Physical Activity, Screen Time and Body Composition in Adolescents: A Population-based Prospective Cohort Study

doi:10.21203/rs.3.rs-2564256/v1

Download PDF

Article

Physical Activity, Screen Time and Body Composition in Adolescents: A Population-based Prospective Cohort Study

https://doi.org/10.21203/rs.3.rs-2564256/v1

This work is licensed under a CC BY 4.0 License

Version 1

posted

You are reading this latest preprint version

Background

Physical activity and sedentary behaviors have been shown to have varying health benefits. Nevertheless, it is unclear how physical activity and screen time affect body composition during adolescence. This study examined the associations of physical activity and screen time with overall and abdominal fat depositions in general pediatric population.

Methods

In a population-based prospective cohort study, among 3.258 adolescents aged 13 years, physical activity and screen time were assessed via self-report questionnaires. Body mass index, dual-energy X-ray absorptiometry-based measures (i.e., fat mass index (FMI), lean body mass index and android/gynoid fat mass ratio (A/G ratio)), and magnetic resonance imaging-based measures (i.e., abdominal subcutaneous fat index, abdominal visceral fat index and abdominal visceral/subcutaneous fat mass ratio) were obtained.

Results

After adjusting for social-demographic and growth-related factors, a change of per additional hour per day spent in physical activity was associated with lower FMI (Z score difference, -0.08; 95% CI, -0.14 to 0.01), A/G ratio (-0.10 [-0.18; -0.03]) and abdominal visceral fat index (-0.11 [0.20; -0.02]), and higher lean body mass index (0.15 [0.08; 0.22]). A change of per additional hour per day spent in screen time was also associated with higher body mass index (0.02 [0.01; 0.04]), FMI (0.02 [0.01; 0.03]), A/G ratio (0.03 [0.02; 0.04]), abdominal subcutaneous fat index (0.03 [0.02; 0.05]) and abdominal visceral fat index (0.03 [0.01; 0.05]), and lower abdominal visceral/subcutaneous fat mass ratio (-0.02 [-0.04; -0.01]).

Conclusion

Adolescents with higher physical activity and lower screen time had lower levels of adiposity both at the general and visceral levels. This information may be used for future work to develop targeted intervention strategies that improve body composition among adolescents.

Health sciences/Health care/Paediatrics

Health sciences/Medical research/Epidemiology

Health sciences/Health care/Public health/Epidemiology

Epidemiology

pediatric obesity

magnetic resonance imaging

lifestyle behaviors

public health

Childhood overweight and obesity have increased rapidly worldwide in the last few decades, and they have harmful psychological and physical health consequences that persist into adulthood (1). In this context, the interest in promoting more physical activity and less screen time has grown in response to this problem (2). However, a better understanding of the associations of physical activity and screen time with obesity is needed to design more effective interventions in the future (3). For instance, prior studies have reported inconsistent associations of physical activity and screen time with overweight and obesity in children, and most studies assessed body mass index (BMI) rather than detailed fat distribution measured using advanced imaging techniques, such as magnetic resonance imaging (MRI) (4, 5). One of the limitations of using BMI to determine obesity is that it does not distinguish between fat and lean mass, which might have different implications on health.

In particular, accumulation of abdominal subcutaneous fat and abdominal visceral fat can lead to insulin resistance, type 2 diabetes, and cardiovascular diseases in both adults and children (6). Studying the associations of physical activity and screen time with general and abdominal obesity measures simultaneously might provide more insights into the consequences of lifestyle behaviors on obesity. Most of previous studies have been performed in children, while adolescence is an important period when several behaviors are established ⁸. In one randomized controlled trail of children with overweight/obesity aged 10 years, the effect of a family-based, multicomponent intervention program on total and abdominal fat was examined (7). This study showed that total and abdominal fat were reduced in both intervention and control groups. Another randomized controlled trail examined the effect of different doses of aerobic training on diabetes risk in 9-year-old sedentary children with overweight. This study showed that visceral fat area via MRI was reduced in the high- and low-does vs control groups (8). Nevertheless, studies with large sample sizes, a measure of unstructured physical activity and screen time, and measures of fat distribution are needed especially in general adolescent population.

Therefore, the current study aimed to examine the associations of physical activity and screen time with adolescent body composition measures estimated by BMI, dual-energy X-ray absorptiometry (DXA) and MRI based primarily on cross-sectional data (i.e., physical activity and screen time at age 13 years), as well as longitudinal data (i.e., physical activity and screen time at ages 6 and 10 years). We hypothesized that at the age of 13 years, higher levels of physical activity and shorter screen time would be associated with lower total body fat mass and lower abdominal fat depots, including decreased abdominal subcutaneous and visceral fat mass.

Design This study was embedded in the Generation R Study, a population-based prospective cohort from fetal life onward in Rotterdam, The Netherlands (9). The study has been approved by the Medical Ethical Committee of the Erasmus MC, University Medical Centre in Rotterdam (MEC-2012-165-NL40020.078.12). Women who lived in Rotterdam, The Netherlands, with a delivery date between April 2002 and January 2006, were eligible for participation. Written informed consent was obtained from all participants and their parents or legal representatives. Adolescents visited our research center at the median age of 13 years. From 6,842 participants, twins (n = 172) and adolescents with missing information on physical activity and screen time (n = 3,100) were excluded. Also, children with missing information on BMI, DXA and MRI (n = 312) were excluded. This resulted in 3,258 children for the current analyses (Fig. 1). It should be pointed out that only 2,121 children had MRI-based fat segmentation data available, as in 1,137 children there was no MRI scan acquired (n = 1,092), MRI sequences acquired were incomplete or of insufficient quality (n = 2), or unsatisfactory fat segmentation was achieved judged by visual evaluation (n = 43).

