Effects of sex, age and BMI on screening tests for impaired glucose tolerance

https://doi.org/10.1016/j.diabres.2003.10.019Get rights and content

Abstract

The discriminating abilities of fasting plasma glucose (FPG) and HbA1c were compared on screening tests for impaired glucose tolerance (IGT) and IGT plus diabetes mellitus by the receiver operating characteristic (ROC) curve analysis. Furthermore, effects of sex, age and BMI were examined on sensitivity and specificity of the optimal cutoff points. This study included 997 subjects who were recruited for 75 g OGTT after the first screening of the Japan Diabetes Prevention Program. According to the 1997 criteria of the American Diabetes Association (ADA), 140 subjects were classified as diabetic and 256 as IGT. The areas under the ROC curves of FPG were significantly larger than those of HbA1c. The optimal cutoff points of FPG were 102 mg/dl for IGT and 105 mg/dl for IGT plus diabetes mellitus. Those of HbA1c were both 5.3%. In screening with FPG, females had significantly lower sensitivity and higher specificity than males, and the specificity for IGT plus diabetes mellitus was the lowest in the obese group. In screening with HbA1c, the specificity was low in the older and the obese groups. We concluded that FPG was superior to HbA1c for screening of IGT and IGT plus diabetes mellitus and the optimal cutoff point of FPG would be 102 mg/dl or greater.

Introduction

The increase in type 2 diabetes mellitus is a major health problem worldwide. It has been estimated that worldwide the number of patients with diabetes will be more than double by the year 2025 [1], and in Japan the number will increase from approximately 6.9 million in 1997 to 10.8 million in 2010 [2].

In Japan, the prevalence of impaired glucose tolerance (IGT) is two to three times that of diabetes mellitus [3]. The higher prevalence of IGT may suggest that patients with diabetes will increase in the future with the change in lifestyle. Actually, it is well known that Japanese immigrants in the US have a higher prevalence of diabetes mellitus than Japanese in Japan [4], [5]. Their lifestyle is more western than Japanese natives despite the same genetic backgrounds. Furthermore, subjects with IGT are at a substantial risk of developing cardiovascular diseases [6]. Japanese community-based cohort studies observed that death from cardiovascular diseases was approximately twice that of people with normal glucose tolerance [7], [8].

IGT can be recognized as a sub-clinical state needing intervention from a public health point of view. Recently, the final reports of primary prevention trials of type 2 diabetes mellitus that targeted subjects with IGT were published in succession [9], [10], [11]. One of the reviews about prevention of diabetes mellitus from these results has regarded subjects with IGT as potential candidates for intervention [12]. Diagnosis of IGT requires oral glucose tolerance test (OGTT). It is inefficient to perform OGTT on numerous people, so various methods were proposed to reduce the number of OGTT for screening of IGT [13], [14]. However, the appropriate method has not been established and the available information about screening for IGT is limited.

In the present study, we evaluated the discriminating abilities of fasting plasma glucose (FPG) and HbA1c by the receiver operating characteristic (ROC) curve analysis [15] for screening of IGT and then examined effects of sex, age and BMI on the screening performance.

Section snippets

Study population

We enrolled 997 subjects (461 males and 536 females, aged 30–59 years, without a previous history of diabetes), who were recruited for OGTT after the first screening of the Japan Diabetes Prevention Program (JDPP). JDPP is a randomized clinical trial designed to assess the efficacy of intensive diet and exercise to prevent or delay the onset of type 2 diabetes mellitus in subjects with IGT. This trial started in April 1999 in collaboration with 27 centers in communities in various districts

Results

The characteristics of the study subjects by glucose tolerance are shown in Table 1. According to the 1997 criteria of ADA, 140 subjects were classified as diabetes (14.0%), 256 as having IGT (25.7%), 87 as having IFG (8.7%) and 514 as having NGT (51.6%). One-way analysis of variance was significant for all variables including age (P<0.0001).

Fig. 1 shows ROC curves of FPG and HbA1c for screening of IGT. Both curves lie above the diagonal line, which means both tests can discriminate between

Discussion

The present results allow some considerations about the proposed screening for IGT.

First, this study showed that the discriminating ability of FPG was superior to that of HbA1c, although each test can discriminate between IGT and non-IGT (NGT plus IFG). This is not due to the difference of CVs between the two parameters because CVs of the both were not so different. Although it is widely accepted that FPG is recommendable as a screening variable for diabetes mellitus [20], it is not always so

Acknowledgements

This study was supported by a Grant for Research on Health Services from the Ministry of Health, Labour and Welfare, Japan. Authors are indebted to the collaborative centers in various districts in Japan.

References (30)

  • M Tominaga et al.

    Impaired glucose tolerance is a risk factor for cardiovascular disease, but not impaired fasting glucose: The Funagata Diabetes Study

    Diab. Care

    (1999)
  • W.C Knowler et al.

    Reduction in the incidence of type 2 diabetes with lifestyle intervention or metformin

    N. Engl. J. Med.

    (2002)
  • J Tuomilehto et al.

    Prevention of type 2 diabetes mellitus by changes in lifestyle among subjects with impaired glucose tolerance

    N. Engl. J. Med.

    (2001)
  • American Diabetes Association and National Institute of Diabetes, Digestive and Kidney Diseases, The prevention or...
  • K.M Narayan et al.

    A two-step strategy for identification of high-risk subjects for a clinical trial of prevention of NIDDM

    Diab. Care

    (1996)
  • Cited by (0)

    View full text