The main findings in this study were that level of parental education, daily breakfast consumption and MVPA were inversely associated with weight status, and time spent watching TV was positively associated with weight status in a sample of Norwegian 11-year-olds. The association between watching TV and weight status turned out to be significant for boys only.
The prevalence of overweight and obese adolescents in the present study does not differ substantially from other recent studies of Norwegian adolescents
The lack of association between sugar-sweetened beverages and weight status may be surprising given that WHO has categorized it as a probable contributor
]. However, the intake of sugar-sweetened beverages was low in both groups, which has also been observed in another recent Norwegian study
]. Yet, underreporting by the overweight cannot be ruled out, especially due to the high awareness about the sugar-sweetened beverages and health.
The association between weight status and consumption of unhealthy snacks was not significant. As a matter of fact, overweight adolescents tended to eat less snacks than non-overweight. This finding might be explained by underreporting or changed dietary patterns by overweight adolescents as a consequence of dieting. However, prospective studies of snack food consumption have consistently failed to show a link between snack food intake and excess weight
]. Our results add to these findings.
An association between breakfast consumption and weight status has been identified in other cross-sectional studies
]. We found a significant negative association in the univariate analyses, and also when adjusting for other factors. Affenito et al. (2005) found a negative association between breakfast consumption and BMI after adjusting for demographic characteristics in the NHLBI Growth and Health study, but the association did not persist after multivariate control for physical activity and energy intake
]. The authors of the study interpreted this observation to suggest that breakfast consumption is a marker for other healthy behaviors. A large US longitudinal study reported that normal-weight children who never ate breakfast gained weight relative to peers who ate breakfast nearly every day
]. Must et al. (2009) points out that breakfast consumption as well as consumption of sugar-sweetened beverages and TV-viewing seem to be operating indirectly, as proxies for other dietary or activity behaviors, and warrants further studies of these relationships
Regarding time watching TV and playing computer games, the differences between non-overweight and overweight adolescents were quite high, especially during weekdays. Watching TV was strongly associated with being overweight in our study, with 40% increased odds of being overweight with every additional hour of watching TV per day in the total sample. However, gender moderated this association and the subgroup analyses revealed a significant association for boys only. A review from prospective studies on modifiable risk factors in relation to changes in BMI and fatness concluded that sedentary behaviors effect on weight status seems to differ by gender, with many studies but not all showing greater positive associations among girls
]. Possible gender differences are important to be aware of when designing intervention efforts targeting overweight/obesity prevention. Watching TV is the most studied sedentary behavior, and some studies suggest that TV-viewing operates through the impact on energy intake rather than displacement of more energy demanding activities. Cross-sectional data from the Danish part of European Youth Heart Study showed inverse associations between watching TV and both healthy food preferences and healthy food habits in school aged children
]. In a laboratory-based study including 9 to 14-year-old boys, watching TV during a meal seemed to delay normal mealtime satiation and reduce satiety signals from recently consumed food, increasing energy intake
]. It has also been suggested that exposure to commercials advertising energy-dense food while watching TV can work as an indirect mechanism and increase the energy intake even more
]. However, in Norway there are legislations restricting such TV commercials aimed at children and this issue should only be of limited importance.
We found a strong association between use of computer games and weight status in the univariate analysis, but this association did not remain when adjusting for the other factors, indicating a confounding effect of other variables on this relationship. Tremblay et al. (2011) found a dose–response relationship between increased sedentary behavior and unfavorable health outcomes in a large systematic review of sedentary behavior on health indicators based on 232 studies of school-aged children and youth
]. The results are based on data from almost 1 million young people. However, these are self-reported sedentary behaviors measured as a mixture of hours/minutes/times per week of watching TV/playing computer games/screen time or self-reported sedentary time. A Canadian study using an objective measure of sedentary behavior (Actiwatch) found that fat mass and percentage body-fat were positively correlated with time spent sedentary for girls but not boys
]. We found no significant association with time spent sedentary measured by accelerometer and weight status. The lack of consistency observed in the studies of sedentary behavior and sedentary time may reflect the range of variable definitions, measurement challenges, and also the changing nature of electronic media.
