Summary
Objective
Investigate associations of TV viewing time and accelerometry‐derived sedentary time with inflammatory and endothelial function biomarkers in children.
Methods
Cross‐sectional analysis of 164 7–10‐year‐old children. TV viewing time was assessed by parental proxy report and total and patterns of sedentary time accumulation (e.g. prolonged bouts) were assessed by accelerometry. C‐reactive protein (CRP), homeostasis model assessment of insulin resistance, interleukin‐2, ‐6, ‐8, ‐10, tumour necrosis factor alpha, adiponectin, resistin, brain‐derived neurotrophic factor, soluble intercellular and vascular adhesion molecule 1, plasminogen activator inhibitor 1 and soluble E‐selectin were assessed. Generalised linear models assessed the associations of TV viewing and sedentary time with biomarkers, adjusting for sex, waist circumference, moderate‐ to vigorous‐intensity physical activity and diet density.
Results
Each additional h week−1 of TV viewing was associated with 4.4% (95% CI: 2.1, 6.7) greater CRP and 0.6% (0.2, 1.0) greater sVCAM‐1 in the fully adjusted model. The association between frequency and duration of 5–10 min bouts of sedentary time and CRP was positive after adjustment for sex and waist circumference but attenuated after adjustment for diet density.
Conclusions
This study suggests that TV viewing was unfavourably associated with several markers of inflammation and endothelial dysfunction. The detrimental association between 5 and 10 min bouts of sedentary time and CRP approached significance, suggesting that further research with a stronger study design (longitudinal and/or experimental) is needed to better understand how the accumulation of sedentary time early in life may influence short and longer term health.
Keywords: Cardiometabolic, paediatric, sedentary behaviour
Introduction
The prevalence of elevated cardiometabolic risk factor levels are increasing among youth worldwide 1, 2. While the clinical manifestation of cardiometabolic diseases (e.g. myocardial infarction, stroke) typically appears in adulthood, the pre‐clinical phase of many diseases can persist for decades and certain risk factors (i.e. atherosclerotic lesions) may present during childhood 3. Beyond traditional markers of cardiometabolic risk (e.g. cholesterol, triglycerides and blood pressure), emerging research has shown that non‐traditional biomarkers such as inflammatory and endothelial dysfunction markers are implicated in atherosclerosis and metabolic syndrome in youth, and are elevated in overweight and obese compared with normal weight children and adolescents 4, 5, 6, 7. Consequently, examining inflammatory and endothelial dysfunction markers will provide a unique insight into the early aetiology and development of cardiometabolic diseases.
Physical activity (PA) is critical for the prevention of elevated metabolic and cardiovascular health risk factors levels in children 8, yet accounts for a small proportion of their day 9. In contrast, sedentary time (activities characterised by a low‐energy expenditure [<1.5 METS] in a seated or reclined position 10) makes up more than 60% of a child's waking hours 9. Interventions that aim to reduce sedentary behaviours might be valuable for improving children's cardiometabolic health. TV viewing time has a detrimental effect on traditional cardiometabolic risk factors among children and adolescents, independent of PA levels 11, 12, but little is known about associations with non‐traditional risk factors. One study reported that several biomarkers (soluble intracellular adhesion molecule 1, vascular adhesion molecule 1 and E‐selectin) were inversely associated with adolescents' TV viewing time, but not objectively measured sedentary time 13. However, it is unknown whether similar relationships exist for younger children.
To assist with refining the public health guidelines in many countries including Australia, the USA, Canada and the UK, which currently recommend limiting screen time (e.g. TV viewing, video games) to less than 2 h per day and breaking up extended periods of sitting 14, it is also important to consider whether sustained bouts of sedentary time of a certain duration are detrimental to health, or whether it is the total time accumulated (regardless of bout length) that is most strongly related to health outcomes. The primary aim of this study, therefore, was to investigate the associations of TV viewing time and objectively measured sedentary time with inflammatory and endothelial function biomarkers in primary school‐aged children. A secondary aim was to evaluate the associations of patterns of sedentary time accumulation (i.e. prolonged sedentary bouts and breaks in sedentary time) with inflammatory and endothelial function biomarkers.
Methods
Participants
Participants were children aged 7–10 years involved in the Transform‐Us! Study, a 2‐year cluster‐randomised controlled trial, which investigated the impact of a school and home‐based intervention on children's sedentary behaviour and PA 15. Grade 3 children were recruited from 20 schools (primarily in low‐ and middle‐income areas) within a 50‐km radius of the Melbourne Central Business District 15. Baseline data were collected from the children and their parents between February and June 2010. Written informed parental consent was obtained from 599 parents for their child to participate in one or more of the Transform‐Us! assessments. A subsample consented to blood biomarker assessments (n = 351), with 164 children (87 girls) providing complete blood sample and valid accelerometry data for analyses. The study was approved by the Deakin University Human Research Ethics Committee (EC 141‐2009), the Victorian Department of Education and Early Childhood Development (2009_000344) and the Catholic Education Melbourne Office (Project Number 1545).
Exposures
TV viewing
Parents proxy‐reported the number of hours per day their child spends watching TV/videos/DVDs during the week and on the weekend using the validated CLASS questionnaire 16. The weekday and weekend responses were summed to provide a weekly estimate of TV viewing time (h week−1).
