A simple tool detected diabetes and prediabetes in rural Chinese

J Clin Epidemiol. 2010 Sep;63(9):1030-5. doi: 10.1016/j.jclinepi.2009.11.012. Epub 2010 Mar 1.

Abstract

Objective: To develop and evaluate a simple tool, using data collected in a rural Chinese general practice, to identify those at high risk of Type 2 diabetes (T2DM) and prediabetes (PDM).

Study design and setting: A total of 2,261 rural Chinese participants without known diabetes were used to derive and validate the models of T2DM and T2DM plus PDM. Logistic regression and classification tree analysis were used to build models.

Results: The significant risk factors included in the logistic regression method were age, body mass index, waist/hip ratio (WHR), duration of hypertension, family history of diabetes, and history of hypertension for T2DM and T2DM plus PDM. In the classification tree analysis, WHR and duration of hypertension were the most important determining factors in the T2DM and T2DM plus PDM model. The sensitivity, specificity, positive predictive value, negative predictive value, and receiver operating characteristic area for detecting T2DM were 74.6%, 71.6%, 23.6%, 96.0%, and 0.731, respectively. For PDM plus T2DM, the results were 65.3%, 72.5%, 33.2%, 90.7%, and 0.689, respectively.

Conclusion: The classification tree model is a simple and accurate tool to identify those at high risk of T2DM and PDM. Central obesity strongly associates with T2DM in rural Chinese.

Publication types

  • Evaluation Study
  • Research Support, Non-U.S. Gov't

MeSH terms

  • Asian Continental Ancestry Group / statistics & numerical data*
  • Body Mass Index
  • Diabetes Mellitus, Type 2 / classification
  • Diabetes Mellitus, Type 2 / diagnosis*
  • Diabetes Mellitus, Type 2 / epidemiology
  • Female
  • Humans
  • Hypertension / diagnosis*
  • Hypertension / epidemiology
  • Male
  • Middle Aged
  • Obesity, Abdominal / diagnosis
  • Prediabetic State / classification
  • Prediabetic State / diagnosis*
  • Prediabetic State / epidemiology
  • Predictive Value of Tests
  • Rural Health
  • Surveys and Questionnaires
  • Waist-Hip Ratio / statistics & numerical data