Date of Award
Doctor of Philosophy (PhD)
Patricia A. Cowan, Ph.D.
Margaret T. Hartig, Ph.D. Donna K. Hathaway, Ph.D. Matthew W. Strum, Pharm.D. Mona N. Wicks, Ph.D.
Although the American Diabetes Association (ADA) adopted the use of the glycated hemoglobin (A1C) test as a method of diabetes and prediabetes diagnosis, the ADA has not developed firm guidelines concerning the use of the A1C test in children and adolescents, as research has not validated thresholds in this group. Diabetes and prediabetes are diseases influenced by multiple factors, including race and ethnicity, age, vitamin D deficiency, and body mass index (BMI). The purpose of this study was to determine the sensitivity, specificity, positive predictive value (PPV), and negative predictive value (NPV) of the A1C test compared to the gold standard use of the fasting plasma glucose (FPG) and 2-hour oral glucose tolerance test (OGTT) to detect diabetes and prediabetes in a children and adolescents considered to be at higher risk for impaired glucose metabolism. In addition, ROC curve analysis was performed to determine optimal thresholds for the diagnosis of prediabetes in available groups of the research sample. The study also to examined the correlation between A1C and race and ethnicity, age, vitamin D levels, and body mass index, in addition to comparing the relationship of A1C to beta cell dysfunction and insulin sensitivity. A retrospective review of 902 patient electronic medical records in an urban endocrinology clinic was conducted. Based on FPG and 2-hr glucose during the OGTT, patients were classified based on the ADA 2014 criteria as having diabetes or prediabetes Subjects ranged in age from 2-18 (11.6 ± 3.32), were predominantly minority (70.7% African American, 17.3% Hispanic, 12.0% Caucasian) and female (60.7%). The results yielded a high specificity (99.7%) and high negative predictive value (99.9%) for the whole sample, although the results were lower for the African American group. The results also yielded a low specificity (35.3%) but a high negative predictive value (99.8%) for the entire sample. Although results were once again lower for the African American subset. ROC curve analysis for prediabetes yielded a threshold of 5.8% for sample. Multiple regression found some correlation between fasting glucose and A1C, although statistical analysis was not possible for the aggregate sample. No statistically significant association was found between the A1C and age, vitamin D, and BMI in the sample. Correlation analysis found stronger associations between the A1C and beta cell dysfunction versus insulin sensitivity. In this predominantly minority population A1C had a high specificity and sensitivity for the diagnosis of diabetes. While the A1C resulted in a high number of false positives for prediabetes, A1C <5.7% accurately identified individuals with normal glucose tolerance. Children and adolescents considered to be at higher risk for impaired glucose metabolism (family history of diabetes, obesity, minorities, or history of gestational diabetes) with A1C ≥5.7% or with symptoms of diabetes should undergo OGTT testing. In addition, different threshold levels for racial and ethnic groups should be considered in the diagnosis of prediabetes.
Hitt, Jennifer McGuire , "Hemoglobin A1C and the Diagnosis of Diabetes and Prediabetes in Children and Adolescents" (2014). Theses and Dissertations (ETD). Paper 124. http://dx.doi.org/10.21007/etd.cghs.2014.0139.