Effect of past and concurrent body mass index on prevalence of glucose intolerance and type 2 (non-insulin-dependent) diabetes and on insulin response. The Israel study of glucose intolerance, obesity and hypertension.

A representative sample (n = 2140) of the Israeli Jewish population aged 40-70 (excluding known diabetic patients), whose body mass index had been measured 10 years earlier, underwent an oral glucose tolerance test and redetermination of body mass index. Irrespective of weight changes, high concurrent and high past body mass index values (greater than or equal to 27) were associated with similarly increased rates of glucose intolerance as compared with body mass index values less than 27 at both time-points (rate ratio 1.76, 90% confidence limits 1.56-1.99). Glucose intolerance here includes borderline and impaired tolerance as well as Type 2 diabetes. The rate of Type 2 diabetes increased only with increasing past body mass index, while concurrent body mass index had no effect [rate ratios: 2.36 (1.48-3.75) and 1.99 (1.48-2.68) respectively for the medium-(23-26.9) versus-low (less than 23) and high- (greater than or equal to 27) versus-medium past body-mass-index categories]. Weight reduction was associated with only slightly reduced rate of glucose intolerance and had no effect on the rate of diabetes. Mean sum insulin (summed 1 and 2 h levels, mU/l) increased significantly with increasing concurrent body mass index (123, 150 and 190 in the low, medium and high categories) with no effect of past body mass index. It also increased significantly (p less than 0.001) in all concurrent body mass index categories from normal tolerance through borderline to impaired tolerance, and decreased significantly (p less than 0.001) in diabetes relative to impaired tolerance, although it remained above normal.(ABSTRACT TRUNCATED AT 250 WORDS)