposity Indices for Cardiorespiratory Fitness

Variables	VO_2max
Variables	Model 1^a	Model 2^b	Model 3^c
Overall
VPA	0.143 (0.091 to 0.194)	0.112 (0.060 to 0.163)	0.109 (0.058 to 0.160)
MPA	0.051 (0.016 to 0.086)	0.044 (0.011 to 0.077)	0.041 (0.008 to 0.074)
MVPA	0.052 (0.029 to 0.075)	0.043 (0.020 to 0.065)	0.041 (0.018 to 0.061)
LPA	−0.002 (−0.009 to 0.005)	0.000 (−0.007 to 0.007)	0.000 (−0.007 to 0.007)
ST	−0.002 (−0.009 to 0.004)	−0.003 (−0.009 to 0.003)	−0.003 (−0.009 to 0.003)
Boys
VPA	0.179 (0.104 to 0.254)	0.124 (0.047 to 0.201)	0.122 (0.046 to 0.198)
MPA	0.068 (0.018 to 0.119)	0.049 (0.001 to 0.098)	0.044 (−0.005 to 0.093)
MVPA	0.065 (0.032 to 0.097)	0.046 (0.013 to 0.078)	0.044 (0.011 to 0.076)
LPA	−0.004 (−0.015 to 0.007)	−0.000 (−0.010 to 0.010)	−0.002 (−0.012 to 0.008)
ST	−0.002 (−0.012 to 0.008)	−0.003 (−0.013 to 0.006)	−0.002 (−0.011 to 0.008)
Girls
VPA	0.071 (0.003 to 0.139)	0.061 (−0.003 to 0.126)	0.063 (−0.003 to 0.147)
MPA	0.027 (−0.018 to 0.072)	0.028 (−0.015 to 0.070)	0.033 (−0.009 to 0.075)
MVPA	0.027 (−0.005 to 0.057)	0.025 (−0.003 to 0.054)	0.028 (−0.001 to 0.056)
LPA	0.001 (−0.008 to 0.008)	0.000 (−0.008 to 0.008)	0.001 (−0.007 to 0.009)
ST	−0.003 (−0.010 to 0.005)	−0.002 (−0.009 to 0.006)	−0.003 (−0.010 to 0.004)

Bold values are statistically significant (p < 0.05).

Model 1 adjusted for gender (only for overall), age, parental education level, monthly household income, puberty initiation, food intake, and accelerometer wear time.

Model 2 included the variables in model 1 and BMI.

Model 3 included the variables in model 1 and WC.

VO_2max, maximal oxygen uptake.

Simple Mediation Analysis

Overall (Fig. 1), mediation analyses showed significant IE in the relationship of VPA with VO_2max through adiposity, and association between MVPA and VO_2max as well. The percentage of the total effect mediated (% mediated) by adiposity in the correlations of VPA and MVPA with CRF ranged from 18.0 % to 19.6 %. These partial mediating effects were also observed in boys (% mediated ranged from 22.6 % to 28.6 %), but not in girls (Fig. 2). Sensitivity analysis demonstrated similar mediating effects and gender differences when VO_2max was replaced with TL as the CRF indicator in 20MSR test (see Supplementary Figs. S1–S3) (data not shown). To minimize potential bias from BMI and WC, we conducted a sensitivity analysis by using PBF to reflect adiposity and observed a similar mediating role of PBF in the association of VPA, MPA, and MVPA with VO_2max.

Figure 1.

Adiposity mediation models of the relationship between PA variables and VO_2max, controlling for gender, age, parental education level, monthly household income, puberty initiation, food intake, and accelerometer wear time. Analyses were performed by using PROCESS macro presented by Preacher and Hayes. a, b, c, and c′ are expressed as the unstandardized regression coefficient. IE was obtained with 10,000 bootstraps resample. *p < 0.05, **p < 0.01. IE, indirect effect; MPA, moderate physical activity; MVPA, moderate-to-vigorous physical activity; PA, physical activity; VO_2max, maximal oxygen consumption; VPA, vigorous physical activity; WC, waist circumference. A–C: BMI mediation models; D–F: WC mediation models.

Figure 2.

