a1 Department of Medicine, University of California San Francisco, 505 Parnassus Avenue, Room M-987, San Francisco, CA 94131, USA
a2 Division of General Internal Medicine, San Francisco General Hospital, San Francisco, CA, USA
a3 Department of Public Health and Policy, London School of Hygiene & Tropical Medicine, London, UK
a4 Department of Sociology, Cambridge University, Cambridge, UK
a5 Division of Noncommunicable Diseases & Health Promotion, WHO Regional Office for Europe, Copenhagen, Denmark
Objective Ageing and urbanization leading to sedentary lifestyles have been the major explanations proposed for a dramatic rise in diabetes worldwide and have been the variables used to predict future diabetes rates. However, a transition to Western diets has been suggested as an alternative driver. We sought to determine what socio-economic and dietary factors are the most significant population-level contributors to diabetes prevalence rates internationally.
Design Multivariate regression models were used to study how market sizes of major food products (sugars, cereals, vegetable oils, meats, total joules) corresponded to diabetes prevalence, incorporating lagged and cumulative effects. The underlying social determinants of food market sizes and diabetes prevalence rates were also studied, including ageing, income, urbanization, overweight prevalence and imports of foodstuffs.
Setting Data were obtained from 173 countries.
Subjects Population-based survey recipients were the basis for diabetes prevalence and food market data.
Results We found that increased income tends to increase overall food market size among low- and middle-income countries, but the level of food importation significantly shifts the content of markets such that a greater proportion of available joules is composed of sugar and related sweeteners. Sugar exposure statistically explained why urbanization and income have been correlated with diabetes rates.
Conclusions Current diabetes projection methods may estimate future diabetes rates poorly if they fail to incorporate the impact of nutritional factors. Imported sugars deserve further investigation as a potential population-level driver of global diabetes.
(Received January 17 2012)
(Revised April 13 2012)
(Accepted May 02 2012)
(Online publication June 13 2012)