Sedentary time
Sedentary time was objectively assessed every 15 s for 8 consecutive days using hip‐mounted ActiGraph GT3X accelerometers (Pensacola, FL, USA). A recent study recommended the use of 60‐min of consecutive zeros to define accelerometer non‐wear 17. However, our group (unpublished data) found that when ActiGraph data were compared with activPAL data (a direct measure of sitting time), the 20‐min definition resulted in an almost identical wear time and a smaller difference between the sedentary time and sitting time estimates compared with the 60‐min definition. Consequently, non‐wear time was defined as ≥20 min of consecutive zero counts in this study. Sedentary time was defined as ≤100 counts per minute 18. The frequency and duration of time spent in sedentary bouts of medium (5–10 min) and longer (>10 min) duration were also calculated, as others have demonstrated a relationship between sedentary time spent in bouts of these durations and weight status in youth 19. Breaks in sedentary time were defined as the number of times that the accelerometer exceeded 25 counts per 15‐s epoch following a 15‐s epoch of <25 counts per epoch 20. Participants were required to have ≥8 h and ≥7 h of wear time on weekdays and weekend days, respectively, on at least 3 days to be included in the analyses. The lower weekend wear time requirement is due to children typically waking later on weekends. In this study, data from 82.6% of participants (400/484) were analysed. Total sedentary time and prolonged bouts in sedentary time were adjusted for wear time using the residuals obtained by regressing the sedentary variables on wear time. Breaks in sedentary time were additionally adjusted for total sedentary time using the same approach.
Outcome measures
Cardiometabolic biomarkers
A fasted morning blood sample was collected at a commercial pathology laboratory (Melbourne Pathology Clinic; n = 206). Plasma insulin, glucose and C‐reactive protein (CRP) were assessed at Melbourne Pathology. Insulin sensitivity was derived using the homeostasis model assessment of insulin resistance (HOMA‐IR) calculated by fasting insulin multiplied by fasting glucose divided by 22.5 21. Milliplex immunoassay kits (Millipore Corp., Billerica, MD, USA) were used to simultaneously measure plasma levels of cytokines (interleukin‐2 [IL‐2], ‐6 [IL‐6], ‐8 [IL‐8], ‐10 [IL‐10] and tumour necrosis factor alpha [TNF‐α]), adipokines (adiponectin and resistin), neurodegenerative factors (brain‐derived neurotrophic factor [BDNF], soluble intercellular adhesion molecule 1 [sICAM‐1], plasminogen activator inhibitor 1 [PAI‐1], soluble vascular cellular adhesion molecule 1 [sVCAM‐1] and cardiovascular disease factor, soluble E‐selectin (sE‐selectin). The assay was performed according to the manufacturer's instructions and all samples were run in duplicate. Mean inter‐assay coefficient of variation for the 12 factors analysed ranged from 2.4% to 7.6%. Intra‐assay coefficient of variation was determined by replicate analysis (n = 9) of the two provided assay quality controls and ranged from 7.3% to 13.5%. Samples that returned a value below the detectable limits of the assay were allocated the minimum detectable concentration of each assay.
Covariates
Anthropometry
Height and weight were measured using standardised protocols with a portable stadiometer (SECA 220, Los Angeles, California, USA) and digital scales (Wederburn Tanita, Melbourne, Victoria, Australia) to the nearest 0.1 cm and 0.1 kg, respectively. Waist circumference (cm) was measured using a flexible steel tape at the narrowest point between the bottom rib and the iliac crest, in the midaxillary plane. Where a discrepancy of over 1 cm or 1 kg was observed, a third measure was taken. For the above measures, the average was used in the analyses. Body mass index (BMI; kg m−2) categorised participants as healthy weight, overweight or obese based on the age‐ and sex‐specific International Obesity Task Force classification 22.
Moderate‐ to vigorous‐intensity physical activity (MVPA) and diet density
MVPA was derived from accelerometers using the age‐appropriate equation for moderate (4.0–5.9 METs) and vigorous (≥6.0 METs) intensity PA 23. MVPA was standardised for accelerometer wear time using the residuals approach. Children's consumption frequency of key energy‐dense foods and beverages was collected via a parental proxy report using items identified as important contributors to children's energy and fat 24. Eight energy‐dense foods and drinks were included: salty snacks, chocolate and sweets, cakes, pastries, fast food, chips, fruit juice and soft drinks. Responses for food items ranged on a monthly scale from ‘Never or less than once/month’ to ‘6 or more times per day’, while beverage items ranged on a daily scale from ‘My child does not drink this beverage’ to ‘6 or more serves/day’. Responses were summed to provide a total diet density score (range: 9–42).
Statistical analysis
All analyses were performed using Stata v.12 (StataCorp, College Station, TX, USA). Associations of TV viewing time and total objectively measured sedentary time with inflammatory and endothelial dysfunction biomarkers were assessed using generalised linear models with robust standard errors accounting for clustering at the class level. Gamma variance and logarithmic link functions were selected because of the positive skew of outcomes and residuals and the continuous, positive nature of the data. Initial models were adjusted for clustering within school classes. Model 2 additionally adjusted for the child's sex and waist circumference, model 3 for accelerometry‐derived MVPA and model 4 for diet density. The deviance residuals of models were examined to assess the goodness of fit. Generalised linear models were also used to examine associations of patterns of sedentary time accumulation (i.e. prolonged bouts of sedentary time and breaks in sedentary time) with inflammatory and endothelial dysfunction biomarkers. Separate models were fitted for each sedentary time pattern exposure. Four models were estimated as described above. Waist circumference was chosen as a measure of adiposity instead of BMI as a covariate in all models because of stronger correlations with biomarkers in this sample. Interactions of sex and adiposity with sedentary time exposure variables were not statistically significant so data for boys and girls were pooled. Statistical significance was set at P < 0.05.
