To examine the association between language use – predominantly English, English and Spanish equally and predominantly Spanish – and food insecurity among Hispanic adults residing in the USA, 1999–2018.

Pooled cross-sectional study design.

United States.

15 073 Hispanic adults.

Compared with Hispanic adults who predominantly spoke English and after adjusting for age, sex, family income-to-poverty ratio, education level and employment status, Hispanic adults who spoke English and Spanish equally (OR = 1·28, 95 % CI = 1·05, 1·56) or predominantly Spanish (OR = 1·25, 95 % CI = 1·04, 1·49) had higher odds of food insecurity. After stratifying by country of birth, language use was associated with higher odds of food insecurity only for Hispanic adults born outside of the USA, but not for Hispanic adults born in the USA. Hispanic adults born outside of the USA who spoke English and Spanish equally (OR = 1·27, 95 % CI = 1·04, 1·55) or spoke predominantly Spanish (OR = 1·24, 95 % CI = 1·04, 1·48) had higher odds of food insecurity when compared with those who predominantly spoke English.

Foreign-born Hispanic adults who speak predominantly Spanish, or English and Spanish equally, have higher odds of food insecurity. Food and nutrition assistance programmes that serve Hispanic immigrants should make sure to provide linguistically and culturally appropriate services to this population.

To identify predominant dietary patterns among Hispanic women and to determine whether adherence to dietary patterns is predicted by neighbourhood-level factors: linguistic isolation, poverty rate and the retail food environment.

Cross-sectional analyses of predictors of adherence to dietary patterns identified from principal component analysis of data collected using the Study of Women's Health Across the Nation FFQ. Census data were used to measure poverty rates and the percentage of Spanish-speaking families in the neighbourhood in which no person aged ≥14 years spoke English very well (linguistic isolation) and the retail food environment was measured using business listings data.

New York City.

A total of 345 Hispanic women.

Two major dietary patterns were identified: a healthy dietary pattern loading high for vegetables, legumes, potatoes, fish and other seafood, which explained 17 % of the variance in the FFQ data and an energy-dense dietary pattern loading high for red meat, poultry, pizza, french fries and high-energy drinks, which explained 9 % of the variance in the FFQ data. Adherence to the healthy dietary pattern was positively associated with neighbourhood linguistic isolation and negatively associated with neighbourhood poverty. Presence of more fast-food restaurants per square kilometre in the neighbourhood was significantly associated with lower adherence to the healthy diet. Adherence to the energy-dense dietary pattern was inversely, but not significantly, associated with neighbourhood linguistic isolation.

These results are consistent with the hypothesis that living in immigrant enclaves is associated with healthy dietary patterns among Hispanics.