Diet Quality among the Brazilian Population and Associated Socioeconomic and Demographic Factors: Analysis from the National Dietary Survey 2008-2009.

BACKGROUND: Evaluating diet quality is a way to monitor a population's adherence to dietary guidelines. OBJECTIVE: To adapt the Healthy Eating Index-2015 (HEI-2015) to the Brazilian population, to evaluate the validity and reliability of the adapted HEI-2015, and to assess diet quality of Brazilian adults as well as socioeconomic and demographic factors associated with diet quality. DESIGN: In this cross-sectional study, two dietary records from nonconsecutive days were used to obtain food intake information. The collected socioeconomic and demographic data included sex, age, education, per capita income, and residence area. PARTICIPANTS/SETTING: A representative sample (n=27,760) of the Brazilian adult population participated in the Nutrition Dietary Survey 2008-2009. MAIN OUTCOME MEASURES: Total and component scores for the adapted HEI-2015. The validity and reliability of the index were tested. STATISTICAL ANALYSIS PERFORMED: Mean total and component scores were estimated for the adapted HEI-2015. Also, the proportion of subjects that achieved the maximum score for each component was calculated. Regarding validity and reliability analysis, principal components analysis examined the number of dimensions; Pearson correlations were estimated between total score, components, and energy, and Cronbach's coefficient α was estimated. Diet quality was compared among socioeconomic and demographic categories. RESULTS: The mean total score for the adapted HEI-2015 was 45.7 (95% CI: 45.4 to 46.0). Women had higher diet quality scores (46.4; 95% CI: 46.1 to 46.7) than men (44.9; 95% CI: 44.6 to 45.3). Age, education, and per capita income were directly associated with the adapted HEI-2015 total score. Greater than or equal to 50% of subjects received the maximum adapted HEI-2015 component scores for "total protein foods," "seafood and plant proteins," "fatty acids," and "sodium." In contrast, less than 30% of subjects received the maximum component scores for "total vegetables," "dairy," and "saturated fats." Higher percentages of women received the maximum component scores for "total fruits," "whole fruits," "total vegetables," "greens," "dairy," and "sodium," whereas higher percentages of men received the maximum component scores for "total protein foods," "seafood and plant proteins," "fatty acids," "added sugars," and "saturated fats" components. Finally, validity analysis revealed weak correlations between component scores and energy and weak to moderate correlations with total scores; six dimensions were responsible for the total variance in diet quality and the standardized Cronbach's coefficient α was .65 (unstandardized=.64). CONCLUSIONS: Brazilian adults have suboptimal diet quality as assessed by the adapted HEI-2015. Diet quality varied by socioeconomic and demographic factors. Results support the validity and the reliability of the index.