Twins can be used to investigate the biological basis for observed associations between birth weight and later disease risk, as they experience in utero growth restriction compared with singletons, which can differ in magnitude within twin pairs despite partial or total genetic identity. In the present study, sixty monozygotic and seventy-one dizygotic same-sex twin pairs aged 19–50 years and eighty-nine singleton controls matched for age, gestational age, sex, maternal age and parity were recruited from an obstetric database. Associations between fasting lipid levels and birth weight were assessed by linear regression with adjustment for possible confounding factors. Twins were significantly lighter at birth but were not significantly different in adult height, weight or lipid levels from the singleton controls. There was a significant inverse association between birth weight and both total and LDL-cholesterol levels among singleton controls (−0·53mmol/l per kg (95% CI −0·97, −0·09), P=0·02 and −0·39mmol/l per kg (95% CI −0·76, −0·02), P=0·04, respectively), but there was no significant association between birth weight and lipid levels in either unpaired or within-pair analysis of twins. The results suggest that the in utero growth restriction and early catch-up growth experienced by twins does not increase the risk of an atherogenic lipid profile in adult life.