Background: Subnormal head circumference (HC) has been associated with poor neurologic and developmental outcomes. Aim: To examine the correlates and consequences of subnormal HC in a cohort of school age, very low birth weight (VLBW, < 1500 g) children. Study design and outcome measures: We examined developmental outcomes at a mean age of 6.8 years in a cohort of 128 VLBW children born from 1982- 1986 and 58 normal birth weight controls. The VLBW cohort included a regional sample of < 750 g birth weight children and matched children with 750- 1499 g birth weight. Outcomes included an IQ equivalent, neuropsychological skills, academic achievement, adaptive behavior, and attention problems. HC was defined along a continuum and as subnormal vs. normal. Linear and logistic regressions were employed to determine the effects of HC on the outcomes after controlling for confounding variables. Results: Thirty one VLBW children (24% ) had subnormal HC vs. none of the controls. The VLBW children with subnormal HC differed significantly from the normal HC group in birth weight (748 g vs. 977 g, p < 0.001), SGA status (52% vs. 27% , p < 0.05), high neonatal risk (57% vs 29% , P < 0.05), and neurosensory impairment (23% vs 8% , P < 0.05). Even after taking these risk factors into account, subnormal HC was associated with poorer IQ equivalent, perceptual motor skills, academic achievement, and adaptive behavior. Results were similar after excluding the children with neurosensory impairment. Conclusions: Subnormal HC is associated with adverse developmental outcomes in VLBW children, independent of other risk factors. Interventions to improve antenatal and postnatal growth may contribute to better school- age outcomes. Background: Subnormal head circumference (HC) has been associated with poor neurologic and developmental outcomes. Aim: To examine the correlates and consequences of subnormal HC in a cohort of school age, very low birth weight (VLBW, <1500 g) children. Study design and outcome measures: We examined developmental outcomes at a mean age of 6.8 years in a cohort of 128 VLBW children born from 1982- 1986 and 58 normal birth weight controls. The VLBW cohort included a regional sample of <750 g birth weight children and matched children with 750- 1499 g birth weight. Outcomes included an IQ equivalent, neuropsychological skills, academic achievement, adaptive behavior, and attention problems. HC was defined along a continuum and as subnormal vs. normal. Linear and logistic regressions were employed to determine the effects of HC on the outcomes after controlling for confounding variables. Results: Thirty one VLBW children (24%) had subnormal HC vs. none of the controls. The VLBW children with subnormal HC differed significantly from the normal HC group in birth weight (748 g vs. 977 g, pO.001), SGA status (52% vs. 27%, p <0.05), high neonatal risk (57% vs 29% , P <0.05), and neurosensory impairment (23% vs 8%, P <0.05). Even after taking these risk factors into account, subnormal HC was associated with poorer IQ equivalent, perceptual motor skills, academic achievement, and adaptive behavior. Results were similar after excluding the children with neurosensory impairment. Conclusions: Subnormal HC is associated with adverse developmental outcomes in VLBW children, independent of other risk factors. Interventions to improve antenatal and postnatal growth may contribute to better school-age outcomes.