Trends in the dietary patterns of Mexican adults by sociodemographic characteristics

BACKGROUND: Sociodemographic characteristics are associated with the dietary patterns of populations. However, the direction of the association is not consistent among countries: it is contingent on the nutritional transition phase, level of economic development, cultural contexts and both the social and health policies prevailing in each country. The objective of this study was to identify the trends in dietary patterns observed in 2006, 2012 and 2016 among Mexican adults by sociodemographic characteristic. METHODS: To determine and compare dietary patterns, we performed a secondary analysis of dietary and sociodemographic data for adults 20–59 years old. Data were drawn from the 2006 and 2012 National Health and Nutrition Surveys (ENSANUTs) together with the 2016 Half-Way National Health and Nutrition Survey (ENSANUTMC). To estimate the dietary patterns, we used an adapted version of the Healthy Eating Index-2015 (HEI-2015) and a quantile-based regression model to compare the HEI medians by sociodemographic characteristic. RESULTS: From 2006 to 2016, the quality of the diet of Mexican adults scored under 50 points on a scale of 0 to 100, markedly below the maximum scores for the majority of HEI-2015 components. Diet quality varied according to age, sex, socioeconomic status (SES), area (urban/rural) and region of residence, with the highest quality observed among older individuals (within the 40–59 age group), women, people of lower SES and residents of rural areas, particularly in southern Mexico. Although this trend remained constant overall throughout 2006, 2012 and 2016, specific HEI-2015 components showed an opposite trend by sociodemographic strata. CONCLUSION: The diet quality of Mexican adults was suboptimal from 2006 to 2016, with notorious disparities persisting over time among sociodemographic strata. Our results can serve as a basis for formulating recommendations on ways to improve the population diet, where those components diverging the most from adequate scores could be highlighted in public-health messages.