Objective: We have investigated the frequency of autistic spectrum symptoms in extremely preterm children at school age.
Method: The EPICure Study is a prospective study of births <26 weeks gestation in the UK and Ireland in 1995. Parents of 219 (71%) 307 survivors at 11 years were asked to complete the social communication questionnaire (SCQ) as part of a comprehensive assessment. SCQ data were obtained for 183 extremely preterm children and 137 classmates born at term.
Results: Extremely preterm children had significantly higher SCQ scores than classmates, indicating a higher level of autistic spectrum symptoms (difference in means 4.6; 95% CI 3.4 to 5.8). 29 (16%) extremely preterm children had scores above the threshold for autistic spectrum disorder (ASD) compared with four (3%) classmates (odds ratio (OR) 6.3; 95% CI 2.2 to 18.3) and 14 (8%) extremely preterm children had scores above the threshold for autism compared with no classmates. Children born ⩽24 weeks were six times more likely to score above the threshold for autism than those born at 25 weeks (OR 1.6 to 22.4) and extremely preterm boys were more likely to score above the threshold for autism than extremely preterm girls (OR 3.4; 95% CI 1.0 to 11.2). Extremely preterm birth remained a significant predictor of autistic spectrum symptoms after adjusting for IQ and sex.
Conclusions: Autistic features are more frequent in extremely preterm children than term peers, independent of cognitive ability and sex. The increased prevalence of autistic spectrum symptoms in this population suggests that early aberration in brain development may be responsible for ASD problems.