Data sources and analytical techniques

Data from the 2004 Canadian Community Health Survey (CCHS): Nutrition were used to produce overweight and obesity rates for 2- to 17-year-olds. The 2004 CCHS was designed to gather information at the provincial level about the nutritional status of the Canadian population (see /concepts/hs-es/index-eng.htm#content ). The survey does not include residents of the three Territories, Indian reserves and some remote areas, and regular members of the Canadian Armed Forces. The response rate was 76.5%. Measured height and weight were obtained for 66% of the 2- to 17-year-olds who responded to the 2004 CCHS, a total of 8,661 individuals (see Limitations).

Overweight and obesity rates of American children and adolescents were estimated from the 1999-2002 National Health and Nutrition Examination Survey (NHANES). The NHANES obtained measured height and weight data for 7,297 children and adolescents.

Historical estimates of Canadian overweight and obesity rates, based on directly measured height and weight, are from on the 1978/79 Canada Health Survey, and for 12- to 17-year-olds, from the 1981 Canada Fitness Survey and the 1988 Campbell’s Survey on Health and Well Being. For 12- to 17-year-olds, rates based on self-reported data are from the 2000/01 and 2003 CCHS and the 1994/95, 1996/97 and 1998/99 National Population Health Survey (NPHS). For 2- to 11-years olds, rates based on data reported by parents are from the 1994/95, 1996/97, 1998/99, 2000/01 and 2002/03 National Longitudinal Survey of Children and Youth (NLSCY). In 2002/03, NLSCY estimates could be made only for children aged 2 to 5, since the cross-sectional file does not include records for children aged 6 or older.

Descriptive statistics from the 2004 CCHS were used to estimate the proportion of 2- to 17-year-olds who were overweight and obese in relation to selected characteristics. All estimates were based on the 8,661 children and adolescents for whom measured data on height and weight were obtained. Since they accounted for only 66% of children and adolescents who responded to the 2004 CCHS, an adjustment was made to minimize non-response bias. A special sampling weight was created by redistributing the sampling weights of the non-respondents to the respondents, using response propensity classes. Variables such as province, age, sex, household income, race, education, physical activity, fruit and vegetable consumption and chronic conditions were used to create the classes. The classes were created with the CHAID (Chi-Square Automatic Interaction Detector) algorithm available in Knowledge Seeker¹⁶ to identify the characteristics that best split the sample into groups that were dissimilar with respect to response/non-response. This adjusted sampling weight was used to produce all estimates in this analysis. Standard errors and coefficients of variation were estimated using the bootstrap technique, which accounts for the survey design effects.^{17, 18, 19}

The American and Canadian historical estimates are based on weighted data. The criteria specified by the International Obesity TaskForce (IOTF) were used to define overweight and obesity among youth (see Calculating overweight and obesity in children and adolescents).

Standard errors and coefficients of variation for estimates from the 1978/79 Canada Health Survey and the National Health and Nutrition Examination Survey (NHANES), 1999-2002 were estimated with SUDAAN, which uses a Taylor series linearization method to account for the complex survey sample design.

The body mass index (BMI) distribution (Chart 3) was smoothed by calculating three-point averages. For example, the percentage of the population with a BMI of 23 was calculated by summing the percentage with a BMI of 22, the percentage with a BMI of 23 and the percentage with a BMI of 24, and then dividing the result by 3.

Chart 3. Percentage distribution of body mass index (BMI), household population aged 12 to 17, Canada excluding territories, 1978/79 and 2004