RESUMO
AIM: To evaluate the association of Spanish compared to English primary household language on preterm (PT) infants' language outcomes in the United States and to examine associations with socio-economic factors. METHODS: This was a retrospective cohort of PT infants born <32 weeks gestation from Spanish-speaking (n = 95) and English-speaking homes (n = 1030) born 2005-2019. Language (primary outcome) and cognitive and motor skills (secondary outcomes) were measured using the BSID-III at 18-24 months corrected age. Group differences were evaluated using bivariate comparisons and logistic regression analyses. RESULTS: Mothers reporting Spanish-speaking homes had higher rates of public insurance and lower educational achievement. Group newborn characteristics were similar. Preterm infants from Spanish-speaking homes had significantly lower BSID-III language composite, cognitive composite, receptive and expressive scores compared to infants from English-speaking homes. Logistic regression modelling identified independent negative effects of Spanish-speaking household OR 3.26 (CI 1.89-5.62) and public insurance OR 2.31 (CI 1.71-3.12) with a protective benefit derived from breast milk OR 0.68 (CI 0.50-0.92) when adjusting for medical morbidities, socio-economic factors and gestational age. CONCLUSION: Public health policies and interventions in the United States should target language and cognitive outcomes of PT infants from Spanish-speaking homes.