Physical activity and screen time Information on physical activity and screen time was obtained by self-reported questionnaires at age 13 years. This questionnaire was designed based on the Activity Questionnaire for Adults and Adolescents (AQuAA) (10) and the Health Behaviours in School-Aged Children (HBSC) questionnaire (11). The AQuAA questionnaire has been shown to be reproducible although there was low validity measured with the use of accelerometers (10). The widely used HBSC questionnaire, using the Multistage Fitness Test for physical activity and a television watching diary for screen time, has demonstrated adequate reliability and validity (12, 13). To assess the total amount of physical activity, children reported their strenuous physical activity and walking or cycling to and from school respectively. Children were asked to response strenuous physical activity (i.e., ‘How many hours a week do you usually exercise in your spare time in a way that makes you breathless?’). The response categories for strenuous physical activity included: ‘never’, ‘less than 1 hour’, ‘1–2 hour’, ‘2–3 hour’, ‘3–4 hour’, ‘4–6 hour’, ‘more than 7 hours’. The average time per day spent during leisure time to activities that make the respondent running out of breath was calculated using the following formula: hours per day spent on strenuous physical activity / 7. Children also reported the way of traveling to and from school (i.e., ‘How do you usually get from home to school?’, ‘How do you usually go home from school?’) and the travelling time (‘How long does it take you to travel home from school?’). The response categories for the way traveling to and from school included: ‘walk’, ‘bicycle’, ‘bus, train, tram, metro or boat’, ‘car, motorcycle or scooter’, ‘others’. The response categories for the traveling time included: ‘less than 5 minutes’, ‘5 to 15 minutes’, ’15 to 30 minutes’, ’30 minutes to 1 hour’, ‘more than 1 hour’. The average time per day for walking or cycling to and from school was calculated using the following formula: [(hours per time walking or cycling to school + hours per time walking or cycling from school) * 5] / 7. A total daily physical activity time was calculated by adding up children’s strenuous physical activity and walking or cycling to and from school.

Regarding screen time, participants were asked, for weekdays and weekend days separately, for the number of hours during leisure time they played computer games, used a computer for chatting, web-browsing, e-mail or homework, and watched television (including videos/DVDs). A total daily screen time was calculated by using the following formula: [(hours per weekday) * 5 + ((hours per weekend day) *2)] / 7. Information of questionnaires on physical activity and screen time at ages 6 and 10 years was used for additional analysis (Supplemental Appendix 1 and Supplemental Appendix 2).

Body composition measures Children visited our research center at age 13 years, their anthropometrics were measured, and DXA and MRI examinations were performed. DXA-based measures included fat mass index (FMI), lean body mass index (LBMI), android/gynoid fat mass index (A/G ratio), and MRI-based measures included abdominal subcutaneous fat index and abdominal visceral fat index. Particularly, a deep learning-based image segmentation method, using a 2D-Competitive Dense Fully Convolutional Network (CDFNet) as proposed in the FatSegNet method (14), was applied to quantify abdominal subcutaneous fat and visceral fat on MRI scans automatically (Fig. 2). DXA and MRI measurements at age 10 years were used for additional adjustment. Detailed information is described in the Supplemental methods.

Covariates Information on maternal age, pre-pregnancy BMI, educational level (low: no education or primary education; middle: secondary Phase 1 or 2 finished; high: vocational or university degree), parity (nulliparous or multiparous), psychiatric symptoms (yes; no), smoking (yes; no) were obtained from questionnaires during pregnancy.

Child’s age, sex, gestational age at birth, and birth weight were obtained from midwife and hospital registries at birth. Child ethnic background was based on parental countries of birth obtained from questionnaire at enrollment. Ever breastfeeding (yes; no) was obtained from Questionnaires after birth. Puberty stage at the age of 13 years were obtained based on self-reported pubic hair growth according to the Tanner scale (15, 16). According to the distribution within our cohort, puberty status was classified as early puberty (Tanner stages 1–3), or advanced puberty (Tanner stages 4–5). Children’s breakfast skipping behavior was assessed by single-item question in a parent-reported questionnaire. Parents reported the times that their children have breakfast (more than a glass of milk or fruit juice) for weekdays and weekend days separately. We defined breakfast skipping behavior as present if breakfast skipping occurred at least once in a week versus never.

