Obesity has been recognized as a significant public health challenge worldwide. The obesity epidemic is of concern in Canadian adolescents due to the dramatic increase in prevalence over the past 25 years [¹]. Further, numerous health consequences of adolescent obesity exist including risk for cardiovascular disease, type 2 diabetes [²], and persistence of obesity into adulthood [³, ⁴]. Equally concerning are psychosocial consequences such as discrimination and teasing that may occur, which can have a long-term effect on self-esteem and body image [⁵]. Obesity presents an enormous burden to our health care system, estimated in the billions of dollars, which is only expected to increase with the dramatic rise in obesity prevalence [⁶].

Given the numerous consequences of obesity, surveillance systems and measurement tools are needed to determine the effectiveness of obesity prevention programs and interventions [⁷]. However, it is becoming increasingly difficult to collect relevant and geographically diverse information in a time- and cost-effective manner. Web-based methods of surveillance provide a cost-effective means to survey large numbers of participants concurrently throughout wide geographic areas. Access to hard-to-reach areas is enhanced, thus improving the ability to obtain a representative sample [⁸]. The Web Survey of Physical Activity and Nutrition (Web-SPAN), a self-administered web-based survey for adolescents, has previously been validated [⁹]. However, Web-SPAN has not yet been used to determine the relationship between bodyweight status and health behaviours such as dietary intake and suboptimal meal behaviours. Therefore, this study provides the much-needed insight on how web-based methods of surveillance can be used to further our understanding of the factors that contribute to overweight/obesity in youth and will help inform intervention development and refinement.

Although several studies have examined the relationship between overweight/obesity and obesity determinants, few have utilized a web-based platform in order to capture this information. Previous studies have indicated that obesity is a complex condition involving individual/behavioural, environmental, and social factors. Further, it has been proposed that a basic imbalance of energy intake and energy expenditure leads to increased adiposity [¹⁰]. Specifically, individual behaviours such as increased consumption of high calorie foods (which include high calorie beverages) [¹, ¹¹–¹³], physical inactivity, and increased sedentary activity [¹⁴–¹⁶] have been identified as key factors contributing to the development of obesity.

Therefore, the purpose of this study was to use a web-based method of surveillance to investigate the current bodyweight status of Alberta adolescents and to identify whether non-overweight students were different from their overweight or obese peers with respect to macronutrient intakes, consumption of Other Foods (foods not included in Canada's Food Guide), suboptimal meal behaviours including breakfast skipping and consuming meals away from home, and physical activity levels. It was hypothesized that non-overweight students would have different intakes including lower intakes of fat and Other Foods, and a lower frequency of meal skipping and consuming meals away from home. Further, it was hypothesized that non-overweight students would be more active than their overweight and obese peers.

The purpose of this study was to investigate the overweight and obesity status of Alberta adolescents and to identify differences in nutrient intakes, meal behaviours, and physical activity levels between BMI classifications using a web-based survey. One in five students were considered overweight or obese. They had significantly different nutrient intakes, meal behaviours, and physical activity levels than non-overweight students, such that suboptimal behaviours (e.g., skipping breakfast) were more prevalent in students classified as overweight or obese.

Self-reported height and weight were used to calculate BMI and to further classify students as overweight or obese. Although measured height and weight are more precise and accurate than self-reported measurements, self-report is a useful method that offers researchers a time- and cost-effective method to survey a large number of participants simultaneously [²⁴, ³⁹–⁴¹]. Overweight and obesity rates in the current study were similar to the Alberta rates from the 2004 CCHS [¹]. The 2004 CCHS revealed that Alberta youth had a significantly lower prevalence of overweight/obesity than the national average, therefore it was most appropriate to compare to provincial statistics [¹].

The association between dietary fat and weight status in the literature is inconclusive. Several researchers have suggested that an increase in dietary fat intake can be correlated to an increase in body fat in children 4 to 11 years [⁴²–⁴⁴]. However, Gillis and colleagues [⁴⁵] and Troiano et al. [⁴⁶] did not observe the same effect. In the current study, both overweight and obese students consumed significantly more total fat compared to non-overweight students, suggesting a relationship between dietary fat intake and BMI classification. In addition, overweight and obese students had significantly lower intakes of carbohydrate and fibre compared to their peers. Although mean intakes of total fibre were well below the recommended Dietary Reference Intake (DRI) in all BMI categories, higher intakes were observed in non-overweight students, representing an improvement in fibre intake with decreasing BMI. Significant differences were not observed between BMI classification and diet quality; however, recent findings from the 2004 CCHS suggested that youth who consumed fruits and vegetables more frequently (five or more times a day) were less likely to be classified as overweight/obese [¹]. However, it is worth noting that diet quality was very poor in all students, regardless of weight status, which could explain the lack of significance in results [¹⁷]. Further, the food-based diet quality index used in the present study did not determine when students exceeded their CFGHE recommendations, which may provide a better understanding of the relationship between diet quality and overweight/obesity.

