School performance in singletons born after assisted reproductive technology

Abstract

Is school performance in children conceived after assisted reproductive technology (ART) comparable to those conceived after spontaneous conception (SC)? Singleton children born after ART performed better in the crude analysis than singleton children born after SC although after adjustment, small differences were observed in total scores in favour of SC children. While it is well known that ART children, also singletons, have an adverse perinatal outcome, studies on cognitive skills in ART children are inconsistent and only few studies have been published on school performances. Although these studies indicate good school performances in ART children many studies suffer from low participation rate and few participants. This retrospective population-based cohort study included all singleton children in Sweden, born after ART (n = 8323) or SC (n = 1 499 667), between 1985 and 2001. Singleton children born after ART, identified in national IVF registries, were cross-linked with the Medical Birth Registry (MBR), the National Patient Registry (NPR) and the Swedish Cause of Death Registry (CDR) for characteristics and medical outcomes. Data on school performances, parental education and other parental characteristics was obtained through cross-linking to the National School Register and to Statistics Sweden. The control group was identified from the MBR and consisted of all singletons born after SC during the same time period. The primary outcome was school performance after 9 years at primary school and based on a mean total score of 16 subjects (0-320). The secondary outcomes were the mean school grade in specific subjects (mathematics, Swedish, English, physical education), 'qualified to enter secondary school' (i.e. approved in mathematics, Swedish and English) and 'poor school performance' (total score <160). Since the distribution of school grades was skewed, percentiles were used. Simple and multivariable linear regression was used for analysis of percentiles and logistic regression was used for the corresponding analysis of binary outcomes. Adjustments were made for child gender and year of birth, maternal age, parity, maternal smoking, paternal age, parental region of birth, parental education and socioeconomic class. Data on 1 507 990 singletons in the ninth grade and registered by the Swedish School Authority were included. In the crude analysis, mean total scores were significantly higher for ART children (mean total score 230.2 (SD 57.2), corresponding to mean percentiles 60.2 (SD 27.7)), than for their SC counterparts (mean total score of 209.7 (SD 63.9), corresponding to mean percentiles 50.2 (SD 29.0)). However, after adjustments for several confounders, SC children had a significant advantage (adjusted mean difference [percentiles] -0.72, 95% confidence interval [CI] -1.31 to -0.12; P = 0.018). When analysing boys and girls together, no significant difference between children born after ART and children born after SC was found in mathematics, Swedish, English or physical education. Neither was there any significant difference between ART children and SC children in qualifying for secondary school (adjusted odds ratio [AOR] 1.05; 95% CI 0.95-1.17, P = 0.35) or in poor school performance (AOR 0.98; 95% CI, 0.89-1.09, P = 0.73). When cross-linking children with missing data on school performances (2.7% for ART and 2.8% for SC) with the NPR for mental disability, 35% of ART and 34% of SC children with missing data, were registered under such a diagnosis. The main limitation was that test scores were missing in a small percentage in both ART and SC children. Although we were able to crosslink this subpopulation with the NPR using codes for mental disability, such diagnosis only partly explained the missing scores. Other limitations are residual confounding caused by unknown confounders. The findings are generally reassuring and indicate, in the crude analysis, that school performances of ART children compared to children born after SC are better. After adjustment small differences were observed in total scores in favour of SC children. There were no significant differences, when analysing boys and girls together in specific subjects, in secondary school qualification or poor school performance. Financial support was received through Sahlgrenska University Hospital (ALFGBG-70 940), Hjalmar Svensson Research Foundation and Nordforsk, project number 71450. None the authors declare any conflict of interest.