糖尿病不論在全世界及台灣都已經是嚴重的公共衛生問題，雖然在台灣已有關於糖尿病發生率的報導，但是幾乎沒有關於糖尿病前期的探討，且缺乏對於糖尿病前期篩檢工具的評估，因此本研究主要的研究目的有二，第一為估算糖尿病前期的發生率；第二為評估2004年美國糖尿病協會發展的糖尿病前期篩檢工具 (ADART) 預測糖尿病前期發生，並比較和評估ADART及過去其他研究所發展的篩檢工具。
1021位沒有空腹血糖異常或糖尿病的居民追蹤3年後，184位新診斷個案發展為糖尿病前期或糖尿病，累積3年糖尿病前期或糖尿病的粗發生率為18.02% % (95% CI: 15.64-20.36)，在經過調整年齡和性別，累積3年糖尿病前期或糖尿病的發生率為17.83% (95% CI: 15.41-20.24)。
使用羅吉斯迴歸分別比較ADART (模型一)，ADART加上生活型態 (模型二)，及ADART加上生活型態和生理生化值檢驗 (模型三) 三個模型的接收者操作特徵曲線下面積。模型一的曲線下面積男性為0.60 (95% CI 0.54-0.66)，女性為0.72 (95% CI 0.66-0.77)；模型二男性為0.62 (95% CI 0.56-0.68)，女性為0.74 (95% CI 0.68-0.80)；模型三男性為0.64 (95% CI 0.58-0.71)，女性為0.75 (95% CI 0.69-0.80)。不管在男性還是女性三個模型都沒有統計上的顯著差異，顯示ADART用在篩檢女性40歲以上社區居民是否為糖尿病前期是個不錯的篩檢工具。
Diabetes is an important worldwide public health problem. The disease has also become one of most serious health conditions in Taiwan. There were limited studies on diabetes incidence in Taiwan, none of previous studies in Taiwan have reported the incidence rate of pre-diabetes, and the performance of screening tools for identifying new cases of pre-diabetes or diabetes has not been evaluated. Therefore, there are two aims in this study. One is to estimate the incidence of pre-diabetes or type 2 diabetes in Taiwan, and the other is to evaluate the performance of American Diabetes Association Risk Tool (ADART) proposed by American Diabetes Association (ADA) and other instruments published in literature in Taiwan.
This study followed for 3 years a random sample of 1021 residents with normal glycemia and without diabetic medication at baseline. New cases of hyperglycemia (Fasting plasma glucose, (FPG)≥100 mg/dl or medication) and type 2 diabetes (FPG≥126 mg/dl or medication) were ascertained from health check-up and laboratory examinations. Three multivariate logistic regression models were used, considering ADART only, ADART plus lifestyle behaviors, and ADART plus lifestyle behaviors and biomarkers. We also compared the ability of ADART with the other instruments published in literature in screening undiagnosed pre-diabetes or diabetes. The areas under curves (AUC) of ROCs were calculated to compare their relative ability.
Overall, 184 new cases of pre-diabetes and diabetes were identified after a mean follow-up period of 3 years, the age- and gender- weighted cumulative incidence was 17.83 (95% CI: 15.41-20.24) per 100.
The AUCs of model 1 for males and females were 0.60 (95% CI 0.54-0.66) and 0.72 (95% CI 0.66-0.77); of model 2 were 0.62 (95% CI 0.56-0.68) and 0.74 (95% CI 0.68-0.80); of model 3 were 0.64 (95% CI 0.58-0.71) and 0.75 (95% CI 0.69-0.80). The AUCs of model 2 and model 3 were not significantly different from that of model 1 (p = 0.317and 0.106 in males, respectively; p = 0.213 and 0.086 in females, respectively). Conclusion, this study demonstrates that ADART is a good screening instrument for predicting the 3-year incidence of hyperglycemia for woman aged forty years and over in Taiwan.
