Mapping Anxiety in Minimally Verbal Autistic Pupils

Background: Autistic individuals are significantly more likely to experience clinically significant levels of anxiety than neurotypical peers. Better support for autistic individuals in managing anxiety would bring numerous benefits notably in support with behaviours that challenge. Addressing anxiety is also a priority for parents of autistic individuals. Better support requires a better understanding of anxiety ‘triggers’ for autistic individuals. Recent research in psychopathology has moved towards an understanding that conditions involve highly connected networks of symptoms. These networks are created by interactions and causal influences between symptoms. From this perspective, anxiety may develop as a result of prolonged interactions between various triggers. A networked understanding of anxiety may give a better understanding of how to provide more effective support.

Objectives: This study, co-produced by school-based professionals and parents, examined the viability of better understanding anxiety in minimally verbal autistic adolescents through the use of perceived causal relationship (PCR) scaling and network analysis in order to provide more direct support.

Methods: Anxiety triggers were identified for 18 pupils (16 male and 2 female) attending a government funded, special school in inner-city London. Pupils’ ages ranged from 14 years 10 months to 17;7 (M=15;8, SD=1;9). Parents and professionals rated how frequently pupils experienced anxiety as a result of each trigger. The eight response options ranging between “Not at all in the past month” and “Daily or almost daily for most of the day” and scored 0-7, respectively. Parents and professionals then responded to causal association questions - the extent to which “trigger X” caused “trigger Y” and vice-versa. Response options were rated from 0-10 with 0, 5 and 10 denoting “Not at all”, “Moderately cause,” and “Strong cause” respectively. Mean Causal Association and Mean Causal Effect scores were calculated. To map anxiety a directed network of PCR scores was generated using the R-package qgraph. Nodes represented triggers and PCR scores were represented by edges between the nodes. The direction of the edges represented the direction of the perceived causal effect and the thickness indicated its strength. Centrality measures were used to examine which triggers were most influential (central) in the network.

Results: The network maps provided a clear, visual representation of the inter-relationship between anxiety triggers. Each pupil was readily identifiable from the maps giving a measure of face validity. Qualitative data collected from professionals and parents endorsed the network map as providing a clearer indication of specific triggers to target for support.

Conclusions: This study represents the first time network analysis has been used to map anxiety in minimally verbal autistic adolescents. It has promise as a method of assessing the inter-relationship of anxiety triggers as a means to direct more effective support. Future research should examine the extent to which targeting a high-centrality trigger for support initiates a beneficial therapeutic cascade that mitigates the impact of other triggers, addressing anxiety more effectively in terms of time and resources and more rapidly reducing individuals’ distress.