Two urbanized Micronesian populations were recently studied by population-based diabetes surveys. These were Nauruans living on the island of Nauru, and Gilbertese resident on the islet of Betio, in the Republic of Kiribati (1982 and 1981, respectively). Nauruans are known to suffer from a very high prevalence of non-insulin-dependent (type II) diabetes mellitus (NIDDM). In the present study, the effects of suspected environmental risk factors for diabetes were controlled for, in an attempt to elucidate any residual difference in the prevalence of diabetes between the two groups, which might be of genetic origin. As almost all Nauruans lead a physically inactive lifestyle, only inactive subjects in either population were selected for study. The total study sample consisted of 2306 subjects. After further controlling for the effects of age and obesity, the odds of diabetes for Nauruans, as compared with Gilbertese, were threefold. The multiple logistic regression model showed ethnicity (i.e., being Nauruan) to be the strongest of the predictor variables examined in both sexes. In a random subsample of approximately one-third of the total subjects (N = 694), stratified with respect to age and sex, daily intake of total energy and of three dietary components was assessed. The dietary variables were carbohydrate, fat, and dietary fiber. Fat intake was found to be a weak, but significant predictor of diabetes in females after controlling for age, although dietary fat was not predictive of diabetes after also controlling for ethnicity, or for body mass. None of the other dietary variables had any predictive power in either sex.(ABSTRACT TRUNCATED AT 250 WORDS)
Download full-text PDF |
Source |
---|---|
http://dx.doi.org/10.2337/diab.33.10.1002 | DOI Listing |
Enter search terms and have AI summaries delivered each week - change queries or unsubscribe any time!