Although results did not indicate significance for the consumption of Other Foods subcategories, univariate analysis revealed that obese students consumed significantly more high calorie beverages compared to non-overweight students. The relationship between high calorie beverages (e.g., regular soft drinks) and body weight is of interest due to the tremendous increase in soft drink consumption in the adolescent population. In the United States, data from the National Heart, Lung, and Blood Institute Growth and Health Survey assessed longitudinal changes in beverage consumption among adolescent girls. Girls were assessed initially at age 9 or 10 years and annually up until 19 years of age. Data from multiple 3-day food records indicated that regular soft drink consumption nearly tripled over the 10-year period. In that study, for every 100 grams of regular soft drinks consumed, BMI increased by 0.01 unit and average daily caloric intake increased by approximately 82 kilocalories [¹²]. A more recent Australian study indicated that schoolchildren aged 4 to 12 years who consumed greater intakes of high calorie beverages (i.e., fruit juice/drink, soft drink) were more than twice as likely to be overweight/obese [¹³]. A meta-analysis which examined the effects of soft drink consumption on nutrition and health revealed positive associations between soft drink consumption and body weight, although results varied depending on the methodology used to assess body weight [⁴⁷].

Results of the present study indicated that non-overweight students had a higher frequency of consuming breakfast and snacks when compared to either overweight or obese students. Breakfast skipping is highly prevalent in the adolescent population, and consistent with previous reports, has been associated with increased prevalence of overweight [⁴⁸]. Although snack consumption has been previously associated with higher intakes of Other Foods [⁴⁹], the current results suggest that snacks may be associated with a lower BMI classification. Additional research that examines meal patterns may be helpful in clarifying the association between snack consumption and weight. Although significant differences were not observed between BMI classification and consumption of meals and snacks away from home, previous studies have found a positive association [⁵⁰, ⁵¹]. This suggests that location alone may not be adequate to explain the relationship between meal behaviours and overweight/obesity. Information regarding the type (e.g., fried versus grilled, white versus whole wheat, whole versus skim milk, juice versus fruit) and quantity of the food consumed away from home may be necessary to provide a true understanding of the relationship between BMI classification and meal location. As well, in the current study, the frequency of family meals was not examined. Family meals may have provided a better understanding of BMI classification than consumption of meals away from home at independent locations.

Results from the current study indicated that both non-overweight and overweight students were more active than obese students. Further, differences were observed by gender where both non-overweight and overweight boys were more active than obese boys, and non-overweight girls were more active than obese girls. These findings are consistent with results from the 2004 CCHS which indicated that sedentary adolescent boys were more likely than active boys to be obese [¹]. However, these results are dissimilar to other reports from the Canadian Health Measures Survey that utilized objective measuring devices for physical activity such as accelerometers [⁵²]. In a study done by Thompson et al. [¹⁴], no significant differences were found in physical activity levels among BMI classifications.

A limitation of this study was the use of self-reported survey data. Although validated survey components were used, many adolescents experience recall bias, response bias, and underreporting of total energy intake which is known to occur in the adolescent population [⁵³]. Strengths of this study include the large sample size, the use of validated questionnaires to assess behaviours, and the web-based methodology which facilitated access to a large number of students over a large geographic area. The survey also provided anonymity and privacy which have been shown to be important determinants of adolescents' decision to report honest answers [⁵⁴]. Further, students were not aware in advance what day the survey was to be completed, therefore reducing reactivity.

In conclusion, this study examined the overweight and obesity status of Alberta adolescents and the associations between BMI classification and lifestyle behaviours including nutrient intakes, meal behaviours, and levels of physical activity using a web-based survey. Web-SPAN was a time- and cost-effective method that was used to assess health behaviours of adolescents throughout a wide geographic area, including hard-to-reach areas. Significant differences in intake, meal behaviours, and physical activity levels existed between BMI classifications. This research supports the need to target suboptimal behaviours such as high calorie beverage consumption, total fat intake, breakfast skipping, and low levels of physical activity in order to promote healthy weights in the adolescent population. Because behaviours established during adolescence have been shown to track into adulthood, it is important to target these behaviours during early adolescence. In Alberta, vending machines have been removed from many schools, and daily physical activity is mandatory in grades 1–9; however, this research supports the need to focus on older adolescents, including those in high school (grades 10 through 12). Additional nutrition education in schools may also help target suboptimal nutritional intakes such as high intakes of total fat and low intakes of fibre, by educating students on the importance of a healthy diet. Thus, priorities that aim to improve the whole school environment such as comprehensive school health which includes school nutrition policies, removal of vending machines, and mandatory physical activity may help to improve overall weight status in Alberta adolescents. Further, because diet quality was poor among all Alberta adolescents, regardless of weight status, priorities should be developed that target students in all grade levels in order to improve the overall health and well-being of all adolescents.

The authors declare that they have no conflict of interests.

This research was supported by Alberta Innovates-Health Solutions (AIHS; formerly the Alberta Heritage Foundation for Medical Research [AHFMR]) and the Canadian Institutes of Health Research/Heart and Stroke Foundation (CIHR/HSF). K. E. Storey received funding from the Danone Institute of Canada (Doctoral Student Award), CIHR/HSF, and AIHS Health Research Studentship. R. C. Plotnikoff and K. D. Raine were funded by Salary Awards from the CIHR and AIHS; K. D. Raine was also funded from an HSF Salary Award.