The Socioeconomic Disparities in Intakes and Purchases of Less-Healthy Foods and Beverages Have Changed over Time in Urban Mexico

Abstract

Background: To our knowledge, the association between diet and socioeconomic status (SES), using both purchase and intake data, in the Mexican population has not been examined, which is particularly important given the high prevalence of diet-related diseases in Mexico.

Objective: Our objective was to examine the SES-diet relation using household food purchases and individual food intake data.

Methods: We analyzed purchases of packaged food and beverages of 5240 households with the use of the 2012-2014 Nielsen Mexico Consumer Panel Service Dataset, representative of urban areas. Likewise, we examined 9672 individuals over 2 y with food and beverage intake information collected using a single 24-h recall as part of the Mexican National Health and Nutrition Survey 2012. Multivariate linear regression models were conducted to predict per capita daily purchases and intakes of food and beverages classified as healthy and less healthy by SES, and adjusting for sociodemographic variables.

Results: Per capita daily purchases of healthy and less-healthy foods were, on average, 142% and 55% higher in high- than in low-SES households, respectively, from 2012 to 2014 (P < 0.05). Intakes of healthy and less-healthy foods in urban areas were, on average, 7% and 136% higher in high- than in low-SES groups (P < 0.05). Per capita daily purchases of healthy beverages were, on average, 56% higher in high- than in low-SES households from 2012 to 2014 (P < 0.05), whereas purchases of less-healthy beverages were 27% and 17% higher in low- than in high-SES households in 2012 and 2014, respectively (P < 0.05). Per capita daily intake of healthy beverages was 33% higher in high- than in low-SES groups (P < 0.05).

Conclusion: Higher-SES groups from urban areas had greater purchases and intakes of less-healthy foods and healthy beverages. Lower-SES households had greater purchases of less-healthy beverages, but also had the largest reduction in these purchases from 2012 to 2014, which could be associated with the beverage tax implemented in Mexico in 2014.

Keywords: Mexican population; food intake; food purchases; healthy and less-healthy foods; socioeconomic disparities.

FIGURE 1

Daily per capita purchases of healthy (A) and less-healthy (B) foods in urban Mexican households, according to SES, 2012–2014. Values are means ± SEs. n = 959 in the low-, n = 3133 in the medium-, and n = 1721 in the high-SES group in 2012; n = 1094 in the low-, n = 2872 in the medium-, and n = 1809 in the high-SES group in 2013; and n = 1087 in the low-, n = 2815 in the medium-, and n = 1755 in the high-SES group in 2014. Results are derived from the authors’ own analyses and calculations based on data from Nielsen (The Nielsen Company, 2016) through its Mexico Consumer Panel Service for the food and beverage categories from January 2012 to December 2014. Nielsen is not responsible for and had no role in preparing the results reported herein. SES classification is based on the socioeconomic index provided by Nielsen. Multivariate linear regression models were used to predict per capita daily purchases of healthy and less-healthy foods according to SES and year of purchases, adjusting for household composition, region of residence, rate of minimum wage, unemployment rate, and cluster at the household level. Within the SES group, labeled means without a common letter differ among years, P < 0.05, with the use of Bonferroni adjustment. *SES differences in purchases of healthy and less-healthy foods were found at each time point, P < 0.05, with the use of Bonferroni adjustment. SES, socioeconomic status.

FIGURE 2

Daily per capita intakes of healthy (A) and less-healthy foods (B) in the urban Mexican population, categorized according to SES, 2012. Values are means ± SEs. n = 3502 in the low-, n = 3367 in the medium-, and n = 2773 in the high-SES group. Results are derived from the Mexican National Health and Nutrition Survey 2012 (24). Corn tortillas, tortillas, tortilla chips, chocolate, and bread from a bakery were excluded from the healthy and less-healthy food groups. Tertiles of SES are based on the distribution of the socioeconomic index scores computed as part of the Mexican National Health and Nutrition Survey 2012 (range of scores by SES: low, −6.0 to <−0.6; medium, −0.6 to <0.9; high, 0.9 to 4.7) (31). Multivariate linear regression models were used to predict per capita daily intakes of healthy and less-healthy foods according to SES, adjusting for age (categorical), sex, area, and region of residence. Labeled means without a common letter differ among categories of SES, P < 0.05, with the use of Bonferroni adjustment. SES, socioeconomic status.

FIGURE 3

Daily per capita purchases of healthy (A) and less-healthy (B) beverages in urban Mexican households, according to SES, 2012–2014. Values are means ± SEs. n = 959 in the low-, n = 3133 in the medium-, and n = 1721 in the high-SES group in 2012; n = 1094 in the low-, n = 2872 in the medium-, and n = 1809 in the high-SES group in 2013; n = 1087 in the low-, n = 2815 in the medium-, and n = 1755 in the high-SES group in 2014. Results are derived from the authors’ own analyses and calculations based on data from Nielsen through its Mexico Consumer Panel Service for the food and beverage categories from January 2012 to December 2014 (The Nielsen Company, 2016). Nielsen is not responsible for and had no role in preparing the results reported herein. SES classification is based on the socioeconomic index provided by Nielsen. Multivariate linear regression models were used to predict per capita daily purchases of healthy and less-healthy beverages based on SES and year of purchases, adjusting for household composition, region of residence, rate of minimum wage, unemployment rate, and cluster at the household level. Within the SES groups, labeled means without a common letter differ among years, P < 0.05, with the use of Bonferroni adjustment. *SES differences in purchases of healthy and less-healthy beverages were found at each time point, P < 0.05, with the use of Bonferroni adjustment, except for the comparison of per capita daily purchases of less-healthy beverages between low- and medium-SES groups in 2013 (P = 0.27). SES, socioeconomic status.

FIGURE 4

Daily per capita intakes of healthy and less-healthy beverages in the Mexican population from urban areas, categorized by SES. Results are derived from the Mexican National Health and Nutrition Survey 2012 (24). Values are means ± SEs. n = 3502 in the low-, n = 3367 in the medium-, and n = 2773 in the high-SES group. Tertiles of SES are based on the distribution of the socioeconomic index scores computed as part of the Mexican National Health and Nutrition Survey 2012 (range of scores by SES: low, −6.0 to <−0.6; medium, −0.6 to <0.9; high, 0.9 to 4.7) (31). Multivariate linear regression models were used to predict per capita daily intakes of healthy and less-healthy beverages according to SES, adjusting for age (categorical), sex, area, and region of residence. Labeled means without a common letter differ among categories of SES, P < 0.05, with the use of Bonferroni adjustment. SES, socioeconomic status.