Statistical analysis Characteristics of children included and not included in the study were assessed using Student’s t-tests, Mann-Whitney U tests, and Chi-square tests. We used linear and logistic regression models to examine the associations of physical activity and screen time with body composition at age 13 years. We constructed three sets of models. Model 1 (basic model) was adjusted for child’s sex and age. In model 2 (main model), we additionally adjusted for maternal age, prepregnancy BMI, educational level, parity, psychiatric symptoms (GSI) and smoking during pregnancy, and child’s gestational age at birth, birth weight, ethnic background, breastfeeding, puberty status and breakfast skipping. Potential confounders were selected according to literature, if they were related to both physical activity, screen time and respiratory outcomes, or if the effect estimate of the unadjusted analyses changed ≥ 10% when we additionally adjusted for a confounder. For better visualization of causal assumptions and potential confounders, we created a Directed Acyclic Graph using DAGitty version 2.3 (Supplemental Fig. 1). Consequently, all the potential confounders were included in our models. In model 3 (reverse causation model), we additionally adjusted for body composition outcomes at age 10 years to partly reduce in part the chance of reverse causation based on the main model. In addition, physical activity and screen time were categorized into tertile groups (low, middle and high). Interaction terms were tested in the main model. We tested whether the associations of physical activity and screen time with body composition were modified by child’s sex and puberty stage. If the interaction term was significant (P value < 0.10), we performed further stratified analyses.

Some additional analyses were performed. We excluded adolescents of non-European ethnic background because of their potential different body composition based on our main model in our sensitivity analyses. Longitudinal associations of physical activity and screen time at ages 6 and 10 years with body composition measure at age 13 years were examined separately. Not normally distributed outcomes measures (FMI, A/G ratio, abdominal subcutaneous fat index, abdominal visceral fat index and V/S ratio) were log-natural transformed. To enable comparison of effect estimates, we constructed Z score of outcomes. We tested for non-linearity of the associations via applying natural cubic splines. Missing data in covariates (ranging from 0 to 27.7%) were imputed using the Fully Condition Specification method with ten imputations (17). All measures of associations were presented as pooled estimates from the imputed datasets. Statistical analyses were performed using SPSS, version 28.0 for Windows (IBM Crop., Armonk, NY, USA) and R statistical software, version 4.1.2 (R Foundation for Statistical Computing).

Participants characteristics Table 1 shows the maternal and child characteristics. Of the 3,258 children in the cohort, 434 (13.3%) of the adolescents had overweight/obesity at age 13 years. Non-response analyses showed that children included in the analyses were more with European ethnic background, had a higher gestational age at birth, a higher birth weight and more likely to be the first born of their mothers. Also, children included in the analyses had mothers with higher educational level, who more likely were non-smokers during pregnancy (P value < 0.05; Supplemental Table 1). The correlation coefficients of BMI with FMI, LBMI, A/G ratio, abdominal subcutaneous fat index, abdominal visceral fat index and abdominal visceral/subcutaneous fat mass ratio (V/S ratio) were moderate to strong (ranging from − 0.43 to 0.76), and the correlation coefficient between physical activity and screen time was weak (-0.05) (Supplemental Fig. 2).

Table 1

Characteristics of children and their mothers
	n = 3,258^a
Maternal characteristics
Pre-pregnancy age, mean (SD), y	31.7 (4.6)
Pre-pregnancy BMI, median (IQR), kg/m^2b	22.3 (18.2–33.4)
Educational level
Lower	1,285 (41.7)
Higher	1,794 (58.3)
Parity
Nullipara	1,874 (59.4)
Multipara	1,280 (40.6)
Psychiatric symptoms (GSI)
No	2,409 (94.3)
Yes	146 (5.7)
Smoking during pregnancy
No	2,305 (79.1)
Yes	609 (20.9)
Birth and infant characteristics
Sex
Male	1,532 (47.0)
Female	1,726 (53.0)
Gestational age at birth, median (IQR), weeks	40.1 (35.6–42.3)
Birth weight, mean (SD), grams	3459 (558)
Ethnic background
European	2,447 (75.8)
Non-European	780 (24.2)
Ever breastfeeding
No	201 (7.3)
Yes	2,560 (92.7)
Adolescence characteristics
Age, mean (SD), y	13.9 (0.6)
Puberty status
Early puberty	1,173 (49.8)
Advanced puberty	1,182 (50.2)
Breakfast skipping
No	2,634 (84.0)
Yes	502 (16.0)
Physical activity, median (IQR), hours/day	0.7 (0.1–1.8)
Screen time, median (IQR), hours/day	4.4 (1.4–10.9)
BMI^b, mean (SD), kg/m²	19.6 (3.2)
Underweight	357 (12.7)
Normal weight	2,467 (75.7)
Overweight	356 (10.9)
Obese	78 (2.4)
Total fat mass, median (IQR), g	12,034 (6,250 − 30,179)
Lean body mass, mean (SD), g	37,115 (6,227)
A/G ratio, median (IQR)	0.24 (0.15–0.48)
Abdominal subcutaneous fat, median (IQR), g	806 (246-3,736)
Abdominal visceral fat, median (IQR), g	227 (90–799)
V/S ratio, median (IQR)	0.28 (0.13–0.59)
Abbreviations: A/G ratio, android/gynoid fat mass ratio; BMI, body fat mass; V/S ratio, abdominal visceral/subcutaneous fat mass ratio.
^a Characteristics are based on observed, not imputed data.
^b Calculated as weight in kilograms divided by height in meters squared.