Mediation models of the relationship between PA variables and VO_2max stratified by gender, controlling for age, parental education level, monthly household income, puberty initiation, food intake, and accelerometer wear time. Data in roman type refer to boys. Data in italics refer to girls. *p < 0.05, **p < 0.01. A–C: BMI mediation models; D–F: WC mediation models.

Discussion

The main finding of this study was that BMI and WC mediate the association of objectively measured VPA, MPA, and MVPA with CRF in 7- to 12-year-old children. To the best of our knowledge, this is the first study to offer a complete understanding of the intensity-specific influence of PA on CRF through a plausible mechanism of adiposity.

Although CRF is genetically determined, it has also been described to be significantly influenced by modifiable factors, such as PA, ST, and obesity.²⁸ Prior resembling studies have substantiated the key roles of MPA and VPA⁶ and obesity²⁹ in CRF among children, but yet received little consensus regarding ST with CRF. To detail this, various measures or definitions, types, and study designs might be responsible for the controversial findings. Cliff et al.³⁰ corroborated that cross-sectional association between objectively measured ST and CRF was not significant, while van Ekris et al.³¹ revealed a prospectively detrimental correlation based on two studies examining television viewing. The present data coincided with findings of the prior research,³² where objectively measured PA, but not ST, was associated with CRF in 9- to 15-year-old children. These findings indicated that, in contrast to ST reduction, enhancing PA and preventing obesity may be more decisive to promote CRF in children. Notwithstanding, since reallocating equivalent ST into PA had been cross-sectionally and prospectively confirmed to benefit CRF,³³ the possible hazardous long-term effect of ST should not be ignored.

It has been proposed that improvements in CRF appear more responsive to increases in PA interventions intensity than session duration or frequency in adults.¹ Among children, although a review on observational and intervention studies has concluded that both VPA and MPA were significant predictors of CRF and adiposity, it did not involve in the influence of LPA.¹⁹ Likewise, in this study, when PA intensity levels were evaluated across all activity spectrum, per 1 minute increment in MPA and VPA predicts 0.051 and 0.143 mL/(kg·min) higher VO_2max, respectively, while LPA was not associated with CRF. These associations remained nearly unchanged when additionally controlled for adiposity. Despite different PA intensity definitions, our results supported the findings of Collings et al.,⁶ where they proposed PA above three metabolic equivalents (moderate intensity) was necessary to elicit benefits to CRF, and any right shift in the intensity distribution is likely to be beneficial. These results re-emphasized an optimal means by promoting the moderate or higher intensity of PA to maintain healthy CRF. More specifically, as VPA was the most effective³⁴ and time efficient⁶ in comparison with other PA intensity, an appropriate PA intensity, at least moderate, and particularly vigorous, would be the preferential alternative to promote the enhancement of fitness.

CRF develops with the increase of age in childhood, and the rate of growth in CRF decreases gradually during adolescence.¹ Specifically, CRF plateaus in boys at ∼16–18 years of age, whereas it begins to progressively decline in girls at ∼12–14 years of age,¹⁸ suggesting that childhood is a particularly critical period for promoting the development and reserve of CRF. Moreover, primary school children are at the beginning of compulsory education, a critical stage to establish a healthy lifestyle (including PA, ST, and diet).³⁵ Thus, the study evaluated adiposity as an underlying mechanism of the association of MPA and VPA with CRF, to comprehensively understand how different PA intensity levels benefit CRF and better advance more effective PA intervention among children. The results of mediation analysis supported the findings of Arngrimsson and Olafsdottir¹⁵ with a sample of 202 adolescents, in which they found that over a third of the association between overall PA and CRF was mediated by PBF measured by dual-energy X-ray absorptiometry. Moreover, our study expended the mediating role of adiposity in children, and further revealed that adiposity only partially explained the relations of MVPA and VPA with CRF. These findings revealed that children with a combination of elevated MVPA and low level of BMI or WC might exhibit a more prominent favorable CRF profile, compared with those inactive and overweight/obese. As PA is a fundamental means to energy balance, low levels of daily MVPA have been widely implicated as causes of excess body mass accumulation and the development of obesity,³⁶ which were potentially related to low fitness. Meanwhile, adiposity predicts poorer performance during sustained exercise owing to the metabolic cost with higher oxygen demand during an unloaded task.¹³ It is likely that the targets on increasing children's level of MPA or VPA may have a positive influence on their weight control, subsequently leading to improvement in CRF, though the underlying mechanisms have not been fully understood. The findings emphasized a significant implication of adiposity management when advancing PA interventions with moderate or higher intensity. Meanwhile, other weight reduction programs might also contribute to decreasing adiposity and consequently improve CRF in children.³⁷

