Abstract
Importance Previous studies have found that children born with a non-syndromic form of cleft lip and/or palate have lower-than-average educational attainment. These differences could be due to a genetic predisposition to low intelligence and academic performance, factors arising due to the cleft phenotype (such as school absence, social stigmatization and impaired speech and language development), or confounding by the prenatal environment. If the association is not explained by a genetic predisposition, interventions and policies to improve educational attainment in individuals born with a cleft could have wide-ranging knock-on effects on their quality of life.
Objective To assess evidence for the hypothesis that genetic liability to non-syndromic cleft lip with or without cleft palate (nsCL/P) influences educational attainment.
Methods Using summary data from genome-wide association studies (GWAS), we performed Linkage Disequilibrium (LD)-score regression and two-sample Mendelian randomization to evaluate the relationship between genetic liability to nsCL/P (GWAS n=3,987) and educational attainment (GWAS n=766,345), and intelligence (GWAS n=257,828).
Results There was little evidence for shared genetic aetiology between nsCL/P and educational attainment (rg −0.03, 95% CI −0.14 to 0.08, P 0.58; βMR 0.002, 95% CI −0.001 to 0.005, P 0.417) or intelligence (rg −0.01, 95% CI −0.12 to 0.10, P 0.85; βMR 0.002, 95% CI −0.010 to 0.014, P 0.669).
Conclusions and relevance Individuals born with nsCL/P are unlikely to be genetically predisposed to low educational attainment or intelligence. Therefore, clinical-, school-, social- and family-level interventions to improve educational attainment could be effective.
Evidence before this study Previous studies have found that children born with a cleft lip/palate tend to have lower-than-average educational attainment, even in the absence of other birth defects or known syndromes. These differences could be due to a genetic predisposition to low intelligence caused by undiagnosed congenital differences in brain structure or function. Alternatively, these differences might be explained by downstream factors related to having a cleft, such as higher school absences to attend healthcare appointments, social stigmatization, or impaired speech and language development. The differences might also be explained by confounding factors such as family socioeconomic position or parental health behaviours (for example smoking or drinking alcohol). None of these proposed explanations has been investigated in detail and it is currently unclear why individuals born with a cleft tend to do less well at school.
Added value of this study Using a causal inference technique called Mendelian randomization, this study suggests that any association between non-syndromic cleft lip/palate and lower educational attainment is unlikely to be caused by a genetic predisposition to both cleft and low intelligence or academic achievement.
Implications of all the available evidence This study provides reassurance that non-syndromic children born with cleft lip/palate are unlikely to be genetically predisposed to do poorly at school and have lower intelligence. Therefore, interventions and policies to improve educational attainment in children born with a cleft could be effective. The target of these interventions will depend on further research to understand the causes of lower educational attainment in this group. This highlights the need for a largescale longitudinal patient cohort study combining genetic data with detailed information on clinical, social, environmental and developmental factors. The Cleft Collective Cohort Study has been established to address this need.
Funding acknowledgements
The Medical Research Council (MRC) and the University of Bristol support the MRC Integrative Epidemiology Unit [MC_UU_12013/1, MC_UU_12013/9, MC_UU_00011/1, MC_UU_00011/5]. The Scar Free Foundation supports the Cleft Collective (REC approval 13/SW/0064). The Economics and Social Research Council (ESRC) support NMD via a Future Research Leaders grant [ES/N000757/1]. CD is funded by the Wellcome Trust [108902/B/15/Z]. No funding body has influenced data collection, analysis or its interpretation. This publication is the work of the authors, who serve as the guarantors for the contents of this paper. This work was carried out using the computational facilities of the Advanced Computing Research Centre - http://www.bris.ac.uk/acrc/ and the Research Data Storage Facility of the University of Bristol - http://www.bris.ac.uk/acrc/storage/.