Results
Participant characteristics are presented in Table 1. On average, participants spent 56% of their day in sedentary time, with 20% of sedentary time accumulated in bouts of 5–10 min in duration, 16% accumulated in greater than 10 min in duration and the remainder (64%) accumulated in bouts of less than 5 min.
Table 1.
Participant characteristics (n = 164)
| Variables | Mean (SD)* |
|---|---|
| Age (years) | 8.7 (0.4) |
| Height (cm) | 132.9 (6.3) |
| Weight (kg) | 30.9 (6.3) |
| BMI (kg m−2) | 17.4 (2.5) |
| Overweight + obese (%) | 22.2 |
| Waist circumference (cm) | 59.6 (6.3) |
| Questionnaire‐derived variables | |
| Diet density score | 2.3 (4.9) |
| TV viewing (h week−1) | 12.1 (6.4) |
| Accelerometer‐derived variables | |
| Total wear time (min day−1) | 717.3 (69.2) |
| Sedentary time (min day−1) | 401.6 (57.4) |
| Frequency of sedentary bouts of 5–10 min (no. per day) | 11.6 (3.1) |
| Total duration of sedentary bouts of 5–10 min (min day−1) | 78.7 (21.5) |
| Frequency of sedentary bouts of >10 min (no. per day) | 4.3 (1.9) |
| Total duration of sedentary bouts of >10 min (min day−1) | 62.5 (29.5) |
| Breaks in sedentary time (no. per day) | 315.3 (38.9) |
| MVPA (min day−1) | 85.5 (24.8) |
| Biomarkers | |
| CRP (mg L−1) | 0.3 (0.1, 0.6) |
| HOMA‐IR | 1.2 (0.8, 1.6) |
| BDNF (ng mL−1) | 20.9 (15.6, 25.6) |
| Adiponectin (mg mL−1) | 25.7 (18.2, 35.3) |
| Resistin (ng mL−1) | 20.2 (16.1, 24.8) |
| sE‐selectin (ng mL−1) | 60.7 (47.1, 83.5) |
| sICAM‐1 (ng mL−1) | 141.8 (121.9, 169.3) |
| sVCAM‐1 (ng mL−1) | 1479.2 (1328.8, 1659.1) |
| PAI‐1 (ng mL−1) | 237.1 (184.4, 291.8) |
| TNF‐α (pg mL−1) | 9.8 (7.3, 12.2) |
| IL‐2 (pg mL−1) | 1.1 (0.3, 3.0) |
| IL‐6 (pg mL−1) | 0.4 (0.2, 1.3) |
| IL‐8 (pg mL−1) | 5.6 (4.4, 7.7) |
| IL‐10 (pg mL−1) | 13.8 (8.9, 25.3) |
*Mean (SD, standard deviation) for demographic and sedentary variables, median (IQR, interquartile range) for biomarkers.
BDNF, brain‐derived neurotrophic factor; BMI, body mass index; CRP, C‐reactive protein; HOMA‐IR, homeostasis model assessment of insulin resistance; IL, interleukin; MVPA, moderate‐to vigorous‐intensity physical activity; PAI, plasminogen inhibitor; sE‐selectin, soluble E‐selectin; sICAM, soluble intracellular adhesion molecule; sVCAM, soluble vascular adhesion molecule; TNF‐α, tumour necrosis factor.
Table 2 presents the associations of TV viewing and sedentary time with cardiometabolic biomarkers. Several detrimental associations were observed between biomarkers and TV viewing, which persisted after adjustment of accelerometry‐derived MVPA and diet density (Supporting Information Fig. S1). Each additional h week−1 of TV viewing was associated with 4.4% (95% CI: 2.1, 6.7) greater CRP and 0.6% (0.2, 1.0) greater sVCAM‐1. Significant associations were found between objectively measured sedentary time and sE‐selectin and IL‐6 in Model 1, although these were attenuated after adjustment for waist circumference and sex. No other associations between total sedentary time and the remaining biomarkers were observed (P > 0.05).
Table 2.
