The study hypothesis was that bronchopulmonary dysplasia (BPD) adversely affects cognitive performance at school age. This prospective cohort study examined three groups of children. Two groups were selected from a reconstructed cohort of very low birth weight (VLBW) infants previously enrolled in a multi-site, randomized, controlled, clinical trial. Group 1(n=311) included infants without BPD and group 2 (n=95) included infants with BPD. A control group of full term infants without BPD (group 3, n=188) was selected from a cohort assembled for a previous program supported by the Robert Wood Johnson Foundation. All of the children were assessed at 8-10 years of age in the following areas: full, verbal, and performance intelligence quotients (IQs), reading grade and age, math grade and age, and the Developmental Test of Visual-Motor Integration (VMI). Possible confounding variables were also recorded. Linear regression models were designed to determine the effect of BPD on the performance measures while controlling for confounding variables. Children in group 3 had the best performance and children in group 2 had the poorest performance on all eight cognitive measures. These differences were significant (p=0.0001) for the full IQ, performance IQ, reading grade and age, and math grade and age. Children in groups 3 and 1 performed better than children in group 2 for the verbal IQ(p=0.0001) and the VMI (p=0.0012), but for these two measures there was no significant difference in performance between children in groups 3 and 1. The 15 confounding variables affected scoring on each of the performance measures to varying degrees. Controlling for the confounding variables showed that levels of maternal education and history of cerebral palsy or hydrocephalus partially explained certain differences in performance between the three groups. In conclusion, the cognitive performance at school age of infants with a history of BPD was generally poorer than the performance of VLBW and full term children without a history of BPD.