There was some variation, however, in the associations between different PA, adiposity, and CRF depending on gender. Congruent with previous epidemiological studies,^38,39 boys have a higher level of CRF than their girl peers. They were more active and less sedentary, and seemed to respond stronger to moderate or higher intensity of PA intervention stimuli than girls.⁹ Besides, the IE of PA on CRF through adiposity was observed only in boys with adjustment for puberty initiation in this study, which provided a further understanding of gender differences in CRF. Previously, it has been reported that CRF levels in boys were impacted to a greater extent by adiposity than in girls,¹² suggesting a gender-dependent association between adiposity and CRF. Given this, it is likely that boys may benefit from their higher levels of PA, in which adiposity improvement produced by exercise also makes contributions to the higher CRF than those in girls. Besides, the previous study has reported that boys had 8%–18% higher aerobic fitness values than girls even before puberty, of which 3%–5% of the existing gender differences cannot be attributed to the difference in body composition, or PAs.⁴⁰ Collectively, our findings suggested that researchers should make effort to develop gender-specific strategies, to promote health in both boys and girls in the future.

This study was strengthened by integrating intensity-specific influence of PA and taking gender differences into account to explore the role of adiposity in the association between PA and CRF, with both exposure and outcome variables being objectively measured. However, certain limitations warrant consideration with the results. The primary limitation is the cross-sectional design, which prevents us from establishing the direction of the association and ruling out the bidirectional possibility. Second, since the participants were recruited from five primary schools located in a southern state, the findings are only generalizable to similar populations. Third, although PA and ST were measured by accelerometer objectively, there may also be misclassification in PA intensity levels with various processing criteria.⁴¹ In addition, VO_2max was assessed indirectly and could not be expressed relative to lean body mass. However, previous studies have demonstrated high reliability and moderate validity with the 20 m shuttle run test and prediction equation of Léger's presented satisfactory parameters.⁴² Likewise, BMI and WC were most widely used indicators but not direct measurements of adiposity.⁴³ Finally, even though we have considered many potential confounders that were plausibly related to exposures and outcomes, there still exist some unmeasured confounders or residual confounding. Evidence must be interpreted with these caveats in mind, and further prospective studies are needed to examine the effect of adiposity on the association of PA intensity levels and ST with CRF.

Conclusions

Objectively measured VPA, MPA, and MVPA were positively associated with CRF in children. Besides, BMI and WC play a pivotal mediating role in these associations, particularly in boys. Emphasis should be made in promoting VPA or MPA, which might positively influence BMI and WC, to achieve higher CRF development, especially for boys.

Footnotes

Acknowledgments

The authors thank all the participating children and their families. They acknowledge the staff of the Education Bureau, China, the Health Promotion Centre for Primary and Secondary Schools of Guangzhou Municipality, China, school doctors, and teachers for their great contributions to the success of the program.

Funding Information

This study was supported by the National Natural Science Foundation of China (Grant No. 81673193) and the Natural Science Foundation of Guangdong Province, China (Grant No. 2019A1515011462).

Author Disclosure Statement

No competing financial interests exist.

Supplementary Material

References

Ross

, Blair

, Arena

, et al. Importance of assessing cardiorespiratory fitness in clinical practice: A case for fitness as a clinical vital sign: A scientific statement from the American Heart Association. Circulation, 2016; 134:e653–e699.

Lang

, Tomkinson

, Janssen

, et al. Making a case for cardiorespiratory fitness surveillance among children and youth. Exerc Sport Sci Rev, 2018; 46:66–75.

Mintjens

, Menting

, Daams

, et al. Cardiorespiratory fitness in childhood and adolescence affects future cardiovascular risk factors: A systematic review of longitudinal studies. Sports Med, 2018; 48:2577–2605.

