Despite strong prevention efforts and advances in neonatal care in recent decades, low birth weight remains a serious public health problem in the United States, and survivors remain at increased risk for lifelong problems including cognitive deficits. Current regional and local strategies for referral often rely on variable thresholds for birth weight and gestational age that may be poor analogues to cognitive risk. Improving early referral criteria offers many benefits, including improved cognitive outcomes for children and improved cost-effectiveness and resource utilization in resource-limited communities. We hypothesized that social determinants measurable at birth or at birth hospital discharge, when combined with birth weight and gestational age, would offer an improvement over birth weight and gestational age alone in predicting cognitive test scores in school-aged children with low birth weight. We conducted a secondary analysis using a birth cohort of children from the Fragile Families and Child Wellbeing Study. We created a panel of maternal, familial, and community-level social determinant indicators from the data and examined associations with cognitive measures assessed at age of 9 years. The final social determinant model was statistically significant and explained 35% of the total variance in composite test scores. The "standard care" model (birth weight and gestational age) only explained 9% of the variance. Assessment of social determinants may offer improvement over traditional referral criteria to identify children most at risk of cognitive deficits after low birth weight.