Diet does not predict incidence or prevalence of non-insulin-dependent diabetes in Nauruans

Cross-sectional and longitudinal relationships between diet and non-insulin-dependent diabetes (NIDDM) were assessed in Nauruan adults to determine if a particular component of the diet contributed to the high prevalence of NIDDM in this population. In 1982, 24-h dietary recall data were collected from 430 Nauruans over the age of 20, who were participating in a noncommunicable disease (NCD) survey. In 1987 a follow-up survey was performed which included 350 of the subjects from whom dietary data was obtained. Neither cross-sectional nor longitudinal analyses showed any statistically significant associations between any of the specific dietary components studied and NIDDM prevalence or incidence. However, when nutrient intakes were adjusted for energy intake it appeared that the age- and body-mass-index (BMI)-corrected mean intakes of total fat, total carbohydrates, alcohol, sugar and monounsaturated fat were slightly higher in the seven incident cases than in those who remained healthy, while intakes of protein, fibre and cholesterol were lower. Despite the inability to demonstrate an association between NIDDM risk and nutrient intake at the individual level, Nauruans as a population have total energy intakes 115-135% greater than recommended for maintenance of healthy weight, protein intakes about 250% of that required, sugar intakes about twic 1000 e the recommended, fibre intakes only about 30% of current recommended levels and in men a mean alcohol intake more than three times the recommended level. This adverse diet undoubtedly contributes to the high prevalence of obesity in the population and hence, even if there are no direct dietary effects, to the risk of NIDDM and other diet-related diseases.