Preliminary Data for the CAST Test in Three Spanish Population-Based Birth Cohorts At 5 Years of Age, in the INMA (INfancia y Medio Ambiente [Environment and Childhood]) Project

Background: The Childhood Asperger Syndrome Test (CAST) is a 37-item parental self-completion questionnaire designed to screen high-functioning autism spectrum in children 4 to 11 years (Scott, Baron-Cohen, Bolton, & Brayne, 2002). In a population-based school age study established in UK, the authors found a significant gender difference, where boys had a higher CAST score (Mean 5; IQR: 3,8) than girls (Mean 4; IQR: 2,6) (p-value <0.001) (Williams et al., 2008). There are few population-based school age studies that have use CAST and to our knowledge only one in Spanish population, in Gipuzkoa (Fuentes et al.,IMFAR 2010).

Objectives:  To describe the preliminary results of CAST in three Spanish population-based cohorts; including the relation of sociodemographic variables and the multivariable analysis of longitudinal cognitive and clinical test scores.

Methods: The INMA (INfancia y Medio Ambiente [Environment and Childhood]) Project (http://www.proyectoinma.org/) is a network of seven prospective population-based birth cohorts in Spain, which aims to study the role of environmental pollutants (air, water and diet) during pregnancy and childhood in relation to child growth and development.  We used CAST Spanish translated version by Fuentes, J., and Pezzuto, C., retrieved in 2009, from http://www.autismresearchcentre.com/arc_testsin 412 children at 5 years old (Sabadell cohort). We also used Bayley Scales of Infant Development (BSID) at 14 months, and McCarthy Scales of Children Abilities (MSCA) at 5 years to assess global cognitive development. Finally, the diagnostic criteria of DSM-IV for Attention Deficit Hyperactivity Disorder (ADHD) were applied to evaluate the ADHD symptomatology. 

Results:  In our population, the CAST mean score was 5.16 (SD 3.07). We found significantly higher mean scores in males [5.68 (SD 3.48)] compared with females [4.6 (SD 2.45)] (p-value <0.001). The children of mothers of lower socioeconomic class or mothers with lower education level had significantly higher CAST scores (higher symptomatology) than those children of mothers of higher socioeconomical class or university level education. In multivariate models, adjusted for child's sex, maternal social class and maternal education, children whose mothers scored higher in the global severity index of SCL-90-R, defined as worse maternal mental health score, were more likely to have higher CAST score [IRR 1.30; p-value=0.028]. According to the DSM-IV, children with higher score in the symptomatology of ADHD combined criteria, were more likely to have higher score at CAST [IRR 1.59; p-value= 0.006]. We found 4 children who exceeded the clinical cut-off of CAST (0.97%), 3 of them were boys and had similar profile with low cognitive score at BSID at 14 month, high cognitive score at MSCA and high score in ADHD symptoms.

Conclusions: These preliminary results on the CAST show consistency with a previous population study in UK and Spain. The study offers the second Spanish population-based data in 5 year-old children for the CAST test, including sociodemographic and mental health factors, longitudinal cognitive development and co-morbidity with ADHD symptoms. Further data for two INMA cohorts will be available by December this year, adding the Gipuzkoa and Valencia Cohorts to these data (with an estimated total sample of 1200 children).