91 - Non-English Language Preference and Human Milk at Discharge in VLBW infants in California

Publication Number: 3087.91

Laura A. Rose, University of California, San Diego School of Medicine, La Jolla, CA, United States; Annalisa Watson, Stanford University School of Medicine, Mill Valley, CA, United States; Tianyao Lu, Stanford University School of Medicine, Alamo, CA, United States; Mihoko Bennett, Stanford University, Palo Alto, CA, United States; Jochen Profit, Stanford University, Palo Alto, CA, United States; Henry Lee, University of California San Diego, La Jolla, CA, United States

Background: The provision of human milk has benefits for very low birth weight (VLBW; ≤1,500 grams) infants. In California, Hispanic infants have the highest odds of human milk at discharge. In evaluating language and human milk, non-English speaking parents had lower odds of human milk at discharge. This study examines the impact of language on human milk provision.

Objective: To examine how preferred language impacts the rate of human milk at home discharge among VLBW infants born in California.

Design/Methods: We analyzed data from all VLBW infants within the California Perinatal Quality Care Collaborative (CPQCC) network born 2022-2023 (10,542 infants). We excluded infants that died in delivery room (n=647), those still hospitalized after one year (n=53), and those missing characteristics (n=127). We tested for associations with receiving any human milk at discharge using multivariable logistic regression models. The cohort was stratified into subgroups to examine Spanish versus English language preference amongst Hispanic families, and other languages versus English in non-Hispanic families.

Results: According to 2022-2023 CPQCC data including all patients, 87% of families identify English as their preferred language, 10% identify Spanish, and 3% identify another language. An interpreter was required for 85% of all non-English languages. Among all patients, non-English language preference was more common in infants who received prenatal care, were transferred from another hospital, were lower gestational age (GA), were small for GA, had congenital malformation, developed NEC, and had longer stays.
After multivariable logistic regression in the VLBW cohort, human milk at discharge was associated with prenatal care, higher GA, and higher five-minute Apgar score (see Table 1). These factors remained significant within the Spanish versus English preference Hispanic VLBW population (n=5202).
Among VLBW Hispanic infants, those whose families preferred Spanish had similar adjusted odds of receiving human milk at discharge compared to those with English language preference (aOR 0.99 [0.84-1.16]). Among VLBW non-Hispanic infants, those with English language preference had higher adjusted odds of receiving human milk at discharge compared to those with other (non-English, non-Spanish) language preferences (aOR 1.43 [1.07-1.92]).

Conclusion(s): Among the VLBW infant population in CPQCC hospitals, the rate of human milk at discharge was similar for Spanish-speaking and English-speaking Hispanic families. The rate of human milk at discharge was lower for non-Hispanic patients who spoke languages other than English or Spanish.

Table 1: Multivariable Logistic Regression Model for Human Milk Nutrition at Home Discharge among VLBW Infants
Variables associated with provision of human milk at discharge for very low birth weight (VLBW) infants included prenatal care, higher gestational age, and higher five minute Apgar score.