Tomkinson

, Lang

, Tremblay

, et al. International normative 20 m shuttle run values from 1 142 026 children and youth representing 50 countries. Br J Sports Med, 2017; 51:1545–1554.

Matton

, Thomis

, Wijndaele

, et al. Tracking of physical fitness and physical activity from youth to adulthood in females. Med Sci Sports Exerc, 2006; 38:1114–1120.

Collings

, Westgate

, Vaisto

, et al. Cross-sectional associations of objectively-measured physical activity and sedentary time with body composition and cardiorespiratory fitness in mid-childhood: The PANIC study. Sports Med, 2017; 47:769–780.

Pozuelo-Carrascosa

, Garcia-Hermoso

, Alvarez-Bueno

, et al. Effectiveness of school-based physical activity programmes on cardiorespiratory fitness in children: A meta-analysis of randomised controlled trials. Br J Sports Med, 2018; 52:1234–1240.

Braaksma

, Stuive

, Garst

, et al. Characteristics of physical activity interventions and effects on cardiorespiratory fitness in children aged 6–12 years—A systematic review. J Sci Med Sport, 2018; 21:296–306.

Minatto

, Barbosa

, Berria

, Petroski

. School-based interventions to improve cardiorespiratory fitness in adolescents: Systematic review with meta-analysis. Sports Med, 2016; 46:1273–1292.

10.

Fogelholm

, Stigman

, Huisman

, Metsamuuronen

. Physical fitness in adolescents with normal weight and overweight. Scand J Med Sci Sports, 2008; 18:162–170.

11.

Petridou

, Siopi

, Mougios

. Exercise in the management of obesity. Metabolism, 2019; 92:163–169.

12.

Fairchild

, Klakk

, Heidemann

, et al. Exploring the relationship between adiposity and fitness in young children. Med Sci Sports Exerc, 2016; 48:1708–1714.

13.

Norman

, Drinkard

, McDuffie

, et al. Influence of excess adiposity on exercise fitness and performance in overweight children and adolescents. Pediatrics, 2005; 115:e690–e696.

14.

Rauner

, Mess

, Woll

. The relationship between physical activity, physical fitness and overweight in adolescents: A systematic review of studies published in or after 2000. BMC Pediatr, 2013; 13:19.

15.

Arngrimsson

, Olafsdottir

. The relation between physical activity, fitness, and fatness in adolescents: A mediation analysis. Am J Hum Biol, 2016; 28:584–586.

16.

Gorely

, Nevill

, Morris

, et al. Effect of a school-based intervention to promote healthy lifestyles in 7–11 year old children. Int J Behav Nutr Phys Act, 2009; 6:5.

17.

Tomkinson

, Lang

, Tremblay

. Temporal trends in the cardiorespiratory fitness of children and adolescents representing 19 high-income and upper middle-income countries between 1981 and 2014. Br J Sports Med, 2019; 53:478–486.

18.

Armstrong

, Welsman

. Sex-specific longitudinal modeling of youth peak oxygen uptake. Pediatr Exerc Sci, 2019; 31:204–212.

19.

Parikh

, Stratton

. Influence of intensity of physical activity on adiposity and cardiorespiratory fitness in 5–18 year olds. Sports Med, 2011; 41:477–488.

20.

, Cai

, Gui

, et al. Effects of physical activity and sedentary behaviour on cardiometabolic risk factors and cognitive function in children: Protocol for a cohort study. BMJ Open, 2019; 9:e030322.

21.

, Lai

, Li

, et al. Study on the reliability and validity of physical health surveillance system questionnaire for primary school students in Guangzhou. Matern Child Health Care China, 2020; 35:1511–1516.

22.

Troiano

, Berrigan

, Dodd

, et al. Physical activity in the United States measured by accelerometer. Med Sci Sports Exerc, 2008; 40:181–188.

23.

Evenson

, Catellier

, Gill

, et al. Calibration of two objective measures of physical activity for children. J Sports Sci, 2008; 26:1557–1565.

24.

de Onis

, Onyango

, Borghi

, et al. Development of a WHO growth reference for school-aged children and adolescents. Bull World Health Organ, 2007; 85:660–667.

25.

Leger

, Mercier

, Gadoury

, Lambert

. The multistage 20 metre shuttle run test for aerobic fitness. J Sports Sci, 1988; 6:93–101.

