Measured Weights in PEI Adults Reveal Higher Than Expected Obesity Rates

Abstract

Background: The increasing prevalence of obesity and overweight in Canada is a significant health concern. Unfortunately, we know very little about the actual weight status and associated health risks in our population since most surveys use only selfreported body weights and heights and typically do not include a measure of body fat distribution. This paper summarizes the findings of the Prince Edward Island Nutrition Survey. Methods: A random sample of 1,995 adults aged 18-74 were interviewed in their homes and weights, heights and waist circumference measurements were obtained. Relative health risks, population proportions and their corresponding 95% confidence intervals were calculated. Results: Overall, almost one third of PEI adults are obese (BMI 30). This is almost double that reported in the 1995 National Population Health Survey using self-reported heights and weights. More women were classified as being very severely obese (Class III) than men, but for both men and women there appears to be a trend of increased mild obesity with age. Based on BMI and waist circumference, over one third of the population is considered to be at high to extremely high risk for health problems. Conclusion: Self-reported height and weight data appear to result in significant underestimation of the problem of obesity. Given the serious health consequences associated with this condition, it is critical that measured heights and weights be collected in future population-based surveys to ensure that public health interventions are based on accurate prevalence data. La traduction du résumé se trouve à la fin de l'article.

Prevalence of obesity in Canada

“…The inconsistency around obesity prevalence estimates because of self-report for height and weight has recently been investigated (15). Comparing measured data collected in 1995 adults from PEI (one of the 10 Canadian provinces) with self-reported anthropometric data from the 1995 NPHS, MacLellan et al .…”

Gender differences in growth and nutrition in a sample of rural ontario schoolchildren

“…While there are few studies of obesity in rural North American adults, they consistently report a significant gender difference in obesity prevalence. Women have higher obesity prevalence than men in both Canadian and US studies (Borders et al, 2006;Liebman et al, 2003;MacLellan et al, 2004;Self et al, 2005). This gender difference is less consistently observed in studies of rural children.…”

“…There is also national-level literature that demonstrates heightened obesity risk for North Americans experiencing socioeconomic deprivation (Alaimo et al, 2001;Gordon-Larsen et al, 2003;Phipps et al, 2006;Willms et al, 2003). However, researchers stress the need for directly measured data at both the national and local levels in order to develop understandings of the global and proximate causes of obesity in the environment (MacLellan et al, 2004;Tremblay, 2004).…”

Adjusting Divergences between Self-reported and Measured Height and Weight in an Adult Canadian Population

“…Without adjustment, these measures could lead to a significant underestimation of the obesity prevalence due to misclassification in BMI category. 16,23 Such misclassification may affect not only the magnitude but also the direction of the predicted relationships between obesity and illhealth, thereby blurring understanding of the association between obesity and disease risk factors, which may ultimately influence public policy decisions. To the best of our knowledge, this is the first study to make use of telephone surveys in developing correction equations to adjust deviations between self-reported and measured height and weight in the Canadian context.…”

“…10 Other determinants such as income, education and occupation also have been linked to the discordance between self-reported and actual height and weight measurements. 14,17 A number of studies argue that systematic errors in self-reported height and weight lead to an underestimation of the proportion of overweight or obese people in the general population, 23 which indisputably poses challenges in measuring and monitoring the prevalence and trends of obesity, distorts the classification of individuals with regard to chronic conditions, 8 and ultimately, confounds the association of BMI with mortality. 24 Given that most health surveys collect self-reported data, the idea of adjusting self-reported height and weight has been explored in a handful of empirical studies in the Canadian context.…”