Associations of total sedentary time and TV viewing with cardiometabolic biomarkers in school‐aged children
| Model | Sedentary time (h day−1) | TV viewing (h week−1) | |
|---|---|---|---|
| b (95% CI) | b (95% CI) | ||
| CRP (mg L−1) | 1 | 1.139 (0.859, 1.509) | 1.043 (1.017, 1.070) a |
| 2 | 1.180 (0.926, 1.504) | 1.032 (1.010, 1.054) b | |
| 3 | 1.163 (0.850, 1.592) | 1.034 (1.013, 1.056) c | |
| 4 | 1.083 (0.782, 1.502) | 1.044 (1.021, 1.067) a | |
| HOMA‐IR | 1 | 1.076 (0.896, 1.291) | 1.009 (0.996, 1.023) |
| 2 | 1.016 (0.881, 1.170) | 1.003 (0.989, 1.018) | |
| 3 | 0.959 (0.806, 1.142) | 1.004 (0.989, 1.018) | |
| 4 | 0.937 (0.794, 1.107) | 1.002 (0.987, 1.017) | |
| BDNF (ng mL−1) | 1 | 0.998 (0.919, 1.084) | 1.005 (0.996, 1.014) |
| 2 | 0.999 (0.926, 1.079) | 1.003 (0.994, 1.011) | |
| 3 | 0.970 (0.878, 1.072) | 1.003 (0.994, 1.011) | |
| 4 | 0.972 (0.875, 1.080) | 1.003 (0.995, 1.012) | |
| Adiponectin (mg mL−1) | 1 | 0.960 (0.853, 1.082) | 1.004 (0.992, 1.016) |
| 2 | 0.959 (0.854, 1.076) | 1.006 (0.994, 1.019) | |
| 3 | 0.922 (0.801, 1.062) | 1.007 (0.994, 1.019) | |
| 4 | 0.930 (0.811, 1.067) | 1.005 (0.993, 1.018) | |
| Resistin (ng mL−1) | 1 | 1.024 (0.935, 1.122) | 1.007 (0.998, 1.017) |
| 2 | 1.016 (0.926, 1.115) | 1.007 (0.997, 1.016) | |
| 3 | 0.993 (0.850, 1.159) | 1.007 (0.997, 1.017) | |
| 4 | 0.997 (0.862, 1.153) | 1.007 (0.997, 1.018) | |
| sE‐selectin (ng mL−1) | 1 | 1.105 (1.004, 1.216) d | 1.003 (0.991, 1.014) |
| 2 | 1.094 (0.997, 1.201) | 0.998 (0.988, 1.008) | |
| 3 | 1.051 (0.935, 1.182) | 0.999 (0.989, 1.009) | |
| 4 | 1.054 (0.923, 1.204) | 0.998 (0.988, 1.008) | |
| sICAM‐1 (ng mL−1) | 1 | 1.007 (0.931, 1.090) | 1.002 (0.997, 1.009) |
| 2 | 1.000 (0.924, 1.081) | 1.001 (0.995, 1.007) | |
| 3 | 0.989 (0.891, 1.096) | 1.001 (0.995, 1.008) | |
| 4 | 0.987 (0.889, 1.095) | 1.001 (0.994, 1.008) | |
| sVCAM‐1 (ng mL−1) | 1 | 0.991 (0.955, 1.028) | 1.005 (1.001, 1.009) e |
| 2 | 0.995 (0.959, 1.032) | 1.007 (1.003, 1.011) c | |
| 3 | 0.995 (0.938, 1.054) | 1.007 (1.003, 1.011) c | |
| 4 | 0.999 (0.940, 1.062) | 1.006 (1.002, 1.010) f | |
| PAI‐1 (ng mL−1) | 1 | 1.084 (0.982, 1.120) | 1.004 (0.995, 1.013) |
| 2 | 1.084 (0.988, 1.191) | 1.003 (0.994, 1.011) | |
| 3 | 0.987 (0.880, 1.106) | 1.004 (0.995, 1.012) | |
| 4 | 0.974 (0.866, 1.095) | 1.005 (0.996, 1.014) | |
| TNF‐α (pg mL−1) | 1 | 1.063 (0.948, 1.192) | 1.008 (0.998, 1.018) |
| 2 | 1.063 (0.948, 1.192) | 1.007 (0.997, 1.018) | |
| 3 | 1.129 (0.936, 1.362) | 1.006 (0.996, 1.017) | |
| 4 | 1.146 (0.941, 1.394) | 1.007 (0.996, 1.017) | |
| IL‐2 (pg mL−1) | 1 | 0.962 (0.680, 1.362) | 1.041 (0.989, 1.096) |
| 2 | 0.998 (0.701, 1.422) | 1.047 (0.998, 1.098) | |
| 3 | 0.744 (0.494, 1.121) | 1.046 (1.000, 1.095) g | |
| 4 | 0.937 (0.633, 1.386) | 1.025 (0.988, 1.063) | |
| IL‐6 (pg mL−1) | 1 | 1.772 (1.112, 2.826) h | 1.024 (0.979, 1.071) |
| 2 | 1.268 (0.899, 1.788) | 1.039 (0.998, 1.082) | |
| 3 | 0.894 (0.571, 1.400) | 1.030 (0.996, 1.066) | |
| 4 | 0.778 (0.476, 1.271) | 1.015 (0.980, 1.052) | |
| IL‐8 (pg mL−1) | 1 | 0.660 (0.349, 1.251) | 1.027 (0.953, 1.108) |
| 2 | 0.826 (0.549, 1.242) | 1.026 (0.977, 1.078) | |
| 3 | 0.994 (0.613, 1.612) | 1.023 (0.980, 1.068) | |
| 4 | 0.998 (0.606, 1.641) | 1.021 (0.974, 1.071) | |
| IL‐10 (pg mL−1) | 1 | 1.151 (0.884, 1.498) | 1.025 (0.992, 1.059) |
| 2 | 1.124 (0.869, 1.454) | 1.029 (0.997, 1.062) | |
| 3 | 0.841 (0.589, 1.201) | 1.028 (0.998, 1.059) | |
| 4 | 0.772 (0.552, 1.080) | 1.022 (0.991, 1.055) |
b (95% CI), exponentiated regression coefficients and 95% confidence intervals. Model 1 adjusted for clustering within school classes. Model 2 further adjusted for sex and waist circumference. Model 3 further adjusted for moderate‐to vigorous‐intensity physical activity (MVPA). Model 4 further adjusted for diet density. a P < 0.001, b P = 0.019, c P = 0.001, d P = 0.041, e P = 0.009, f P = 0.005, g P = 0.05, h P = 0.016.
