National Statistics and Opportunities for Improved Surveillance,
Prevention, and Control

LIMITATIONS OF CURRENT NATIONAL DATA SOURCES

Our current information on diabetes and its risk factors is limited by a number of factors, summarized below.

Limitations on Diabetes Information

• Incomplete Data Sources

  Diabetes is generally diagnosed and treated in physicians' offices, from which data collection is difficult. Data from hospitalization records capture only people with more severe diabetes. Therefore, the prevalence of diabetes in the Canadian population is underestimated by current data sources. Most "national" data sources are also geographically incomplete in that they exclude the territories.

• Survey Data Are Unsubstantiated

  There are limitations to survey data because the presence of diabetes is based on self-report. Some people with diabetes do not report having diabetes when questioned; conversely, some report diabetes when they do not have the condition.

• Inability to Capture Undiagnosed Diabetes

  Survey data do not detect people with undiagnosed diabetes unless a blood sample is taken. Administrative databases (such as physician billing databases) do not identify these people either.

• Inaccurate Mortality Data

  Mortality data are limited, in that diabetes is often not cited as the underlying cause of death in cases of death due to the complications of diabetes.

• Type 1 and 2 Diabetes Undifferentiated

  Most data sources do not differentiate between Type 1 and Type 2 diabetes.

• Small Samples in National Surveys Population surveys have not had an adequate sample size nationally to allow comparison on a subprovincial or sometimes even provincial level.

• Lack of Aboriginal Data

  Aboriginal status has not been included as a variable in most national surveys. Some reserves were not included in the Aboriginal Peoples Survey 1991, with the result that roughly 58,000 potential respondents had no opportunity to be included in the sample.

• Different Methodologies in National Surveys

  National health surveys that have been conducted over the past 20 years differ in their methodologies such that comparison of diabetes prevalence over time is limited.

  Limitations on Risk Factor Information

• Lack of National Risk Factor Surveillance System

  Canada does not have an established national risk factor surveillance system comparable to the Behavioural Risk Factor Surveillance System (BRFSS) in the United States. However, a similar risk factor surveillance system is currently in development at the Laboratory Centre for Disease Control, Health Canada.

• Physical Measures in Few National Surveys

  Few national surveys in Canada incorporate physical measures, such as blood pressure or weight, and none on the scale of the National Health and Nutrition Examination Survey (NHANES) in the United States. Thus, our data on known or potential risk factors, such as hypertension and central deposition of fat, are limited.

• Social Desirability Bias in Self-Reports

  Reliance on self-report for data on risk factors such as obesity and physical inactivity introduces the possibility of a social desirability bias in the results (i.e., respondents claim to be thinner and more physically active than they actually are), with the result that survey data on these risk factors are probably inaccurate.

[Previous] [Table of Contents] [Next]