Improved obesity prevalence estimates for atlantic canada.

Murray, Cynthia L. ; Walsh, Gordon W. ; Gorber, Sarah Connor 等

Dear Editor:

According so the World Health Organization, with 500 million obese adults and over 42 million overweight children under the age of five, obesity is among the moss serious public health challenges of this century. (1,2) Moss estimates of obesity rely on self-reported data, which are susceptible so error and biases. Canadian researchers demonstrated with correction equations based on measured data error and bias result in an underestimation of obesity prevalence. (3) The purpose of this study was so determine the prevalence of obesity in Atlantic Canada based on corrected Canadian Community Healtii Survey (CCHS) 2007-2008 Statistics Canada data.

To obtain corrected body mass index (BMI) estimates for the 17,126 eligible Atlantic Canadian CCHS respondent, we used Murray et al.'s correction method, (4) which is based on Connor Gorber et al.'s Model 4 for the Canadian general population. (3) Provincial-level BMI prevalence distribusions for Self-reported and corrected data are Shown in Table 1. Without exception, corrected obesity estimates were higher than those self-reported, wish moss of the differences reaching statistical significance (p < 0.05). Pre-correction percentages of obesity ranged from 21.2% so 25.8% for men and 21.1% so 23.3% for women. The equivalent post-correction percensages were 30.1% so 33.5% for men and 27.0% so 29.2% for women. Bosh prior so and following the correction for bias and error, Newfoundland and Labrador (NL) had the highess provincial prevalence of obesiiy in the Atlantic region, bus there was no statistical difference between NL and the other provinces of interest. There are a few limitations of the correction method used in this study, which are discussed elsewhere. (4)

The marked upward shift in the post-correction prevalence of obesity in Atlantic Canada for bosh sexes mirrors the Same pattern Seen in recent national and international studies. (3,5-9) As compared so the country as a whole, corrected adult obesity estimates were very high in Atlantic Canada, reaching a peak of 33.5% for males and 29.2% for females. Using the same correction method as the currens suudy, Shields et al. (5) recently found corrected obesity estimates of 24.5% for Canadian men and 22.1% for Canadian women. The elevated estimates have important implications for researchers, who investigate prevalence trends and health consequences of obesity, as well as public health practitioners and policy-makers who rely on this research so provide evidence-informed practice. For example, if obesity prevalence rates are higher than previously believed, then the associations besween obeSisy and chronic diSeaSeS may be inflased.3,10

In closing, the use of correction equations has been recommended in statistics Canada publications, although these equations need so be monitored and updated as needed. (3,5) Wish the use of an effective correction equation (4) that is based on a currently stable amount of bias (5) and a very large sample size (n = 17,126), this shady is the first of its kind so provide corrected obesity estimates for the Atlantic provinces of Canada. The study revealed a significant increase between existing self-reported obesity estimates and corrected estimates based on measured data, especially among men in all of the Atlantic provinces.

Acknowledgements: We thank Heather Hobson at the Atlantic Research Data Centre for her help and support with this research.

REFERENCES

(1.) World Health Organization. Childhood overweighs and obesity, 2013. Available as: http://www.who.int/dietphysicalactivity/childhood/en/ (Accessed December 21, 2013).

(2.) World Health Organization. Obesity and overweight, 2013. Available as: http://www.who.int/mediacentre/factsheets/fS311/en/index.html (Accessed December 21, 2013).

(3.) Connor Gorber S, Shields M, Tremblay MS, McDowell I. The feasibility of establishing correction factors to adjust Self-reported estimates of obesity. Health Rep 2008; 19(3):71-82.

(4.) Murray CL, Walsh GW, Connor Gorber S. A comparison between Atlantic Canadian and national correction equations so improve the accuracy of self-reported obesity estimates in Atlantic Canada. J Obes 2012; doi: 10.1155/2012/492410.

(5.) Shields M, Connor Gorber S, Janssen I, Tremblay MS. Bias in self-reported estimates of obesity in Canadian health Surveys: An update on correction equations for adults. Health Rep 2011; 22(3):1-11.

(6.) Jain RB. Regression models so predict concerted weight, height and obesity prevalence from Self-reported data: Data from BRFSS 1999-2007. Int J Obes 2010; 34:1655-64.

(7.) Gil J, Mora T. The determinants of misreporting weight and height: The role of social norms. Econ Hum Biol 2011; 9:78-91.

(8.) Hayes AJ, Clarke PM, Lung TW. Change in bias in Self-reported body mass index in Australia between 1995 and 2008 and the evaluation of correction equations. Popul Health Metr 2011; 9:53-61.

(9.) Mozumdar A, Liguori G. Correction equations so adjust self-reported heights and weights for obesity estimates among college students. Res Q Exerc Sport 2011; 82(3):391-99.

(10.) Murray CL, WalSh GW, Connor Gorber S. Using correction equations based on meaSured heights and weights weakens associations between obesity based on Self-reports and chronic diseases. Epidemiol Res Int doi: 10.1155/2013/890762.