Physical activity, screen time and body composition at age 13 years Fig. 3 and Supplemental Table 2 show the change in Z scores of body composition per additional hour per day spent in physical activity or screen time at age 13 years. After adjustment of confounders, a change of per additional hour per day spent in physical activity was associated with lower FMI, A/G ratio and abdominal visceral fat index (Z-score difference (95% CI): -0.11 (-0.20, -0.02) to -0.08 (-0.14, -0.01)), and higher LBMI (Z-score difference (95% CI): 0.15 (0.08, 0.22)). After further adjustment for body composition at age 10 years, the association of physical activity with LBMI did not remain but the direction was the same (Supplemental Table 2). Longer screen time was associated with higher BMI, FMI, A/G ratio, abdominal subcutaneous fat index and abdominal visceral fat index (Z-score difference (95% CI): 0.02 (0.01, 0.03) to 0.03 (0.02, 0.05)). Longer screen time was associated with lower V/S ratio (Z-score difference (95% CI): -0.02 (-0.04, -0.01)). After additional adjustment for body composition at age 10 years, the results were comparable, only the association of screen time with V/S ratio was not significant (Supplemental Table 2). The basic model showed comparable results with the main model, although the Z scores were larger (Fig. 3 and Supplemental Table 2).

The association between categorical physical activity and screen time and body composition showed that high group of physical activity were associated with lower A/G ratio and abdominal visceral fat index (Z-score difference (95% CI): -0.10 (-0.18, -0.01), -0.12 (-0.23, -0.02), respectively), but with higher LBMI (Z-score difference (95% CI): 0.14 (0.05, 0.22)) as compared to low group of physical activity (Table 2). Those who had a low group of screen time had a lower BMI, FMI, A/G ratio, abdominal subcutaneous fat index and abdominal visceral fat index (Z-score difference (95% CI): -0.16 (-0.25, -0.07) to -0.09 (-0.17, -0.01)), but with higher V/S ratio (Z-score difference (95% CI): 0.14 (0.05, 0.24)) as compared to those who had a high group of screen time (Table 2).

Table 2

Associations of physical activity and screen time categorized into tertiles with body composition at 13 years^a
	BMI Z-score difference (95% CI) n = 3,258	FMI Z-score difference (95% CI) n = 3,086	LBMI Z-score difference (95% CI) n = 3,086	A/G ratio Z-score difference (95% CI) n = 3,086	Abdominal subcutaneous fat index Z-score difference (95% CI) n = 2,121	Abdominal visceral fat index Z-score difference (95% CI) n = 2,121	V/S ratio Z-score difference (95% CI) n = 2,121
Physical activity (basic model)
Low (n = 1,066)	Reference	Reference	Reference	Reference	Reference	Reference	Reference
Middle (n = 958)	-0.09 (-0.19, 0.00)	-0.10 (0.19, 0.02)	-0.06 (-0.14, 0.03)	-0.09 (0.18, 0.00)	-0.12 (0.21, 0.02)	-0.11 (0.21, 0.00)	0.06 (-0.04, 0.16)
P value	0.058	0.013	0.217	0.043	0.024	0.046	0.239
High (n = 1,039)	-0.13 (0.22, 0.04)	-0.23 (0.31, 0.15)	-0.01 (-0.09, 0.08)	-0.25 (0.34, 0.17)	-0.26 (0.36, 0.16)	-0.25 (0.35, 0.14)	0.13 (0.04, 0.23)
P value	0.005	< 0.001	0.880	< 0.001	< 0.001	< 0.001	0.007
Physical activity (main model)
Low (n = 1,066)	Reference	Reference	Reference	Reference	Reference	Reference	Reference
Middle (n = 958)	0.05 (-0.04, 0.14)	0.01 (-0.07, 0.09)	0.05 (-0.04, 0.14)	0.02 (-0.07, 0.11)	0.02 (-0.08, 0.12)	-0.01 (-0.12, 0.09)	-0.06 (-0.15, 0.04)
P value	0.269	0.787	0.276	0.621	0.676	0.808	0.243
High (n = 1,039)	0.06 (-0.03, 0.14)	-0.07 (-0.15, 0.01)	0.14 (0.05, 0.22)	-0.10 (0.18, 0.01)	-0.08 (-0.18, 0.01)	-0.12 (0.23, 0.02)	-0.02 (-0.12, 0.07)
P value	0.222	0.079	0.002	0.027	0.091	0.020	0.609
Screen time (basic model)
High (n = 1,031)	Reference	Reference	Reference	Reference	Reference	Reference	Reference
Middle (n = 1,055)	-0.14 (0.23, 0.04)	-0.13 (0.21, 0.05)	-0.06 (-0.15, 0.03)	-0.17 (0.26, 0.08)	-0.21 (0.31, 0.11)	-0.13 (0.23, 0.02)	0.20 (0.11, 0.30)
P value	0.004	0.002	0.173	< 0.001	< 0.001	0.017	< 0.001
Low (n = 1,039)	-0.27 (0.36, 0.17)	-0.21 (0.30, 0.13)	-0.14 (0.23, 0.05)	-0.26 (0.34, 0.17)	-0.31 (0.41, 0.21)	-0.22 (0.33, 0.12)	0.26 (0.16, 0.35)
P value	< 0.001	< 0.001	0.002	< 0.001	< 0.001	< 0.001	< 0.001
Screen time (main model)
High (n = 1,031)	Reference	Reference	Reference	Reference	Reference	Reference	Reference
Middle (n = 1,055)	-0.02 (-0.11, 0.07)	-0.03 (-0.11, 0.05)	0.02 (-0.07, 0.10)	-0.08 (-0.16, 0.01)	-0.11 (0.20, 0.01)	-0.05 (-0.15, 0.05)	0.12 (0.03, 0.21)
P value	0.659	0.403	0.702	0.077	0.026	0.325	0.010
Low (n = 1,039)	-0.11 (0.20, 0.02)	-0.09 (0.17, 0.01)	-0.04 (-0.13, 0.04)	-0.13 (0.22, 0.04)	-0.16 (0.25, 0.07)	-0.11 (0.21, 0.00)	0.14 (0.05, 0.24)
P value	0.013	0.026	0.323	0.004	0.001	0.043	0.003
Abbreviation: A/G ratio, android/gynoid fat mass ratio; BMI, body mass index; FMI, fat mass index; LBMI, lean body mass index; V/S ratio, abdominal visceral/subcutaneous fat mass ratio.
^a Regression coefficients are change in Z-scores with 95% confidence interval (95% CI), derived from linear regression models. The basic model was adjusted for child’s age and sex. The main model was additionally adjusted for maternal pre-pregnancy age, pre-pregnancy BMI, educational level, parity, psychiatric symptoms and smoking during pregnancy, and child’s gestational age at birth, birth weight, ethnicity background, ever breastfeeding, puberty stage and breakfast skipping.