For total physical activity (mcpm) we found a significant difference between genders and between non-overweight and overweight adolescents. We found a significant inverse association between MVPA and being overweight, but no moderating effect of gender on this association. Ekelund et al. (2012) combined data from multiple cohorts of accelerometer assessed physical activity of 20 871 children and adolescents (age 4–18 years) and found that both total physical activity and time in MVPA were significantly and negatively associated with waist circumference
]. However, in a systematic review of prospective studies of objectively measured physical activity and obesity prevention in children, adolescents and adults, the authors conclude that physical activity might not be a key determinant of excessive gain in adiposity
]. Currently, the literature is inconsistent in the relationship between physical activity and adiposity.
Earlier findings from the HEIA study showed differences in anthropometric characteristics and prevalence of overweight when stratified by parental education level
]. The current results show that these differences by parental education level remain when adjusting for behavioral factors. Level of parental education was inversely associated with weight status in the adjusted model, and this association was not moderated by gender. These results confirm previous studies suggesting a social gradient in the problem of overweight/obesity among Norwegian adolescents
]. While earlier international research (from 1941–1989) found inconsistent relationships between SES and childhood adiposity
], more recent research indicates a shift in trends where most studies show inverse relationships
]. The ENERGY Project, a recent school-based survey among 10–12 year olds conducted in seven European countries (n
7234), found more favorable indicators of weight status in children of higher educated parents than in children of lower educated parents
]. Our results support the evidence that adolescents of parents with low education have a higher risk of being overweight than adolescents of parents with higher education independent of behavioral correlates and gender.
Limitations and strengths
There are several limitations that should be considered when interpreting the findings from this study. It is impossible to infer causal relationships and to determine its direction from cross-sectional data. We cannot be certain that other unmeasured confounders could not have impacted our findings. Although accelerometers are considered a preferred tool when assessing physical activity and sedentary time in large surveys, it has weaknesses when it comes to measuring water-activities, cycling, skiing/skating, carrying loads, inclines and upper-body movements. We made no attempts to correct for these weaknesses, as to our knowledge there are no valid techniques to do so for a large scale survey. This may represent a possible underreporting of total physical activity. Furthermore, there was a rather large proportion of invited schools that declined to take part in the study. Recruiting schools in Norway to extra-curricular projects has become a challenging undertaking the last decade, as the curricular demands the last years have increased substantially. In addition, weighing of children is a controversial issue in Norway and has been debated in the national media repeatedly. However, attrition analyses showed no differences between the participating schools (n
37) and schools which declined to participate (n
140) in terms of number of students in 6th
grade and overall size (data not shown)
]. Also, the sample is collected from seven counties surrounding the county of Oslo, and this may limit the possibility to generalize the results for 11-year-olds outside this area. However, comparing the HEIA study sample to nationally representative figures for 9 and 15-year-olds, the measures from the participants in the HEIA study lie adequately between the measures of the 9 and 15-year-olds when it comes to objectively measured height, weight and total physical activity
One of the major strengths of the study is the relatively large sample of adolescents at a very narrow age range. The age of 10–11 years is an important age group for addressing efforts to promote healthy behaviors
]. The fact that many adolescents at this age have entered puberty makes this specific age group considered as hard to study and therefore less studied than pre- and post-pubertal adolescents. We adjusted for pubertal development and taking this into account is one of the strengths in this study. Other strengths of the study are that we measured height, weight, physical activity and sedentary time objectively. All adolescents in the present study had valid accelerometer measures of physical activity and sedentary time. An additional strength is that we were able to collect data on parental education from nearly all the parents giving their consent for their child to participate in the study. Finally, investigating both behaviors related to energy intake and expenditure in the same study concerning weight status can be considered as a strength, as the intrinsic interplay among dietary behaviors, physical activity and sedentary time still needs further understanding.