BDNF, brain‐derived neurotrophic factor; CRP, C‐reactive protein; HOMA‐IR, homeostasis model assessment of insulin resistance; IL, interleukin; PAI, plasminogen inhibitor; sE‐selectin, soluble E‐selectin; sICAM, soluble intracellular adhesion molecule; sVCAM, soluble vascular adhesion molecule; TNF‐α, tumour necrosis factor.
Relationships between sedentary time accumulation and CRP are presented in Table 3. There was a positive association between the frequency of time spent in sedentary bouts of 5–10 min in length and CRP, independent of sex, waist circumference and MVPA (model 3). Each additional 5–10 min bout of sedentary time was associated with 5.4% (95% CI: 0.4, 10.7) greater CRP. However, this association was attenuated after adjustment for diet density (P = 0.056). The duration of time spent in sedentary bouts of 5–10 min in length was positively associated with CRP, independent of sex and waist circumference (model 2). Each additional minute spent in a sedentary bout length of 5–10 min was associated with 0.7% (0.1, 1.4) greater CRP. This association attenuated after adjustment for MVPA (P = 0.063) and diet density (P = 0.084). No significant associations of the frequency and duration of time spent in sedentary bouts of ≥10 min or frequency of breaks with biomarkers were observed independent of confounders (Table 3 and Supporting Information Table S1).
Table 3.
Associations of sedentary time patterns with C‐reactive protein (CRP) in school‐aged children
| Model | Frequency of sedentary bouts of 5–10 min (no. per day) | Duration of sedentary bouts of 5–10 min (min day−1) | Frequency of sedentary bouts >10 min (no. per day) | Duration of sedentary bouts of >10 min (min day−1) | Frequency of breaks in sedentary time (no. per day) | |
|---|---|---|---|---|---|---|
| b (95% CI) | b (95% CI) | b (95% CI) | b (95% CI) | b (95% CI) | ||
| 1 | 1.046 (0.999, 1.095) | 1.006 (0.999, 1.013) | 1.034 (0.923, 1.158) | 1.002 (0.994, 1.009) | 1.003 (0.995, 1.012) | |
| CRP (mg L−1) | 2 | 1.055 (1.010, 1.101) a | 1.007 (1.001, 1.014) c | 1.085 (0.985, 1.196) | 1.005 (0.999, 1.011) | 0.996 (0.990, 1.003) |
| 3 | 1.054 (1.004, 1.107) b | 1.007 (1.000, 1.014) | 1.074 (0.963, 1.198) | 1.004 (0.998, 1.011) | 0.993 (0.985, 1.000) | |
| 4 | 1.050 (0.999, 1.104) | 1.006 (0.999, 1.014) | 1.034 (0.927, 1.154) | 1.002 (0.995, 1.009) | 0.994 (0.986, 1.001) |
b (95% CI), exponentiated regression coefficients and 95% confidence intervals. Model 1 adjusted for clustering within school classes. Model 2 further adjusted for sex and waist circumference. Model 3 further adjusted for moderate‐to vigorous‐intensity physical activity (MVPA). Model 4 further adjusted for diet density. a P = 0.015, b P = 0.033, c P = 0.024.
Discussion
The key finding from this study was that TV viewing time, but not objectively measured sedentary time, was detrimentally associated with several biomarkers, including those related to inflammation (CRP) and endothelial function (sVCAM‐1), independent of MVPA and diet density. Likewise, sedentary time bouts lasting 5–10 min were also associated with greater levels of CRP, although this association was attenuated after adjusting for MVPA and diet density.
The positive association between TV viewing and CRP (independent of MVPA) are in agreement with a large US nationwide study in 10–16‐year‐olds, which reported that the odds of having CRP in the highest quartile increased in a dose‐response manner with TV viewing, independent of PA and other covariates 11. This is an important finding as CRP is an acute‐phase inflammatory marker shown to predict coronary events in adults, independent of conventional risk factors 25, and has been associated with atherosclerosis, endothelial dysfunction, and metabolic syndrome in youth 4, 5, 6. To date, the potential physiological mechanisms that may be involved in the association between CRP and sedentary behaviour are currently poorly understood, and further research is needed. However, CRP has been found to be a strong predictor of several diseases, such as cardiovascular diseases, independent of more traditional markers and other inflammatory markers 4, 5, 25. It is therefore possible that CRP has a higher predictive value for poor health outcomes than the other inflammatory markers assessed in this study. This may partly explain the significant findings observed.
The significant association between TV viewing and soluble cellular adhesion molecules also supports previous adolescent research 13. Cellular adhesion molecules play a role in promoting the adhesion of inflammatory cells on the vascular surface, thereby aiding the development of atherosclerosis 5. Although our findings are modest (4.4% and 0.6% greater CRP and sVCAM‐1, respectively, per h week−1 of TV viewing), the tracking of low‐grade inflammation from childhood through adulthood and associations of markers of inflammation and endothelial function with atherosclerotic lesions 5 suggest that elevated biomarkers early in life may be indicative of cardiometabolic risk later in life. Interestingly, we did not observe associations of TV viewing with other markers of inflammation or endothelial dysfunction. Although we measured several upstream biomarkers of CRP that play a role in systemic inflammation, such as TNF‐α and IL‐6, their role in the development of cardiometabolic risk is poorly understood 5. Likewise, while other biomarkers measured in the current study have been implicated in the development and progression of cardiometabolic conditions (e.g. PAI‐1, adiponectin, HOMA‐IR), further work is necessary to identify the clinical relevance of these markers during childhood 4, 5.