No statistically significant interactions of physical activity or screen time with child’s sex or pubertal stage were observed in the associations with body composition outcomes (P value > 0.10).

Additional analyses When we restricted our analysis to include only children of European ethnicity, we generally observed similar directions and strengths of the effects (Supplemental Table 3). Most of the associations of physical activity at ages 6 and 10 years with body composition outcomes at age 13 years lost significance (Supplemental Table 4, Supplemental Table 5, Supplemental Table 6 and Supplemental Table 7). The associations of screen time at ages 6 and 10 years with body composition measures at age 13 years were generally similar with the main cross-sectional analyses at age 13 years (Supplemental Table 4 and Supplemental Table 6). When screen time was categorized, the association of screen time at age 6 years with the body composition outcomes at age 13 years remained significance (Supplemental Table 5) compared to the results as continuous variables (Supplemental Table 4). However, the categorical screen time at age 10 years was no longer statistically significant (Supplemental Table 7) compared to the results as continuous variable (Supplemental Table 6). There was no indication of nonlinearity for the associations of physical activity or screen time with body composition outcomes (results not shown). As our analysis was hypothesis-driven and outcomes correlated strongly, we did not adjust for multiple testing (18).

The current study was designed to examine the possible associations of physical activity and screen time with childhood obesity in adolescents, taken from the population-based Generation R Study. Our results demonstrated associations of lower physical activity level and longer screen time with higher FMI, and extended these findings to higher A/G ratio and abdominal visceral fat that have been strongly linked to increased risk for cardiovascular diseases and type 2 diabetes (19, 20).

Comparison with previous studies Previous systematic reviews showed that physical activity had a protective effect against body fat in adolescence (2, 4, 21, 22), and longer screen time may contribute to higher level of body fat among youth, although the effect estimates of the associations were relatively small (5, 23–25). In the current study, a significant association was found between screen time and BMI, but not between physical activity and BMI. We found a similar relationship of physical activity and screen time with FMI as previous systematic reviews (2, 4, 5, 21–25). It was unknown to date whether physical activity or screen time is associated with A/G ratio. We found that higher levels of physical activity were associated with a lower A/G ratio, and that longer screen time was associated with higher A/G ratio. In addition, several studies have reported that higher levels of physical activity were associated with higher LBMI, which is consistent with our findings (26–31).

When we look at abdominal fat distribution as outcomes in our study, there was no significant association between physical activity and abdominal subcutaneous fat index, which has been concluded differently by other researchers. Two studies found significant associations of higher levels of physical activity with lower abdominal subcutaneous fat index (32, 33), but two other studies did not (34, 35). We also found higher levels of physical activity were associated with lower abdominal visceral fat index, which was in accordance with previous studies (32, 35, 36). Regarding the association between screen time and abdominal fat distribution, only one prior study examined the relationship between screen time and abdominal visceral fat and found it was significant, which is similar to our findings (37). We found that longer screen time was associated with lower V/S ratio. In adults, it has been reported that the V/S ratio was correlated with adverse cardio-metabolic risk factors (38). Our results appeared opposite in adolescents. It is plausible that V/S does not manifest its adverse effects until adolescents become adults. According to current systematic reviews, no causal relationship was inferred between physical activity and screen time and obesity due to heterogeneity (2, 4, 5, 21–25, 39). No study has examined the longitudinal relationship between physical activity or screen time with detailed abdominal fat distribution. We found similar directions of associations between screen time at ages 6 and 10 years with body composition measures at age 13 years, as our main analysis relate to screen time at age 13 years and body composition at age 13 years.

