Diet Quality and Glycemic Control in Patients with Type 2 Diabetes

Background: The overall diet quality of individuals and populations can be assessed by dietary indexes based on information from food surveys. Few studies have evaluated the diet quality of individuals with type 2 diabetes or its potential associations with glycemic control.

Objective: To evaluate the relationship between diet quality and glycemic control.

Design: Cross-sectional study with consecutive enrollment from 2013 to 2016.

Participants: Outpatients with type 2 diabetes treated at a university hospital in southern Brazil.

Main Outcome Measures: Dietary information was obtained by a quantitative food frequency questionnaire validated for patients with diabetes. Overall diet quality was evaluated by the Healthy Eating Index 2010. Glycemic control was assessed by fasting plasma glucose and glycated hemoglobin.

Statistical Analyses: A receiver operating characteristic curve was constructed to find the optimal Healthy Eating Index cutoff point to discriminate diet quality, considering good glycemic control as glycated hemoglobin level <7%. Patients were then classified as having lower vs higher diet quality, and the two groups were compared statistically. Logistic regression models were constructed with glycated hemoglobin level ≥7% as the dependent variable, adjusted for age, current smoking, diabetes duration and treatment, physical activity, body mass index, high-density lipoprotein cholesterol level, and energy intake.

Results: A total of 229 patients with type 2 diabetes (median age=63.0 years [interquartile range=58.0 to 68.5 years]; diabetes duration=10.0 years [interquartile range=5 to 19 years]; body mass index 30.8±4.3; and glycated hemoglobin=8.1% [interquartile range=6.9% to 9.7%]) were evaluated. A Healthy Eating Index score >65% yielded the best properties (area under the receiver operator characteristic curve=0.60; sensitivity=71.2%; specificity=52.1%; P=0.018). Patients with lower-quality diets were younger and more likely to be current smokers than patients with higher-quality diets. After adjusting for confounders, patients with lower-quality diets had nearly threefold odds of poorer glycemic control (2.92; 95% CI 1.27 to 6.71; P=0.012) than those in the higher-quality diet group.

Conclusions: Lower diet quality, defined as an Healthy Eating Index 2010 score <65%, was associated with poor glycemic control in this sample of outpatients with type 2 diabetes.