We failed to observe any association between total sedentary time and inflammatory or endothelial function biomarkers. Similarly, others have failed to observe an association between traditional cardiovascular risk markers (e.g. triglycerides, blood pressure, weight status, vascular function) and total sedentary time in youth after accounting for PA 11, 26, 27. The lack of associations may be due to relatively low levels of sedentariness during childhood compared with adulthood; thus, deleterious associations with cardiometabolic health in adulthood are likely to reflect increasing lifelong exposure to sedentary behaviour 13. More detrimental associations may have been observed had our population been more sedentary or we had investigated a population at higher risk of cardiovascular disease.
Relatively, low sedentariness or high MVPA engagement in childhood compared with adulthood cannot explain the stronger adverse relationships seen between health outcomes and subjective reports of children's TV viewing and objectively measured sedentary time. Unfavourable observations between TV viewing and cardiometabolic risk 11, 12, but not total sedentary time 11, 27, 28, may indicate that certain behaviours associated with TV viewing, such as caloric consumption, and not the actual act of being sedentary itself, may explain the negative associations observed 29. TV viewing was positively associated with diet density in our sample (r = 0.23, P < 0.001), yet significant detrimental associations between TV viewing and inflammation and endothelial dysfunction persisted even after controlling for diet density. Although we used a simple caloric intake measure, this suggests that the relationship between TV viewing and health is not entirely mediated by dietary intake. Future studies using a more sophisticated measure of dietary intake are warranted. Further, the potential mechanisms driving an association between cardiometabolic health and TV viewing, but not total sedentary time, remains unclear. Studies specifically investigating objectively measured sitting time would greatly add to this evidence base, as accelerometers cannot differentiate between sitting and standing and TV viewing may not just reflect time spent sitting as viewers may be standing or moving while watching.
Little research has examined associations between sedentary time accumulation and children's cardiometabolic health. Experimental research in adults has shown a detrimental effect of prolonged sedentary bouts on metabolic health 30; however, similar results have yet to be demonstrated in healthy children 29. While there is currently no consensus on how to define a ‘prolonged’ sedentary bout in children, spending more time and having a greater frequency of 5–10 min sedentary bouts was associated with higher levels of CRP. Sedentary bouts of this duration have been detrimentally associated with BMI z‐score in children 19. We found no associations with cardiometabolic biomarkers and longer bouts of sedentary time (>10 min), although few children engaged in sedentary bouts of this duration (median of 4 bouts per day of >10 min of sedentary time vs. 11 bouts per day of 5–10 min of sedentary time). Although the precise biological mechanisms remain to be determined, evidence from animal models suggest that the decreased muscle contractile activity associated with prolonged sedentariness may disrupt the body's cardiometabolic regulatory processes, while breaking up prolonged periods of sedentariness increases muscle contractile activity 31. Further study is necessary to determine whether such behaviours in childhood are related to adverse health outcomes in adulthood.
Our study had several strengths, including the use of an objective measure of sedentary time and the novel investigation of sedentary patterns with respect to multiple cardiometabolic biomarkers. Despite the use of a simple measure of dietary intake, we were able to account for high‐density dietary items in our analyses. Limitations of our study include the cross‐sectional nature of the study design, which precludes us from making judgments regarding causality. No objective functional measurements of the vasculature were performed, thus structural changes that are markers of early microvascular dysfunction were not examined. Future research should consider utilising such objective assessments. Sedentary time estimates may have included both standing time and some light‐intensity PA as hip‐mounted accelerometers cannot determine posture 18. Future research would benefit from activity monitoring devices that can also assess posture (i.e. inclinometers). The commonly used non‐wear definition of 20 min may have also resulted in reduced sedentary time estimates, which could have influenced the associations observed. Finally, our sample included relatively healthy children and only one‐fifth were overweight or obese. Whether these findings persist in a higher risk cohort remains to be evaluated.
Conclusions
TV viewing was unfavourably associated with several markers of inflammation and endothelial dysfunction independent of PA and energy‐dense dietary intake. This study is the first to suggest a detrimental association between prolonged bouts of sedentary time and markers of inflammation in healthy children, but whether this would be related to adverse long‐term health outcomes into adulthood requires further study. Nevertheless, given recent public health recommendations for children to break up prolonged periods of sitting, public health policies would benefit from further work directed at understanding how the accumulation of sedentary time may influence health outcomes.
Conflict of Interest Statement
The authors declared no conflicts of interest.
Author contributions
LG had full access to all of the data in the study and takes responsibility for the integrity of the data and the accuracy of the data analysis. Study concept and design: JS. Acquisition, analysis and interpretation of data: all authors. Drafting of the manuscript: LG. Clinical revision of the manuscript for intellectual content: all authors. Statistical analysis: LG, EC. Obtained funding: JS. Administrative, technical or material support: NDR, LA, PD. Study supervision: JS.
