Limits...
Optimal Hemoglobin A1c Levels for Screening of Diabetes and Prediabetes in the Japanese Population.

Shimodaira M, Okaniwa S, Hanyu N, Nakayama T - J Diabetes Res (2015)

Bottom Line: A total of 1372 individuals without known diabetes were selected for this study.A 75 g oral glucose tolerance test (OGTT) was used to diagnose diabetes and prediabetes.Thus, HbA1c may be inadequate as a screening tool for prediabetes.

View Article: PubMed Central - PubMed

Affiliation: Department of Internal Medicine, Iida Municipal Hospital, 438 Yawata-machi, Iida, Nagano 395-8502, Japan.

ABSTRACT
The aim of this study was to evaluate the utility of hemoglobin A1c (HbA1c) to identify individuals with diabetes and prediabetes in the Japanese population. A total of 1372 individuals without known diabetes were selected for this study. A 75 g oral glucose tolerance test (OGTT) was used to diagnose diabetes and prediabetes. The ability of HbA1c to detect diabetes and prediabetes was investigated using receiver operating characteristic (ROC) analysis. The kappa (κ) coefficient was used to test the agreement between HbA1c categorization and OGTT-based diagnosis. ROC analysis demonstrated that HbA1c was a good test to identify diabetes and prediabetes, with areas under the curve of 0.918 and 0.714, respectively. Optimal HbA1c cutoffs for diagnosing diabetes and prediabetes were 6.0% (sensitivity 83.7%, specificity 87.6%) and 5.7% (sensitivity 60.6%, specificity 72.1%), respectively, although the cutoff for prediabetes showed low accuracy (67.6%) and a high false-negative rate (39.4%). Agreement between HbA1c categorization and OGTT-based diagnosis was low in diabetes (κ = 0.399) and prediabetes (κ = 0.324). In Japanese subjects, the HbA1c cutoff of 6.0% had appropriate sensitivity and specificity for diabetes screening, whereas the cutoff of 5.7% had modest sensitivity and specificity in identifying prediabetes. Thus, HbA1c may be inadequate as a screening tool for prediabetes.

No MeSH data available.


Related in: MedlinePlus

ROC curve analysis for the ability of HbA1c to predict prediabetes defined by OGTT values.
© Copyright Policy - open-access
Related In: Results  -  Collection


getmorefigures.php?uid=PMC4465763&req=5

fig2: ROC curve analysis for the ability of HbA1c to predict prediabetes defined by OGTT values.

Mentions: AUCs of ROC analysis of HbA1c for the diagnosis of diabetes and prediabetes were 0.918 (95% CI, 0.879–0.958) and 0.714 (95% CI, 0.685–0.743), respectively (Figures 1 and 2). The sensitivity decreased, whereas the specificity increased as the cutoff levels increased. In the analysis stratified by age, the AUCs of ROC were approximately 0.9 for diabetes and 0.7 for prediabetes (Table 2). The optimal HbA1c cutoff levels as identified by the maximal Youden index were 6.0% for diabetes and 5.7% for prediabetes. In diabetes, the HbA1c cutoff of 6.0% showed high sensitivity (83.7%) and specificity (87.6%), with a low proportion (16.3%) of false-negative results in disease identification (Table 3). In contrast, in prediabetes, the HbA1c cutoff of 5.7% had modest sensitivity (60.6%) and specificity (72.1%), with a high proportion (39.4%) of false-negative results (Table 4). At these HbA1c cutoffs (6.0% for diabetes and 5.7% for prediabetes), accuracy was high (87.4%) for diabetes and low (67.6%) for prediabetes; however, the agreement between HbA1c categorization and OGTT-based diagnoses was low in both diabetes (κ = 0.399) and prediabetes (κ = 0.324).


Optimal Hemoglobin A1c Levels for Screening of Diabetes and Prediabetes in the Japanese Population.

Shimodaira M, Okaniwa S, Hanyu N, Nakayama T - J Diabetes Res (2015)

ROC curve analysis for the ability of HbA1c to predict prediabetes defined by OGTT values.
© Copyright Policy - open-access
Related In: Results  -  Collection

Show All Figures
getmorefigures.php?uid=PMC4465763&req=5

fig2: ROC curve analysis for the ability of HbA1c to predict prediabetes defined by OGTT values.
Mentions: AUCs of ROC analysis of HbA1c for the diagnosis of diabetes and prediabetes were 0.918 (95% CI, 0.879–0.958) and 0.714 (95% CI, 0.685–0.743), respectively (Figures 1 and 2). The sensitivity decreased, whereas the specificity increased as the cutoff levels increased. In the analysis stratified by age, the AUCs of ROC were approximately 0.9 for diabetes and 0.7 for prediabetes (Table 2). The optimal HbA1c cutoff levels as identified by the maximal Youden index were 6.0% for diabetes and 5.7% for prediabetes. In diabetes, the HbA1c cutoff of 6.0% showed high sensitivity (83.7%) and specificity (87.6%), with a low proportion (16.3%) of false-negative results in disease identification (Table 3). In contrast, in prediabetes, the HbA1c cutoff of 5.7% had modest sensitivity (60.6%) and specificity (72.1%), with a high proportion (39.4%) of false-negative results (Table 4). At these HbA1c cutoffs (6.0% for diabetes and 5.7% for prediabetes), accuracy was high (87.4%) for diabetes and low (67.6%) for prediabetes; however, the agreement between HbA1c categorization and OGTT-based diagnoses was low in both diabetes (κ = 0.399) and prediabetes (κ = 0.324).

Bottom Line: A total of 1372 individuals without known diabetes were selected for this study.A 75 g oral glucose tolerance test (OGTT) was used to diagnose diabetes and prediabetes.Thus, HbA1c may be inadequate as a screening tool for prediabetes.

View Article: PubMed Central - PubMed

Affiliation: Department of Internal Medicine, Iida Municipal Hospital, 438 Yawata-machi, Iida, Nagano 395-8502, Japan.

ABSTRACT
The aim of this study was to evaluate the utility of hemoglobin A1c (HbA1c) to identify individuals with diabetes and prediabetes in the Japanese population. A total of 1372 individuals without known diabetes were selected for this study. A 75 g oral glucose tolerance test (OGTT) was used to diagnose diabetes and prediabetes. The ability of HbA1c to detect diabetes and prediabetes was investigated using receiver operating characteristic (ROC) analysis. The kappa (κ) coefficient was used to test the agreement between HbA1c categorization and OGTT-based diagnosis. ROC analysis demonstrated that HbA1c was a good test to identify diabetes and prediabetes, with areas under the curve of 0.918 and 0.714, respectively. Optimal HbA1c cutoffs for diagnosing diabetes and prediabetes were 6.0% (sensitivity 83.7%, specificity 87.6%) and 5.7% (sensitivity 60.6%, specificity 72.1%), respectively, although the cutoff for prediabetes showed low accuracy (67.6%) and a high false-negative rate (39.4%). Agreement between HbA1c categorization and OGTT-based diagnosis was low in diabetes (κ = 0.399) and prediabetes (κ = 0.324). In Japanese subjects, the HbA1c cutoff of 6.0% had appropriate sensitivity and specificity for diabetes screening, whereas the cutoff of 5.7% had modest sensitivity and specificity in identifying prediabetes. Thus, HbA1c may be inadequate as a screening tool for prediabetes.

No MeSH data available.


Related in: MedlinePlus