A survey conducted in 1988-1989, in the city of Madras, South India, s
howed that the prevalence of diabetes mellitus in adults was 8.2% and
prevalence of impaired glucose tolerance (IGT) was 8.7%. The present s
urvey was another cross-sectional study conducted 5 years later in the
same urban area to study the temporal changes in the prevalence of di
abetes and IGT. The two sample populations surveyed were similar in ag
e structure and socioeconomic factors. In the second survey in 1994-19
95, a total of 2183 subjects, 1081 men and 1102 women, with a mean age
of 40+/-12 years were tested by an oral glucose tolerance test; fasti
ng and 2-h post-glucose plasma glucose were measured. Anthropometric m
easurements, details of physical activity and clinical history of diab
etes were recorded. Age-standardised prevalence of diabetes had increa
sed to 11.6% from 8.2% in 1989 and IGT was 9.1%, similar to 8.7% in 19
89. Multiple regression analysis showed age, waist:hip ratio, body mas
s index (BMI) and female sex were correlated to diabetes. Family histo
ry of diabetes showed interaction with age and BMI. Prevalence of IGT
correlated to age, BMI and waist:hip ratio. This study highlights the
rising trend in the prevalence of non-insulin-dependent diabetes (NIDD
M) in urban Indians. The persistent high prevalence of IGT may also be
a predictor of a further increase in NIDDM in the future. No signific
ant differences in the anthropometric data were noted in this compared
to the previous study.