P2.F116: Leveraging Hospital Patient Racial and Ethnic Composition to Assess Role of Cultural Familiarity in Asian American, Native Hawaiian & Pacific Islander Very Low Birth Weight Infant Care and Outcomes

Background: Disparities in breastmilk feeding at discharge exist across Asian American, Native Hawaiian & Pacific Islander (AANHPI) Very Low Birth Weight (VLBW: < 1500 grams) infants in California. Cultural familiarity in hospital settings is suggested as one factor promoting health equity; however, there are knowledge gaps on how it influences care and outcomes of AANHPI populations. The volume of patients per hospital may be a good surrogate measure of cultural familiarity. This study's objective is to describe the association between the hospital volume of AANHPI VLBW infants, their care, and outcomes.

Methods: Data were obtained from the California Perinatal Quality Care Collaborative/Vital Statistics data registry, including infants 22-29 weeks estimated gestational age or birth weight 401-1500g, born 2011-2019. We excluded infants who died within 12 hours of life. Race/ethnicity data were obtained from birth certificates. We categorized 142 hospitals into high, middle, and low tertile groups based on their AANHPI VLBW admission volume. The high tertile included hospitals with the highest proportion (25.3%) compared with the middle (11.8%) and low (5.4%) tertiles. Care and outcome measures included survival without major morbidity and Baby-MONITOR quality measure subcomponents: process measures included any breastmilk at discharge, hypothermia on admission, timely eye exam for retinopathy of prematurity, healthcare-associated infection, and use of antenatal steroids; outcome measures included in-hospital survival, chronic lung disease, pneumothorax, and growth velocity. Logistic regression models were fitted to compare the measures between the low and high tertiles. Trend tests were performed to examine the distributions of the measures across tertile groups.

Results: This study included 43,067/48,642 infants/admissions, of which 6,534 (15.2%)/7,217 (14.8%) were AANHPI. Table 1 describes the study population at the hospital level. AANHPI infants were more likely to be treated in non-safety net and urban hospitals. Table 2 shows process and care outcome comparisons across hospital tertiles. Infant process and outcome measures by AANHPI tertile, although statistically significant due to the large sample, showed little clinical significance, except for two measures; for AANHPI infants, receiving care in a high tertile AANHPI NICU resulted in meaningfully increased rates of breastmilk use at discharge [(79.2% (high) vs. 77.0 (middle) vs. 69.6% (low), relative risk (95%CI; low vs. high) [0.88 (0.83, 0.93)], p <.001 for trend)] and high growth velocity [(13.3 g/kg/d (high) vs. 12.8 (middle) vs. 12.8 (low), risk difference (95%CI; low vs. high) [-0.46 (-0.67, -0.25)], p <.001 for trend)].

Conclusion: This hypothesis-generating study demonstrates an association between racial/ethnic composition of patient volume for two feeding-related measures. Future studies should investigate whether patient volume is indeed a good measure for cultural familiarity. Additionally, further analyses on disaggregated AANHPI data adjusting for multi-level confounding variables are needed to further understand the impact of cultural familiarity.