Cognitive development in children up to age 11 years born after ART-a longitudinal cohort study.

study questionHow does the cognitive development of children conceived after ART (IVF and ICSI) - measured as cognitive skills at age 3, 5, 7 and 11 years - differ over time from those born after natural conception (NC)? SUMMARY ANSWER: Improved measures of cognitive development up to age 5 years were recorded in children conceived with ART compared to NC, which attenuates by 11 years, with ART children still scoring slightly better than NC children. WHAT IS KNOWN ALREADY: Results on the cognitive outcomes of children conceived after ART have been highly contradictory. Some have shown that ART children have an impaired behavioural, socio-emotional and cognitive development and higher risk of mental disorders. Others have reported no increased risk or difference. Cognitive development has not been previously examined using latent growth curve models from ages 3 to 11 years, also including appropriate attention to confounding parental characteristics. STUDY DESIGN, SIZE, DURATION: Longitudinal data for the first five waves (2000-2012) of the UK Millennium Cohort Study were used, which is a two-stage sample of all infants born in 2000-2001 and resident in the UK at 9 months of age, drawn from the Department of Social Security Child Benefit Registers. A final sample of N = 15 218 children (125 IVF and 61 ICSI), from 14 816 families was used. Information was available for all waves for 8298 children. Four additional follow-up surveys were conducted in 2003, 2005, 2007 and 2012. PARTICIPANTS/MATERIALS, SETTING,

methodsOur sample includes children born within a union (married or cohabiting parents) and where information on cognitive scores was available for at least two measurement points. Cognitive development was assessed with the British Ability Scales. At age 3 and 5 years (wave 2 and 3), children completed the naming vocabulary component, which measures expressive verbal ability. At age 7 years (wave 4), verbal cognitive abilities were assessed through the word reading test, and at age 11 years (wave 5) through a verbal similarity test. Two-tailed Student's t-tests examined differences between ART and NC groups. Growth curve models (random-coefficient, latent trajectory models) were used to study the effect of ART, confounding parental characteristics and health outcomes at birth, both at a baseline level of cognitive ability at age 3 years and on its growth rate. MAIN RESULTS AND THE ROLE OF CHANCE: At age 3 and 5 years, children conceived with the aid of ART have higher verbal cognitive abilities than NC children (P < 0.001) but this consistently decreases over time and diminishes by age 11 years. Parental environment and resources are pivotal in children's cognitive development. LIMITATIONS, REASON FOR CAUTION: The sample size of the ART cohort of children is small across each time period (N = 150-180) in comparison with NC children (N = 10 496-11 955). Owing to a limited sample size, we are also unable to compare IVF versus ICSI treatment. WIDER IMPLICATIONS OF THE

findingsWith the increasing use of IVF and ICSI, these results indicate that there are no detrimental effects on children's early cognitive outcomes up to age 11 years, and highlight the importance of parental characteristics. STUDY FUNDING/COMPETING INTEREST(S): Funding for this project was provided by the European Union's Seventh Framework Program (FP7 2007-2013) (No. 320116 Families and Societies), ESRC/NCRM SOCGEN Grant (ES/N0011856/1) and the SOCIOGENOME ERC Consolidator Grant (ERC-2013-CoG-615603) (to M.C.M.). The authors have no competing interests to declare. TRIAL REGISTRATION NUMBER: N/A.