摘要：The objective of this paper is to describe a method of estimating the prevalence of non-insulin dependent diabetes mellitus (NIDDM) in a community. We first counted subjects with previously diagnosed NIDDM and those newly diagnosed by use of the 75-gram oral glucose tolerance test (OGTT). We then used the OGTT data to predict the prevalence of NIDDM at different levels of fasting plasm glucose (FPG) among the subjects who did not undergo the OGTT. Data were derived from 3, 614 residents aged 20 years and older in a rural area where population-based screening for NIDDM was performed using FPG or post-prandial plasma glucose(PG). Of the 318 participants who underwent an OGTT in 1992, 35 (11.0%) were judged to have NIDDM by the 2-hour PG criterion. We applied the above-described method to the 1, 179 subjects who had an FPG but failed to undergo the OGTT. The estimated prevalence of NIDDM, which included the previously diagnosed group, the newly diagnosed group and the expected group, was calculated. The estimated prevalence of diabetes in this population aged 20 years and older turned out to be 6.1% (95% Cl: 5.3-6.9). As a validation study, the same method was applied to 220 subjects who did not undergo the OGTT in 1992 but did so in 1994. The actual prevalence of NIDDM among these subjects was compared to the prevalence expected from the 1992 data by our method. The actual value was within the 95% confidence interval of the expected value. J Epidemiol , 1996 ; 6 : 114-119.
