The Risk Score In Screening For Type 2 Diabetes Mellitus In Chinese Adult Population And The Optimal HbAlc Cutoff Point For Diagnosing Diabetes

BackgroundDiabetes mellitus (DM) was defined as "a metabolic disorder of multiple aetiology characterized by chronic hyperglycaemia with disturbances of carbohydrate, fat and protein metabolism resulting from defects in insulin secretion, insulin action, or both". Based on the clinical stages and aetiological types, there are four types of diabetes, type 1, type 2, pregrance and others and type 2 diabetes accounts for 90-95% of DM.If it is not well treated, diabetes might result in a long-term damage, dysfunction, and failure of various organs, especially the eyes, kidneys, nerves, heart, etc. DM has becomes the third chronic disease following tumor and cardiovasular disease.During the past few decades, the number of patients with diabetes mellitus has been seen a dramatic increase in China. In 2008, the prevlance of DM in Chinese adult population(above 20 years old) was more than doubled compared with 1996. And the epidemic of diabetes casts a heavy burden to the famlity and society. According to the WHO estimate, China will lose the national income about＄555.7 billion in the next ten years. In a word, DM has become a seriously public health problem in China.In the past few decades, numerous studies had been performed to investigate the risk factors and prevention of diabetes. It is generally agreed that type 2 diabetes is a multi-etiological disorder, resulted from the genetic background, environmental and social factors such as physical inactivity, unhealthy dietary habits and obesity. As well as education level and socio-economic position. The results of Daqing in China, DPP in Finland and DPS in USA even suggested that type 2 diabetes can be prevented or delayed by intervention such as proper diet, activity and lifestyle.The typical symptoms for diabetes are increased thirst, increased hunger, frequent urination, unexplained weight loss, fatigue, etc. But, these symptoms are usually not present in all patients with type 2 diabetes, which makes the diseases undiagnosed for a long time. So it is important to find a sensitive and acceptable screening method for earlier identification and interventions. The screening tools for diabetes include blood glucose, haemoglobin A1e (HbA1c) and urinary glucose test and risk assessment questionnaires or scores.Urinry glucose test has been rarely used in epidemiological studies of diabetes because of its low sensitivity. Neither blood glucose nor HbA1c test is suitable for general investigation because of their high cost, time-consuming, invasive injury and not easy to be accepted, even with high sensitivity and specificity. Four risk score without biochemical test about diabetes have been developed to serve as a screening tool for diabetes. Most of these risk scores were developed in western countries. Based on this point, Qingdao Diabetes Prevention Project developed a simple screening risk score for diabetes even though it has not been widely used in China.In 2010, American Diabetes Association(ADA) recommanded that haemoglobin Alc (HbAlc) test had been adopted as a diagnostic criterion for diabetes, the optimal HbAlc cutoff points was≥6.50%. Because of ethic variation of HbAlc, ADA suggested that every country should look for the optimal HbAlc cutoff points based on population.A study on early measurement of diabetes using a simple method will supply insights into the intervention of diabetes, and the study of optimal HbAlc cutoff points will provide new starting points for HbA1c diagnosis of diabetes in China.Objectives1 To determine the secular trend in prevalence of diabetes in Chinese adults in Qingdao, China.2 To evaluate the performance of fasting capillary plasma glucose (FCG), HbA1c,and risk score as screening tests for detecting diabetes and impaired fasting glucose/impaired glucose tolerance(IFG/IGT).3 To compare the performance of the Qingdao diabetes risk score with other risk score in western countries.4 To compare the prediction of FCG, HbA1C, and risk score as screening tests for detecting diabetes and IFG/IGT.5 To investigate the optimal HbA1c cutoff point of diabetes diagnosis in China. 6 To compare the different risk factors of diabetes diagnosed by glucose and HbA1c,.Methods1 Study populationA total of 6100 individuals aged 35years old were randomly slected and 5355 individuals took part in the survey, giving a response rate of 87.8%.2 MethodsData was collected by questionnaire interview (demographic characteristics.familly history of diabetes, excise, diet and depressive investigation), anthropomelric measurements (height,weight,waist, heart pulse and blood pressure) and laboratory measmements,such as fasting plasma glucose(FPG), total cholesterol(TC) triglyeride(TG), high density lipoprotein cholesterol(HDL-C), low-density lipoprotein cholesterol (LDL-C), C-reactive protein (CRP), gamma-glutamyltransferase (GGT).3 Classification of diabetesDiabetes was defined according to the 2006 World Health Organization (WHO)/ International Diabetes Federation (IDF) criteria and International Expert Committee Report on the role of the HbAlc assay for diabetes. Subjects who reported a history of diabetes and who were under treatment with either insulin or oral anti-diabetic agents were considered as previously diagnosed diabetes, regardless of their fasting plasma glucose levels. Diagnosed diabetes was excluded from the data analysis to reduce the potential effluence of the anti-diabetic medications and the diabetic complications. Newly diagnosed diabetes was defined