26.

Mayorga-Vega

, Aguilar-Soto

, Viciana

. Criterion-related validity of the 20-m shuttle run test for estimating cardiorespiratory fitness: A meta-analysis. J Sports Sci Med, 2015; 14:536–547.

27.

Preacher

, Hayes

. Asymptotic and resampling strategies for assessing and comparing indirect effects in multiple mediator models. Behav Res Methods, 2008; 40:879–891.

28.

Teran-Garcia

, Rankinen

, Bouchard

. Genes, exercise, growth, and the sedentary, obese child. J Appl Physiol (1985), 2008; 105:988–1001.

29.

Tsiros

, Coates

, Howe

, et al. Adiposity is related to decrements in cardiorespiratory fitness in obese and normal-weight children. Pediatr Obes, 2016; 11:144–150.

30.

Cliff

, Hesketh

, Vella

, et al. Objectively measured sedentary behaviour and health and development in children and adolescents: Systematic review and meta-analysis. Obes Rev, 2016; 17:330–344.

31.

van Ekris

, Altenburg

, Singh

, et al. An evidence-update on the prospective relationship between childhood sedentary behaviour and biomedical health indicators: A systematic review and meta-analysis. Obes Rev, 2016; 17:833–849.

32.

Joensuu

, Syvaoja

, Kallio

, et al. Objectively measured physical activity, body composition and physical fitness: Cross-sectional associations in 9- to 15-year-old children. Eur J Sport Sci, 2018; 18:882–892.

33.

Santos

, Marques

, Minderico

, et al. A cross-sectional and prospective analyse of reallocating sedentary time to physical activity on children's cardiorespiratory fitness. J Sports Sci, 2018; 36:1720–1726.

34.

Aadland

, Kvalheim

, Anderssen

, et al. The multivariate physical activity signature associated with metabolic health in children. Int J Behav Nutr Phys Act, 2018; 15:77.

35.

Day

, Sahota

, Christian

. Effective implementation of primary school-based healthy lifestyle programmes: A qualitative study of views of school staff. BMC Public Health, 2019; 19:1239.

36.

Elder

, Roberts

. The effects of exercise on food intake and body fatness: A summary of published studies. Nutr Rev, 2007; 65:1–19.

37.

van Leeuwen

, Andrinopoulou

, Hamoen

, et al. The effect of a multidisciplinary intervention program for overweight and obese children on cardiorespiratory fitness and blood pressure. Fam Pract, 2019; 36:147–153.

38.

Potter

, Spence

, Boule

, et al. Associations between physical activity, screen time, and fitness among 6- to 10-year-old children living in Edmonton, Canada. Appl Physiol Nutr Metab, 2017; 42:487–494.

39.

Hussey

, Bell

, Bennett

, et al. Relationship between the intensity of physical activity, inactivity, cardiorespiratory fitness and body composition in 7-10-year-old Dublin children. Br J Sports Med, 2007; 41:311–316.

40.

Dencker

, Thorsson

, Karlsson

, et al. Gender differences and determinants of aerobic fitness in children aged 8–11 years. Eur J Appl Physiol, 2007; 99:19–26.

41.

Migueles

, Cadenas-Sanchez

, Ekelund

, et al. Accelerometer data collection and processing criteria to assess physical activity and other outcomes: A systematic review and practical considerations. Sports Med, 2017; 47:1821–1845.

42.

Menezes

, Jesus

, Leite

. Predictive equations of maximum oxygen consumption by shuttle run test in children and adolescents: A systematic review. Rev Paul Pediatr, 2019; 37:241–251.

43.

Javed

, Jumean

, Murad

, et al. Diagnostic performance of body mass index to identify obesity as defined by body adiposity in children and adolescents: A systematic review and meta-analysis. Pediatr Obes, 2015; 10:234–244.

Supplementary Material

Please find the following supplemental material available below.

For Open Access articles published under a Creative Commons License, all supplemental material carries the same license as the article it is associated with.

For non-Open Access articles published, all supplemental material carries a non-exclusive license, and permission requests for re-use of supplemental material or any part of supplemental material shall be sent directly to the copyright owner as specified in the copyright notice associated with the article.

0.00 MB

0.16 MB

0.02 MB