Supporting information
Figure S1. Partial regression plots of TV viewing time and cardiometabolic biomarkers, accounting for sex, waist circumference, accelerometry‐derived moderate‐to vigorous‐intensity physical activity (MVPA), diet density and clustering by school class. (A) C‐reactive protein (C RP, mg L−1) and (B) Soluble vascular adhesion molecule‐1 (sVCAM‐1, ng mL−1). a P = 0.02, b P = 0.03; b (95% CI), exponentiated regression coefficients and 95% confidence intervals.
Table S1. Associations of sedentary time patterns with cardiometabolic biomarkers in school aged children. a P = 0.02, b P = 0.03; b (95% CI), exponentiated regression coefficients and 95% confidence intervals. Model 1 adjusted for clustering within school classes. Model 2 further adjusted for sex and waist circumference. Model 3 further adjusted for moderate‐to vigorous‐intensity physical activity (MVPA). Model 4 further adjusted for diet density.
Acknowledgements
This study is supported by a National Health and Medical Research Council (NHMRC) Project Grant (2009–2013; ID533815) and a Diabetes Australia Research Trust (DART) grant. The funding bodies played no role in the study design; data collection, analysis and interpretation; in the writing of the manuscript; or in the decision to submit the manuscript for publication. LG is supported by an Australian Endeavour Research Fellowship and a Canadian Institutes of Health Research Michael Smith Foreign Study Supplement. NDR is supported by an Australian Research Council Discovery Early Career Research Award (DE120101173). LA is supported by a National Health and Medical Research Council Postgraduate Scholarship (APP1074484). SR is supported by a National Heart Foundation Post Graduate Scholarship. DD is supported by an Australian Research Council Future Fellowship. JS is supported by a National Health and Medical Research Council Principal Fellowship (APP1026216).
Gabel, L. , Ridgers, N. D. , Della Gatta, P. A. , Arundell, L. , Cerin, E. , Robinson, S. , Daly, R. M. , Dunstan, D. W. , and Salmon, J. (2016) Associations of sedentary time patterns and TV viewing time with inflammatory and endothelial function biomarkers in children. Pediatric Obesity, 11: 194–201. doi: 10.1111/ijpo.12045.
References
- 1. de Ferranti SD, Gauvreau K, Ludwig DS, Neufeld EJ, Newburger JW, Rifai N. Prevalence of the metabolic syndrome in American adolescents: findings from the Third National Health and Nutrition Examination Survey. Circulation 2004; 110: 2494–2497. [DOI] [PubMed] [Google Scholar]
- 2. McMahon SK, Haynes A, Ratnam N, et al Increase in type 2 diabetes in children and adolescents in Western Australia. Med J Aust 2004; 180: 459–461. [DOI] [PubMed] [Google Scholar]
- 3. Berenson GS, Wattigney WA, Tracy RE, et al Atherosclerosis of the aorta and coronary arteries and cardiovascular risk factors in persons aged 6 to 30 years and studied at necropsy (The Bogalusa Heart Study). Am J Phys Anthropol 1992; 70: 851–858. [DOI] [PubMed] [Google Scholar]
- 4. Balagopal P, de Ferranti SD, Cook S, et al Nontraditional risk factors and biomarkers for cardiovascular disease: mechanistic, research, and clinical considerations for youth: a scientific statement from the American Heart Association. Circulation 2011; 123: 2749–2769. [DOI] [PubMed] [Google Scholar]
- 5. Canas JA, Sweeten S, Balagopal PB. Biomarkers for cardiovascular risk in children. Curr Opin Cardiol 2013; 28: 103–114. [DOI] [PubMed] [Google Scholar]
- 6. Lambert M, Delvin EE, Paradis G, O'Loughlin J, Hanley JA, Levy E. C‐reactive protein and features of the metabolic syndrome in a population‐based sample of children and adolescents. Clin Chem 2004; 50: 1762–1768. [DOI] [PubMed] [Google Scholar]
- 7. Mendelson M, Michallet A‐S, Monneret D, et al Impact of exercise training without caloric restriction on inflammation, insulin resistance and visceral fat mass in obese adolescents. Pediatr Obes 2014; doi: 10.1111/ijpo.255. [DOI] [PubMed] [Google Scholar]
- 8. McMurray RG, Ondrak KS. Cardiometabolic risk factors in children: the importance of physical activity. Am J Lifestyle Med 2013; 7: 292–303. [Google Scholar]
- 9. Colley RC, Garriguet D, Janssen I, Craig CL, Clarke J, Tremblay MS. Physical activity of Canadian children and youth: accelerometer results from the 2007 to 2009 Canadian Health Measures Survey. Health Rep 2011; 22: 15–23. [PubMed] [Google Scholar]
- 10. Sedentary Behaviour Research Network . Letter to the editor: standardized use of the terms ‘sedentary’ and ‘sedentary behaviours’. Appl Physiol Nutr Metab 2012; 37: 540–542. [DOI] [PubMed] [Google Scholar]
- 11. Carson V, Janssen I. Volume, patterns, and types of sedentary behavior and cardio‐metabolic health in children and adolescents: a cross‐sectional study. BMC Public Health 2011; 11: 274. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 12. Ekelund U, Brage S, Froberg K, et al TV viewing and physical activity are independently associated with metabolic risk in children: the European Youth Heart Study. PLoS Med 2006; 3: e488. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 13. Martinez‐Gomez D, Eisenmann JC, Healy GN, et al Sedentary behaviors and emerging cardiometabolic biomarkers in adolescents. J Pediatr 2012; 160: 104–110, e2. [DOI] [PubMed] [Google Scholar]
- 14. Okely AD, Salmon J, Vella SA, et al A Systematic Review to Inform the Australian Sedentary Behaviour Guidelines for Children and Young People. Report prepared for the Australian Government Department of Health. 2012.
