Background: The longitudinal impact of fetal growth on attention problems in males and females is unclear. This study aims to evaluate the impact of fetal growth assessed by neonatal anthropometry throughout childhood and adolescence in males and females separately. Methods: We compared neonatal anthropometry (birth weight (BW), head circumference (HC), proportion of optimal birthweight (POBW)) and asymmetry (head-to-abdominal circumference ratio (HC/AC) and ponderal index (PI)) at birth with parental assessment of the child behavior checklist attention-problem syndrome (CBCL-AP) raw score measured at ages five, eight, 10, 14 and 17. We used univariable and multivariable linear mixed-effects modeling. Sensitivity analyses included excluding pre-term births, teacher ratings and treating the CBCL-AP as an ordinal variable. Results: In males, a 1-SD lower BW, increased CBCL-AP by 0.234 (95%CI [-0.422, -0.0497]). In males, a 1-SD lower HC increased CBCL-AP by 0.316 (95%CI [0.495, 0.133]). In males, there was a U-shaped relationship between HC/AC and CBCL-AP throughout childhood and adolescence; a curvilinear relationship was observed between POBW and CBCL-AP. In females, a 1 SD lower HC increased CBCL-AP 0.424 (95%CI [0.726, 0.133]), but every increased year of age reduced the effect by 0.027 (95% CI: 0.006-0.05). In females, there was no clear relationship between BW, POBW or HC/AC and CBCL-AP. In males and females, PI was not significantly associated with CBCL-AP. The exclusion of pre-term births and analysis of teacher-rated attention problems was consistent with the primary results. Conclusions: Using a longitudinal design, our study suggests a male vulnerability to attention problems throughout childhood and adolescence from neonatal anthropometry. The relationships in females appear to be limited to childhood.