Background and Objectives:To examine the correlation between the concentration of vitamin D (VD) in venous blood at approximately 4 weeks of age (±4 weeks of age) and neonatal outcomes in preterm infants (birth weight <1500 g or gestational age <32 weeks) in two neonatal intensive care units (NICUs) located in Shenzhen, China. Methods and Study Design: Preterm infants were split into two groups based on their VD concentration at ±4 weeks of age: VD insufficiency (VDI) group (≤20 ng/mL) and VD sufficient (VDS) group (>20 ng/mL). Binary logistic regression analysis was used to examine relationships between outcomes and VDI. Results: Of 230 infants in total, 119 (51.7%) were assigned to the VDI group and 111 to the VDS group (48.3%). No correlation was found between serum VD at ±4 weeks of age and gestational age (p>0.05). The starting point of the two groups for oral VD intake did not differ significantly (p>0.05). At ±4 weeks of age, oral VD dose (P<0.05) was greater in the VDS group. Gestational diabetes mellitus was associated with VDI (OR=1.94, 95% CI 1.01-3.75, p=0.047) after controlling for this risk. Following correction for gestational age and oral VD dosage at ±4 weeks old, VDI was also linked to a significant risk of retinopathy of prematurity (OR=2.00, 95% CI 1.08-3.68, p<0.027). Conclusion: Preterm newborns (gestational age <32 weeks or birth age <1500 g) in NICUs in Shenzhen, China continue to have significantly high VDI. Higher VDI is associated with gestational diabetes mellitus and retinopathy of prematurity.