- 15. Salmon J, Arundell L, Hume C, et al A cluster‐randomized controlled trial to reduce sedentary behavior and promote physical activity and health of 8–9 year olds: the Transform‐Us! Study. BMC Public Health 2011; 11: 759. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 16. Salmon J, Timperio A, Telford A, Carver A, Crawford D. Association of family environment with children's television viewing and with low level of physical activity. Obes Res 2005; 13: 1939–1951. [DOI] [PubMed] [Google Scholar]
- 17. Chinapaw MJ, de Niet M, Verloigne M, De Bourdeaudhuij I, Brug J, Altenburg TM. From sedentary time to sedentary patterns: accelerometer data reduction decisions in youth. PLoS ONE 2014; 9: e111205. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 18. Ridgers ND, Salmon J, Ridley K, O'Connell E, Arundell L, Timperio A. Agreement between activPAL and ActiGraph for assessing children's sedentary time. Int J Behav Nutr Phys Act 2012; 9: 15. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 19. Carson V, Stone M, Faulkner G. Patterns of sedentary behavior and weight status among children. Pediatr Exerc Sci 2014; 26: 95–102. [DOI] [PubMed] [Google Scholar]
- 20. Healy GN, Dunstan DW, Salmon J, et al Breaks in sedentary time: beneficial associations with metabolic risk. Diabetes Care 2008; 31: 661–666. [DOI] [PubMed] [Google Scholar]
- 21. Conwell LS, Trost SG, Brown WJ, Batch JA. Indexes of insulin resistance and secretion in obese children and adolescents: a validation study. Diabetes Care 2004; 27: 314–319. [DOI] [PubMed] [Google Scholar]
- 22. Cole TJ, Bellizzi MC, Flegal KM, Dietz WH. Establishing a standard definition for child overweight and obesity worldwide: international survey. BMJ 2000; 320: 1240–1243. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 23. Freedson P, Pober D, Janz KF. Calibration of accelerometer output for children. Med Sci Sports Exerc 2005; 37(11 Suppl.): S523–S530. [DOI] [PubMed] [Google Scholar]
- 24. Australian Bureau of Statistics . Australian National Nutrition Survey (1995). Australian Government: Canberra, 1997, pp. 1–170. [Google Scholar]
- 25. Hirschfield GM, Pepys MB. C‐reactive protein and cardiovascular disease: new insights from an old molecule. QJM 2003; 96: 793–807. [DOI] [PubMed] [Google Scholar]
- 26. De Bourdeaudhuij I, Verloigne M, Maes L, et al Associations of physical activity and sedentary time with weight and weight status among 10‐ to 12‐year‐old boys and girls in Europe: a cluster analysis within the ENERGY Project. Pediatr Obes 2013; 8: 367–375. [DOI] [PubMed] [Google Scholar]
- 27. Chaput J‐P, Lambert M, Mathieu M‐E, Tremblay MS, O'Loughlin JO, Tremblay A. Physical activity vs. sedentary time: independent associations with adiposity in children. Pediatr Obes 2012; 7: 251–258. [DOI] [PubMed] [Google Scholar]
- 28. Nettlefold L, McKay HA, Naylor P‐J, Bredin SSD, Warburton DER. The relationship between objectively measured physical activity, sedentary time, and vascular health in children. Am J Hypertens 2012; 25: 914–919. [DOI] [PubMed] [Google Scholar]
- 29. Saunders TJ, Chaput J‐P, Goldfield GS, et al Prolonged sitting and markers of cardiometabolic disease risk in children and youth: a randomized crossover study. Metabolism 2013; 62: 1423–1428. [DOI] [PubMed] [Google Scholar]
- 30. Dunstan DW, Kingwell BA, Larsen R, et al Breaking up prolonged sitting reduces postprandial glucose and insulin responses. Diabetes Care 2012; 35: 976–983. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 31. Hamilton MT, Hamilton DG, Zderic TW. Role of low energy expenditure and sitting in obesity, metabolic syndrome, type 2 diabetes, and cardiovascular disease. Diabetes 2007; 56: 2655–2667. [DOI] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Supplementary Materials
Figure S1. Partial regression plots of TV viewing time and cardiometabolic biomarkers, accounting for sex, waist circumference, accelerometry‐derived moderate‐to vigorous‐intensity physical activity (MVPA), diet density and clustering by school class. (A) C‐reactive protein (C RP, mg L−1) and (B) Soluble vascular adhesion molecule‐1 (sVCAM‐1, ng mL−1). a P = 0.02, b P = 0.03; b (95% CI), exponentiated regression coefficients and 95% confidence intervals.
Table S1. Associations of sedentary time patterns with cardiometabolic biomarkers in school aged children. a P = 0.02, b P = 0.03; b (95% CI), exponentiated regression coefficients and 95% confidence intervals. Model 1 adjusted for clustering within school classes. Model 2 further adjusted for sex and waist circumference. Model 3 further adjusted for moderate‐to vigorous‐intensity physical activity (MVPA). Model 4 further adjusted for diet density.