27298
Child Characteristics Associated with Educational Placement in Children with Autism Spectrum Disorders

Poster Presentation
Thursday, May 10, 2018: 5:30 PM-7:00 PM
Hall Grote Zaal (de Doelen ICC Rotterdam)
J. Christiansen1, S. Parlar2 and L. Pedersen1, (1)Centre for Autism, Herlev, Denmark, (2)Nexus, Copenhagen, Denmark
Background: With the Danish “Inclusion Law” of 2012 school type placement for Danish children with disabilities has been dictated by a quota: 96% of all children should be in regular classes. Prior to the introduction of the law the question of school type placement for children with disabilities was based on individual needs.

Objectives: To examine child characteristics associated with educational placement of Danish children with ASD, prior to 2012, where individual needs were considered.

Methods: Our sample consisted of 44 low functioning (IQ<70) and 95 high functioning (IQ≥70) children with ASD (all assessed at the same site between 1996-2012). 41/44 low functioning children were in special education. Child characteristics for the remaining 95 high functioning children were analyzed with multiple binary logistic regression to determine predictors of school placement. School placement was either regular classes (n=53; mean age 11.5) or special education (special school/class (n=35) or “other” (n=7))(n=42; mean age 11.4). For all 95 children we had information on: gender, IQ, ADOS, ADI-R, co-morbidity, somatic problems, medication and behavioral problems and for 74 children we had information on Vineland (regular classes n=39; special education: n=35). A variable only entered the logistic regression if it differed significantly (t-test or Chi-square test) between the two school type placements.

Results: The logistic model was at chance when predicting whether or not a child was in special education. Nevertheless, knowing that a child had poor reciprocal social interaction (high scores on ADI-R algorithm scale: “reciprocal social interaction”) and the co-morbidity ADHD allowed the logistic model to predict, well above chance, that a child was not in regular class. With these two variables the overall percent correct predictions rose from 56% (percent correct prediction when guessing that all 95 children are in a regular class) to around 70%.

Conclusions: When individual needs are considered almost all low functioning children are in special education. Furthermore, for high functioning children, child characteristics such as ADHD and poor reciprocal social interaction are part of the factors that lead to special education.