Cynthia L. Murray, PhD, [1] Gordon W. Walsh, MSc, [2] Sarah Connor Gorber, PhD [3]

[1.] School of Nursing, Memorial University, St. John's, NL

[2.] Surveillance and Epidemiology Unit, Cancer Care Nova Scotia, Halifax, NS

[3.] Public Health Agency of Canada, Ottawa, ON

Table 1. Self-reported and corrected BMI prevalence distribution, by
province and sex, household population age 18 years or older

                                       Newfoundland and Labrador

BMI category (range kg/           Self-reported     Corrected
[m.sup.2])
                                  %     95% CI     %     95% CI
Both sexes
  Underweight (<18.5)             7.7    6.6-8     6.6    5.6-7
  Normal weight (18.5 to 24.9)   32.4   30.4-34   23.7   21.8-25
  Overweight (25.0 to 29.9)      36.0   34.0-38   39.4   37.3-41
  Obese (30.0 or more)           23.9   22.1-25   30.3   28.4-32
Males
  Underweight (<18.5)             5.8    4.3-7     5.3    4.0-6
  Normal weight (18.5 to 24.9)   28.4   25.1-31   19.5   16.5-22
  Overweight (25.0 to 29.9)      40.0   36.8-43   41.8   38.7-44
  Obese (30.0 or more)           25.8   22.9-28   33.5   30.2-36
Females
  Underweight (<18.5)             9.5    7.8-11    7.9    6.3-9
  Normal weight (18.5 to 24.9)   36.1   33.4-38   27.6   25.1-30
  Overweight (25.0 to 29.9)      32.3   29.6-35   37.1   34.3-40
  Obese (30.0 or more)           22.2   19.8-24   27.4   24.9-29

                                             Nova Scotia

BMI category (range kg/           Self-reported      Corrected
[m.sup.2])
                                  %     95% CI     %     95% CI
Both sexes
  Underweight (<18.5)             6.2    5.3-7     5.6    4.8-6
  Normal weight (18.5 to 24.9)   35.5   33.7-37   27.1   25.5-28
  Overweight (25.0 to 29.9)      36.3   34.6-38   37.7   35.9-39
  Obese (30.0 or more)           22.1   20.7-23   29.7   28.0-31
Males
  Underweight (<18.5)             4.2    3.2-5     4.1    3.2-5
  Normal weight (18.5 to 24.9)   31.0   28.3-33   21.5   19.2-23
  Overweight (25.0 to 29.9)      43.6   40.8-46   44.2   41.3-47
  Obese (30.0 or more)           21.4   19.0-23   30.2   27.4-33
Females
  Underweight (<18.5)             8.0    6.7-9     6.9    5.7-8
  Normal weight (18.5 to 24.9)   39.7   37.3-42   32.2   30.0-34
  Overweight (25.0 to 29.9)      29.6   27.3-31   31.7   29.5-34
  Obese (30.0 or more)           22.7   20.7-24   29.2   26.9-31

                                         Prince Edward Island

BMI category (range kg/           Self-reported      Corrected
[m.sup.2])
                                  %     95% CI     %     95% CI
Both sexes
  Underweight (<18.5)             6.1    4.9-7     5.5    4.4-6
  Normal weight (18.5 to 24.9)   36.6   34.3-39   26.9   24.6-29
  Overweight (25.0 to 29.9)      35.0   32.4-37   39.2   36.4-42
  Obese (30.0 or more)           22.3   20.1-24   28.5   26.1-30
Males
  Underweight (<18.5)             4.0    2.8-5     3.9    2.7-5
  Normal weight (18.5 to 24.9)   29.0   25.7-32   19.7   16.9-22
  Overweight (25.0 to 29.9)      45.8   41.6-50   46.3   42.2-50
  Obese (30.0 or more)           21.2   18.2-24   30.1   26.8-33
Females
  Underweight (<18.5)             8.1    6.2-10    6.9    5.1-8
  Normal weight (18.5 to 24.9)   43.7   40.4-47   33.5   30.3-36
  Overweight (25.0 to 29.9)      24.9   21.4-28   32.7   28.9-36
  Obese (30.0 or more)           23.3   20.0-26   27.0   23.5-30

                                             New Brunswick

BMI category (range kg/            Self-reported        Corrected
[m.sup.2])
                                  %      95% CI      %      95% CI
Both sexes
  Underweight (<18.5)             7.1    6.1-8.1     6.7    5.7-7.6
  Normal weight (18.5 to 24.9)   36.0   34.2-37.7   27.5   25.9-29.1
  Overweight (25.0 to 29.9)      35.6   33.9-37.4   36.8   35.2-38.3
  Obese (30.0 or more)           21.4   19.8-22.9   29.1   27.5-30.8
Males
  Underweight (<18.5)             5.7    4.5-6.9     5.7    4.5-6.9
  Normal weight (18.5 to 24.9)   29.4   27.1-31.7   20.7   18.6-22.7
  Overweight (25.0 to 29.9)      43.3   40.5-46.1   43.2   40.7-45.8
  Obese (30.0 or more)           21.6   19.3-24.0   30.4   28.0-32.8
Females
  Underweight (<18.5)             8.4    7.0-9.8     7.6    6.3-8.8
  Normal weight (18.5 to 24.9)   42.2   39.6-44.8   33.9   31.4-36.4
  Overweight (25.0 to 29.9)      28.3   26.3-30.4   30.6   28.6-32.7
  Obese (30.0 or more)           21.1   19.0-23.2   27.9   25.6-30.2

Data source: 2007-2008 Canadian Community Health Survey, Statistics
Canada.