Associations between Diet Behaviors and Measures of Glycemia, in Clinical Setting, in Obese Adolescents

Abstract

Objective: To determine the influence of dietary behaviors, assessed in a clinical setting, on measures of glycemia in overweight and obese adolescents.

Study design: The study is a retrospective, cross-sectional chart review. Eligible participants were overweight youth (N = 146, age 9-21 years) who attended the Youth Diabetes Prevention Clinic in Indianapolis, IN. Glycemic status was assessed during a 2-hour oral glucose tolerance test (OGTT). In the Bright Futures Questionnaire, a recommended clinical tool for assessing unhealthy behaviors in youth, nutrition-specific questions were modified to quantify dietary habits. Associations between dietary habits and measures of glycemia were determined using multiple linear regression models. Skewed data are presented as geometric means and 95% confidence intervals.

Results: Of the 146 adolescents who were assessed [60% girls, age 13.7 years (13.3, 14.0), BMI 33.9 kg/m(2) (33.3, 34.5)], 40% were diagnosed with prediabetes. Higher intake of dessert foods was associated with increased glucose levels at 2 hours following the OGTT (β = 0.23, p = 0.004), and higher intake of packaged snack foods was associated with elevated levels of hemoglobin A1c (β = 0.04, p = 0.04), independent of adiposity.

Conclusions: In obese youth, high intakes of dessert and packaged snack items were associated with elevated concentrations of glucose at 2 hours following the OGTT and hemoglobin A1c. Findings demonstrate the usefulness of a modified Bright Futures Questionnaire, used in a clinical setting, for identifying dietary behaviors associated with hyperglycemia in obese adolescents. ClinicalTrials.gov registration number: NCT02535169.

Conflict of interest statement

Author Disclosure Statement No competing financial interests exist.

Figures

Figure 1.

OGTT glucose concentrations among adolescents who reported varied frequencies of dessert servings per day. Values are means and confidence intervals. Means without a common letter are significantly different (p = 0.02, a total of 141 subjects). p-Values are based on Tukey's honest significant difference (HSD) post hoc test. OGTT, oral glucose tolerance test.