Comparison of the prevalence in two diabetes surveys in Pu-Li, Taiwan, 1987–1988 and 1991–1992

The objective of this study was to investigate the prevalence of non-insulin-dependent diabetes mellitus (NIDDM) in Pu-Li, Taiwan from 1991-1992, and to compare the results with a similar study conducted in 1987-1988. We also wished to compare different approaches in asking about patient history and to determine how this effects data authenticity. Both were community-based cross-sectional studies with stratified cluster sampling of residents age > or = 30. Blood samples were taken for screening and 75 g oral glucose tolerance tests were performed for diagnosis. The total number of eligible subjects in the second study was 2719 (1424 men, 1295 women). Complete data and samples were collected for 1118 (536 men, 582 women). The response rate was 41.1% (37.6% for men, 44.9% for women). The crude prevalence was 10.3% (5.6% known, 4.7% new). Using standard world population (Segi), the age-adjusted prevalence rate was 8.3% (4.0% known, 4.3% new). The 1991-1992 study had a response rate (crude 41.1%, adjusted 51.3%) which was slightly lower than the 1987-1988 study (crude 44.8%, adjusted 55.9%). The age-adjusted prevalence rates for new NIDDM were similar (4.4 vs. 4.3%) while the age-adjusted prevalence of known NIDDM in the second survey (4.0%) was lower than the first survey (6.9%), which apparently was overestimated due to the simplicity of questions regarding history. In conclusion, prevalence of new DM in this area appears to be stable, and when doing a survey regarding previous DM, it is better to include treatment history rather than depending on self-reporting of NIDDM alone.