Abstract
This study was performed on 2, 562 inhabitants (999 male and 1, 563 female subjects) in a rural area in western Hiroshima prefecture who underwent on oral glucose tolerance test in a mass health examination for adult diseases. World Health Organization (WHO) criteria and American Diabetes Association (ADA) criteria on fasting plasma glucose levels were used to classify the individuals. The observed prevalences using the ADA criteria were 4.3% in males and 2.8% in females, and were lower than the prevalences with the WHO criteria (7.8% in males and 5.8% in females). The sensitivity and specificity of the ADA criteria to identify cases diagnosed by the WHO criteria were 40.5% and 99.2%, respectively. The sensitivity classified by obesity was the highest in the moderately obese group (55.3%). It was 36.6% in the normal group and 30.8% in the obese group. When classified by body mass index and age, the sensitivity was low in the elderly subjects (70 years old and older) regardless of the BMI. The results suggest that subjects with glucose intolerance among obese and/or elderly subjects may be excluded or neglected in the diagnosis of diabetes when the ADA criteria are used.
