Proportion of type 2 diabetes cases resulting from impaired fetal growth.

OBJECTIVE During the past decade, several researchers have demonstrated a higher risk of type 2 diabetes in relation to lower birth weight. This theory, referred to as the "thrifty phenotype" hypothesis, postulates that impaired fetal growth predisposes individuals to the development of diabetes and other metabolic abnormalities. This article examines the importance of fetal growth in the etiology of diabetes by estimating the proportion of diabetes cases associated with this exposure. RESEARCH DESIGN AND METHODS The importance of an exposure or its correlate as a potential cause of a disease can be assessed by estimating the proportion of cases that could be prevented if the exposure was eliminated from a defined population. This proportion is referred to as the population-attributable fraction (PAF). Published studies of the association between diabetes and birth weight were reviewed and selected for further analysis if data were presented that enabled PAF calculation. In addition, PAFs were calculated for higher birth weight cutoffs because researchers have postulated that the lowest birth weight category may not capture all cases of fetal growth retardation. Studies have shown that exposure classified in this broader manner can produce unbiased PAF estimates, even if many subjects are falsely classified as exposed. RESULTS PAFs for the lowest birth weight category ranged from 0.01 to 0.25. In this analysis, PAFs for diabetes did not exceed 0.35. In contrast, >50% of diabetes cases in the First National Health and Nutrition Examination Survey Epidemiologic Follow-Up Study were attributable to excess adiposity as reflected by a BMI of > or =26 kg/m2 (PAF >0.50). CONCLUSIONS Impaired fetal growth or its correlates account for a minority (<0.50) of type 2 diabetes cases.