Interpretation of the results The current findings highlight the importance of adequate physical activity time to reduce specific depots of abdominal fat in adolescents. To test our hypothesis, we used not only BMI, but also DXA- and MRI-based measures to assess the overall body composition, including total body fat and muscle mass, and abdominal fat distribution, including abdominal subcutaneous fat and visceral fat. Obesity, especially abdominal or central obesity, is an established risk factor for cardiovascular disease, type 2 diabetes and cancer (19, 20, 38). Visceral adipose tissue is known for being more strongly associated with inflammatory response, glucose and lipid profiles in animal models and humans than subcutaneous adipose tissue (38). In line with our expectations, higher levels of physical activity were associated with a reduction in total body fat and abdominal visceral fat, as well as an increase in total lean body mass. It was also expected that higher levels of physical activity might be related to lower abdominal subcutaneous fat since physical activity was previously associated with total fat mass, and a strong correlation existed between total body fat and abdominal subcutaneous fat (40). However, this association was found to be non-significant in our study.

In our study, we found significant associations of screen time with BMI and FMI but not with LBMI. Also, the associations between screen time and body composition measures including BMI, FMI, A/G ratio, abdominal subcutaneous fat index, abdominal visceral fat index and V/S ratio were a bit stronger compared to the associations of physical activity with these body composition measures. This may be due to the fact that our questionnaire could not take into account the intensity of physical activity. It is worth noting, though, that previous studies reported that increased TV watching time is associated with more snacking or consumption of unhealthy food caused by advertisements, thus leading to a high level of energy intake (41, 42). Future studies are needed to explore the possible pathway between screen time and body composition outcomes. It is crucial to realize that, to some extent, any change either in physical activity or in screen time may partly be replaced by, or compensated with its counterpart (43–45). There is currently insufficient evidence to support the effectiveness of particular strategies in preventing obesity, even though it seems logical to argue that sitting for long periods of time might be counter-acted by strategies promoting physical activity (44, 45).

Strengths and limitations The major strength of this study was that it is embedded in a population-based, prospective cohort study with a large sample size. To characterize the abdominal fat distribution, we used MRI, currently the most accurate available non-invasive technique, to measure abdominal subcutaneous and visceral fat compartments. Furthermore, we considered several potential covariates, such as puberty status and breakfast skipping. To our knowledge, this is the first study to examine physical activity and screen time with both DXA- and MRI-based measures of general and visceral adiposity among adolescents. Our results might bring important implications for public health intervention with regard to physical activity and screen time in order to develop and maintain a healthy body composition distribution in adolescence. However, some methodological limitations need to be addressed. Since the main analyses in our study were cross-sectional, we cannot determine causality. We therefore additionally adjusted for the body composition outcomes at age 10 years to reduce in part the chance of reverse causation, and comparable results were found. The use of self-reported lifestyle behaviors in the current study may result in recall bias and may not be as accurate as objective data provided by accelerometers. However, the validated questionnaires remains an effective method for large-scale population studies and might be better at capturing daily habits as accelerometers are insensitive to many types of physical activity, including bicycling (46). Since this study was not limited to high-risk population, the findings are generalizable to a more general population despite the potential existence of selection bias in our cohort study.

In adolescents, higher levels of physical activity and reduced screen time are associated with lower levels of general and visceral adiposity measures. As part of lifestyle intervention programs, educational and health institutions should consider promoting physical activity and reducing screen time during adolescence as a potential approach to obesity prevention.

A/G ratio, android/gynoid fat mass ratio; BMI, body mass index; DXA, dual-energy X-ray absorptiometry; FMI, fat mass index; LBMI, lean body mass index; MRI, magnetic resonance imaging; V/S ratio, abdominal visceral/subcutaneous fat mass ratio

Acknowledgements

The Generation R study is managed by the Erasmus Medical Centre in close collaboration with the School of Law and the Faculty of Social Sciences at the Erasmus University, Rotterdam, the Municipal Health Service, Rotterdam area, and the Stichting Trombosedienst and Artsenlaboratorium Rijnmond (Star-MDC), Rotterdam. We gratefully acknowledge the contribution of children and their parents, general practitioners, hospitals, midwives and pharmacies in Rotterdam.

Author contributions

TW, JYH, HR, SK and EHGO participated in the design of the study and contributed to the investigation, methodology, formal analysis, writing of original draft, interpretation of results and manuscript editing. VWVJ participated in the design of the study and contributed to the interpretation of results and manuscript editing. MWV and MRA contributed to the interpretation of results and the manuscript editing. All authors have read and approved the final version of the manuscript, and agree with the order of presentation of the authors.

Funding

The Generation R Study is made possible by financial support from the Erasmus Medical Centre, Rotterdam, the Erasmus University Rotterdam and the Netherlands Organization for Health Research and Development. Tong Wu is supported by China Scholarship Council (CSC) PhD Fellowship for his PhD study in Erasmus Medical Center, Rotterdam, the Netherlands. The scholarship file number is 201906260304, CSC URL: http://www.csc.edu.cn/. The researchers are independent from the funders. The study sponsors had no role in the study design, data analysis, interpretation of data, or writing of this report.

Conflict of interest

The authors declare no competing interests.

