Insistence on Sameness in Parents and Their Children with ASD

Background: Previous research has established the existence of the broader autism phenotype (BAP) (e.g., Piven et al., 1997), but work is emerging on its relationship with specific child ASD features. Insistence on sameness (IS) may be a heritable feature of ASD due to its relative independence from mediating variables, such as verbal IQ. It has been suggested that subtyping based on degree of IS to create more homogeneous groupings of individuals may increase the ability to detect genetic factors (Hus et al., 2007).

Objectives:  Characterize IS in parents of children with ASD and examine its relationship with child IS. We hypothesized a positive association between parent and proband IS.

Methods: Participants were 2760 families who participated in the Simons Simplex Collection. Parent IS was measured using the Broader Autism Phenotype Questionnaire (BAPQ). A factor analysis was conducted to create an IS scale from BAPQ items. BAPQ-IS scores were divided into groups of the lowest and highest 25% and middle 50% (25-50-25). Proband IS was operationalized using subsets of items from the ADI-R and the RBS-R. Linear regression models were used to examine the relationship between parent IS and proband IS. Potential predictors were entered into a structured hierarchical regression model, in which Block 1 included proband verbal IQ, sex, and age (which are known to affect the expression of ASD and other psychological symptoms), and Block 2 included father and mother BAPQ-IS raw scores. Significant predictors were then entered into Forward Stepwise models to determine the relative contributions of these individual variables to proband IS. 

Results: Our factor analysis generated a BAPQ-IS scale consisting of a subset of eight items from the original BAPQ-Rigid scale. Rigid items involving keeping things the same and disliking changes in routine were retained in our IS factor, and items involving disliking new or unfamiliar things were excluded. Our empirically derived BAPQ-IS factor differed only slightly from a previous BAPQ-IS scale generated through clinical consensus (Levin-Decanini et al., 2014); it contained the six items identified by Levin-Decanini et al. plus two additional items. Parent IS and proband IS were minimally related, although results reached statistical significance. Only proband RBS-R-IS scores had more than 2% of its variance explained by parental IS once significant proband variables were accounted for, and in all cases, parent IS explained the smallest amount of variance in each proband IS variable. 

Conclusions: The relative lack of association between parent IS and proband IS was surprising given the literature documenting positive correlations between parental rigidity and BAP characteristics and proband ASD (Losh et al., 2008; Piven et al., 1997). Although sample size was large, it was a simplex family sample, which may have resulted in a weaker relationship between parent and child IS compared to previous studies of multiplex families (e.g., Piven et al., 1997). Scale and measurement differences (e.g. characterization of adult versus child IS symptoms) may contribute to these results. Further research is needed on the validity and reliability of measuring IS and BAP features over the course of development.