if FPG level of≥7.0 mmol/1 and/or 2hPG level of≥11.1 mmol/1 regardless of the HbAlc concentration. Alternatively, newly diagnosed diabetes was determined if the calculated HbAlc> 6.5% in spite of the glucose levels. Individuals who had a FPG of 6.1-6.9mmol/l and/or 2hPG of 7.8-11.0 mmol/1 were defined as impaired fasting glucose (IFG) and/or impaired glucose tolerance (IGT). Both IFG and IGT were considered as pre-diabetes in the current study. Waist circumference≥85 cm for Chinese men and≥80 cm for Chinese women was defined as central obesity according to the 2003 definition of metabolic syndrome in China.4 Statistical analysesA chi-square test (χ2)for categorical variables and the general linear model (GLM) procedure for continuous variables were used to compare differences in prevalence and in age-adjusted means among different glucose categories.Receiver operating characteristic curve (ROC) was plotted for the risk score, FCG and HbAlc against the presence of the newly diagnosed diabetes. The optimal cutoff point of HbAlc was identified as this point that the sum of the sensitivity and the specificity is maximum.To compare the different risk factors of diabetes diagnosed by glucose and HbA1c,.,Stepwise logistic regression analysis was performed to estimate the odds ratio (OR) and 95% confidence intervals (CI). Age, education level, alcohol-drinking, smoking, family history of diabetes, BMI, systolic blood pressure (SBP), LDL-C, TG, TC, CRP and GGT were fitted in the multivariate model.C-statistics were used to compare the areas under the receiver operating characteristic curves (AUC) using STATA 8.2 (Stata Corporation, College Station, Texas, USA). All analyses were performed using SPSS (Version15.0; SPSS Inc, Chicago, IL, USA). A p-value less than 0.05 (two tailed) was considered statistically significant.Results1 Prevalance of diabetes and IFG/IGTThe age-standardized prevalence of diabetes and IFG/IGT were 16.0% and 27.8% in the population aged 35 to 74 years, respectively, and the prevalence of diabetes (χ2=159.51, P=0.00) and IFG/IGT (χ2=104.90, P=0.00) raised with aging; and higher in the urban participants than in the rural (χ2=22.76, P=0.00;χ2=35.87, P=0.00). The prevalence did not differ between men and women (χ2=2.24,P=0.134;χ2=1.27, P=0.258).2 FCG, HbAlc and risk score screening for type 2 diabetes and IFG/IGTThe Aear Under Curve (AUC)of FCG in screening for newly diagnosed diabetes was 0.77 (0.73-0.82) in men and0.75 (0.73-0.77) in women for FCG. They were 0.58 (0.53-0.62) in men and 0.58 (0.54-0.63) in women in screening for IFG/IGT. The AUC of HbA1c in screening newly diagnosed diabetes was 0.67 (0.63-0.72) in men and 0.67 (0.63-0.71) in women for HbA1c. They were 0.51(0.47-0.55) in men and 0.52 (0.49-0.55) in women in screening for IFG/IGT.The AUC of risk score in screening for newly diagnosed diabetes was 0.68 (0.64-0.72) in men and 0.69 (0.67-0.71) in women for risk score. They were 0.59 (0.56-0.64) in men and 0.60 (0.57-0.63) in women in screening for IFG/IGT.The AUC was lower for HbA1c than for FCG for detecting newly diagnosed diabetes (P< 0.01, in men; P< 0.01, in women) and the AUC of risk score in screening newly diagnosed diabetes was slightly lower than that of the FCG test (P< 0.01 in men; P< 0.01 in women) but not different from that of the HbA1c (P> 0.05 in men; P> 0.05 in women).3 Comparison of the Chinese risk score with other existing onesWe applied also other previously risk scores which were derived from other population to our study population of the 2006 survey. Four such risk scores (Danish Risk Score, Indian Risk Score, Rotterdam study Risk Score and Thai Risk Score) were applicable to the data and thus, validated. The performance of these scores was poorer in Chinese population of the 2006 survey than in their original populations. None of the existing risk scores had a significantly larger AUC than the simple risk score developed in this study.4 The optimal cutoff point of HbA1C in diabetes diagnosisA HbA1c level of 6.5% recommended by the International Expect Committee has a sensitivity and specificity of 33.5% and 87.8% among the participants, and the optimal HbA1C cutoff point for newly diagnosed diabetes in this study population at 5.6% was lower than the recommended value of 6.5%. By taking a cutoff point of 5.6% with the sensitivity of 72.9% and specificity of 61.5% in men,61.8% and 67.8% in women.5 To compare the different risk factors of diabetes diagnosed by glucose and HbA1CThe risk factors of diabetes diagnosed by glucose in men and women was aging, obesity, family history of diabetes, GGT, TG; the protective factors was education in men was income in women. As for as diabetes diagnosed by HbA1c, the protective factors was as same as diabetes diagnosed by glucose, but the risk factors was obesity in women not in men.Conclusions1 The prevalence of type 2 diabetes and IFG/IGT in Chinese was high and increasing in Qingdao city. It is urgent to take some measures to prevent the epidemic of diabetes.2 The performance of risk score comprising modifiable lifestyle factors is well, and it is a good alternative screening test in screening programs covering large population. It can also serve as a useful health promotion tool in a prevention program.3 None of the existing risk scores in western countries had a significantly larger AUC and simple than Qingdao risk score developed in this study,.4 The optimal HbA1c cutoff point of diabetes diagnosis in China was 5.60%.5 The difference in risk factors of diabetes diagnosed by glucose and HbA1c suggested the phenotypes of diabetes defined by the two glycemic criteria differ.