In order to make national, regional and global predictions for the prevalence of diabetes, a number of assumptions needed to be made, and therefore the results are subject to a number of limitations. In addition to those highlighted in the Methodology section, some of these are that:

• The studies included in this section often used differing screening techniques. The majority of studies used an OGTT to screen for diabetes. However, some studies used a fasting blood glucose (FBG), some a two-hour blood glucose (2BG), some a random blood glucose (RBG), and some based their data on self-report (SR). It is difficult to control for this unless, for example, only those studies that used an OGTT were included. This would also have the effect of excluding findings from countries lacking OGTT data, which would result in data for those countries being extrapolated from another country. The other consequence of incorporating studies that had no OGTT data is that impaired fasting glucose (IFG) rather than IGT represented the non-diabetic, but abnormal, glucose metabolism.
• There were inconsistencies in the diagnostic criteria adopted, resulting from the updating of the diagnostic criteria in 1997   1 . The use of a lower fasting diagnostic criterion for diabetes will tend to result in a higher prevalence of diabetes and lower prevalence of IGT. The diagnostic criteria used for each country are indicated in the Tables on data sources.
• Studies from several countries (Canada, France, Germany, Israel, Italy, Netherlands, New Zealand, Norway) only provided data on self-reported diabetes. To account for undiagnosed diabetes, the prevalence of diabetes for Canada was multiplied by a factor of 1.5, in accordance with findings from the USA   2 , and for the other countries doubled, based on data from a number of countries   3   4   5   6   7 .
• If a country lacked data, it was assumed that their age and sex-specific prevalence rates of diabetes mellitus were the same as those rates in another socio-economically, ethnically and geographically similar country.
• Some of the studies were performed more than a decade ago, and thus may not reflect current prevalence rates. The prevalences and numbers of persons predicted based on such studies are likely to be conservative estimates.

With the forces of globalization and industrialization proceeding at an increasing rate, the prevalence of diabetes is predicted to increase dramatically over the next few decades. The resulting burden of complications and premature mortality will continue to present itself as a major and growing public health problem for most countries.

It is hoped that this report will assist in monitoring the trends of diabetes prevalence over time, by adopting the same methodology for future reports. It should stimulate research in those countries lacking data, as well as encourage further and improved research in those countries where available data may not be representative of national rates.

Finally, this report should act as a stimulus for intervention. Perhaps the most essential aspect of research is the action taken as a result of findings. Diabetes requires culturally appropriate intervention in order to reduce the enormous personal suffering and economic burden that grows with this epidemic.