Time-Trends in Food Insecurity Among US-Born and Foreign-Born Hispanic Adults by Language Use: An Analysis of National Health and Nutrition Examination Survey Data, 1999-2018

Background: Historically, food insecurity prevalence has been higher in Hispanic households than in non-Hispanic White households. Food insecurity prevalence among Hispanic adults, US-born and foreign-born, may vary by language use. Objective: To explore whether or not the relationship between language use and food insecurity varied over time (1999-2018) among US-born and foreign-born Hispanic adults. Design: Trends analysis and multivariable logistic regression modeling using pooled cross-sectional data. Participants and setting: Fifteen thousand sixty-two Hispanic adults participating in the United States National Health and Nutrition Examination Survey (1999-2018). Main outcome measures: Food insecurity prevalence, assessed with the US Household Food Security Survey Module. Statistical analysis: Unadjusted food insecurity trends from 1999 to 2018 by language use (mostly English, both languages equally, or mostly Spanish) among US-born and foreign-born Hispanic adults were analyzed using piecewise-linear regression of log prevalence rates. Multivariable logistic regression models adjusted for sociodemographic characteristics and with an interaction term between language use and time were used to determine if odds of food insecurity among US-born and foreign-born Hispanic adults varied by language use between 1999 and 2018. Results: Hispanic adults’ food insecurity prevalence followed an upward linear trend from 1999 to 2018; this was significant for US-born mostly English-speakers (P < 0.001), US-born mostly Spanish-speakers (P = 0.013), and foreign-born mostly Spanish-speakers (P < 0.001). In fully adjusted logistic regression models, foreign-born Hispanic adults who spoke both languages equally (odds ratio 1.8, 95% CI 1.2 to 2.6) and those who spoke mostly Spanish (odds ratio 1.9, 95% CI 1.4 to 2.8) had significantly higher food insecurity odds, compared with mostly English-speakers. No variations in associations across time were observed between language use and food insecurity (interaction P value > 0.1). Conclusions: Hispanic adults’ unadjusted food insecurity trends from 1999 to 2018 varied by language use. When adjusted for sociodemographic characteristics and compared with mostly English-speakers, food insecurity odds were significantly higher only among foreign-born Hispanic adults who spoke either both languages equally or mostly Spanish. Food assistance programs should linguistically adapt their services for Hispanic adults.