Fitzmaurice C, Abate D, Abbasi N, Abbastabar H, Abd-Allah F, Abdel-Rahman O, et al. Global, regional, and national cancer incidence, mortality, years of life lost, years lived with disability, and disability-adjusted life-years for 29 cancer groups, 1990 to 2017: a systematic analysis for the global burden of disease study. JAMA oncology. 2019;5(12):1749-68.
Wilks DC, Besson H, Lindroos AK, Ekelund U. Objectively measured physical activity and obesity prevention in children, adolescents and adults: a systematic review of prospective studies. Obesity reviews. 2011;12(5):e119-e29.
Nader PR, O'Brien M, Houts R, Bradley R, Belsky J, Crosnoe R, et al. Identifying risk for obesity in early childhood. Pediatrics. 2006;118(3):e594-e601.
Ramires VV, Dumith SC, Gonçalves H. Longitudinal association between physical activity and body fat during adolescence: A systematic review. Journal of Physical Activity and Health. 2015;12(9):1344-58.
Tripathi M, Mishra SK. Screen time and adiposity among children and adolescents: a systematic review. Journal of Public Health. 2020;28(3):227-44.
Gaggini M, Saponaro C, Gastaldelli A. Not all fats are created equal: adipose vs. ectopic fat, implication in cardiometabolic diseases. Hormone molecular biology and clinical investigation. 2015;22(1):7-18.
Labayen I, Medrano M, Arenaza L, Maíz E, Osés M, Martínez-Vizcaíno V, et al. Effects of exercise in addition to a family-based lifestyle intervention program on hepatic fat in children with overweight. Diabetes Care. 2020;43(2):306-13.
Davis CL, Pollock NK, Waller JL, Allison JD, Dennis BA, Bassali R, et al. Exercise dose and diabetes risk in overweight and obese children: a randomized controlled trial. Jama. 2012;308(11):1103-12.
Kooijman MN, Kruithof CJ, van Duijn CM, Duijts L, Franco OH, van IJzendoorn MH, et al. The Generation R Study: design and cohort update 2017. European journal of epidemiology. 2016;31(12):1243-64.
Chinapaw MJ, Slootmaker SM, Schuit AJ, van Zuidam M, van Mechelen W. Reliability and validity of the Activity Questionnaire for Adults and Adolescents (AQuAA). BMC Med Res Methodol. 2009;9:58.
Currie C, Inchley J, Molcho M, Lenzi M, Veselska Z, Wild F. Health Behaviour in School-aged Children (HBSC) Study Protocol: Background, Methodology and Mandatory items for the 2013/14 Survey. St Andrews: CAHRU. 2014.
Schmitz KH, Harnack L, Fulton JE, Jacobs DR, Jr., Gao S, Lytle LA, et al. Reliability and validity of a brief questionnaire to assess television viewing and computer use by middle school children. J Sch Health. 2004;74(9):370-7.
Booth ML, Okely AD, Chey T, Bauman A. The reliability and validity of the physical activity questions in the WHO health behaviour in schoolchildren (HBSC) survey: a population study. Br J Sports Med. 2001;35(4):263-7.
Estrada S, Lu R, Conjeti S, Orozco‐Ruiz X, Panos‐Willuhn J, Breteler MMB, et al. FatSegNet: A fully automated deep learning pipeline for adipose tissue segmentation on abdominal dixon MRI. Magnetic resonance in medicine. 2020;83(4):1471-83.
Marshall WA, Tanner JM. Variations in pattern of pubertal changes in girls. Archives of disease in childhood. 1969;44(235):291.
Marshall WA, Tanner JM. Variations in the pattern of pubertal changes in boys. Archives of disease in childhood. 1970;45(239):13-23.
Sterne JAC, White IR, Carlin JB, Spratt M, Royston P, Kenward MG, et al. Multiple imputation for missing data in epidemiological and clinical research: potential and pitfalls. Bmj. 2009;338.
Feise RJ. Do multiple outcome measures require p-value adjustment? BMC medical research methodology. 2002;2(1):1-4.
Silveira EA, Kliemann N, Noll M, Sarrafzadegan N, de Oliveira C. Visceral obesity and incident cancer and cardiovascular disease: An integrative review of the epidemiological evidence. Obesity Reviews. 2021;22(1):e13088.
Lebovitz HE, Banerji MA. Point: visceral adiposity is causally related to insulin resistance. Diabetes care. 2005;28(9):2322-5.
Fogelholm M, Kukkonen‐Harjula K. Does physical activity prevent weight gain–a systematic review. Obesity reviews. 2000;1(2):95-111.
Wareham NJ, van Sluijs EMF, Ekelund U. Physical activity and obesity prevention: a review of the current evidence. Proceedings of the Nutrition Society. 2005;64(2):229-47.
Biddle SJH, Bengoechea García E, Pedisic Z, Bennie J, Vergeer I, Wiesner G. Screen time, other sedentary behaviours, and obesity risk in adults: a review of reviews. Current obesity reports. 2017;6(2):134-47.
Biddle SJH, García Bengoechea E, Wiesner G. Sedentary behaviour and adiposity in youth: a systematic review of reviews and analysis of causality. International Journal of Behavioral Nutrition and Physical Activity. 2017;14(1):1-21.
Marshall SJ, Biddle SJH, Gorely T, Cameron N, Murdey I. Relationships between media use, body fatness and physical activity in children and youth: a meta-analysis. International journal of obesity. 2004;28(10):1238-46.
Collings PJ, Westgate K, Väistö J, Wijndaele K, Atkin AJ, Haapala EA, 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 Medicine. 2017;47(4):769-80.
Jiménez-Pavón D, Fernández-Vázquez A, Alexy U, Pedrero R, Cuenca-García M, Polito A, et al. Association of objectively measured physical activity with body components in European adolescents. BMC Public Health. 2013;13(1):1-9.
Joensuu L, Syväoja H, Kallio J, Kulmala J, Kujala UM, Tammelin TH. Objectively measured physical activity, body composition and physical fitness: Cross-sectional associations in 9-to 15-year-old children. European journal of sport science. 2018;18(6):882-92.
Lohman TG, Ring K, Schmitz KH, Treuth MS, Loftin M, Yang S, et al. Associations of body size and composition with physical activity in adolescent girls. Medicine and science in sports and exercise. 2006;38(6):1175.
Ness AR, Leary SD, Mattocks C, Blair SN, Reilly JJ, Wells J, et al. Objectively measured physical activity and fat mass in a large cohort of children. PLoS medicine. 2007;4(3):e97.
Riso E-M, Kull M, Mooses K, Jürimäe J. Physical activity, sedentary time and sleep duration: Associations with body composition in 10–12-year-old Estonian schoolchildren. BMC Public Health. 2018;18(1):1-7.
Medrano M, Cadenas‐Sánchez C, Oses M, Villanueva A, Cabeza R, Idoate F, et al. Associations of fitness and physical activity with specific abdominal fat depots in children with overweight/obesity. Scandinavian journal of medicine & science in sports. 2022;32(1):211-22.
Murabito JM, Pedley A, Massaro JM, Vasan RS, Esliger D, Blease SJ, et al. Moderate‐to‐vigorous physical activity with accelerometry is associated with visceral adipose tissue in adults. Journal of the American Heart Association. 2015;4(3):e001379.
Saelens BE, Grow HM, Stark LJ, Seeley RJ, Roehrig H. Efficacy of increasing physical activity to reduce children's visceral fat: A pilot randomized controlled trial. International Journal of Pediatric Obesity. 2011;6(2):102-12.
Saelens BE, Seeley RJ, Van Schaick K, Donnelly LF, O'Brien KJ. Visceral abdominal fat is correlated with whole-body fat and physical activity among 8-y-old children at risk of obesity. The American journal of clinical nutrition. 2007;85(1):46-53.
Moschonis G, Kaliora AC, Karatzi K, Michaletos A, Lambrinou C-P, Karachaliou AK, et al. Perinatal, sociodemographic and lifestyle correlates of increased total and visceral fat mass levels in schoolchildren in Greece: the Healthy Growth Study. Public health nutrition. 2017;20(4):660-70.
Silva BGCd, Silva ICMd, Ekelund U, Brage S, Ong KK, De Lucia Rolfe E, et al. Associations of physical activity and sedentary time with body composition in Brazilian young adults. Scientific reports. 2019;9(1):1-10.
Matsuzawa Y, Shimomura I, Nakamura T, Keno Y, Kotani K, Tokunaga K. Pathophysiology and pathogenesis of visceral fat obesity. Obesity research. 1995;3(S2):187s-94s.
Parikh T, Stratton G. Influence of intensity of physical activity on adiposity and cardiorespiratory fitness in 5–18 year olds. Sports Medicine. 2011;41(6):477-88.
Tran TT, Yamamoto Y, Gesta S, Kahn CR. Beneficial effects of subcutaneous fat transplantation on metabolism. Cell metabolism. 2008;7(5):410-20.
Pearson N, Biddle SJH. Sedentary behavior and dietary intake in children, adolescents, and adults: a systematic review. American journal of preventive medicine. 2011;41(2):178-88.
Fletcher E, Leech R, McNaughton SA, Dunstan DW, Lacy KE, Salmon J. Is the relationship between sedentary behaviour and cardiometabolic health in adolescents independent of dietary intake? A systematic review. obesity reviews. 2015;16(9):795-805.
Marshall SJ, Merchant G. Advancing the science of sedentary behavior measurement. American Journal of Preventive Medicine. 2013;44(2):190-1.
Chastin SFM, Egerton T, Leask C, Stamatakis E. Meta‐analysis of the relationship between breaks in sedentary behavior and cardiometabolic health. Obesity. 2015;23(9):1800-10.
Shrestha N, Pedisic Z, Neil-Sztramko S, Kukkonen-Harjula KT, Hermans V. The impact of obesity in the workplace: a review of contributing factors, consequences and potential solutions. Current obesity reports. 2016;5(3):344-60.
Trost SG. Objective measurement of physical activity in youth: current issues, future directions. Exercise and sport sciences reviews. 2001;29(1):32-6.

There is NO conflict of interest to disclose

Supplementalinformation.docx

Download PDF

Version 1

posted

You are reading this latest preprint version

Physical Activity, Screen Time and Body Composition in Adolescents: A Population-based Prospective Cohort Study

Status:

Version 1

Abstract

Background

Methods

Results

Conclusion

Figures

Introduction

Methods

Results

Discussion

Conclusion

Abbreviations

Declarations

References

Additional Declarations

Supplementary Files

Status:

Version 1