A pilot randomised controlled trial of a group based social skills intervention for adults with autism spectrum disorder

Highlights

• Both social skills training and facilitating positive social interaction may lead to benefits for adults with ASD.

• Social skills training showed no benefits over facilitated interaction, although more research is needed to confirm this.

• The use of active control groups is recommended to control for non-specific aspects of an intervention.

Abstract

Background

Social skills training is commonly used for children and adolescents with ASD. There are few studies considering its efficacy in adults and those that exist usually compare it to standard treatment, as opposed to controlling for non-specific improvements that may result from receiving an intervention. We conducted a pilot study to investigate whether social skills training is more effective than non-specific support for adults with ASD.

Methods

19 adults with ASD were randomised to either social skills training or non-specific social interaction groups for 16 weeks. The Reading the Mind in the Eyes and Reading the Mind in the Voice Tasks were used to assess social cognition while the Social Responsiveness Scale (SRS-2) and the Weiss Functional Impairment Rating Scale (WFIRS-S) were used to measure real-life change pre- and post-treatment.

Results

Taken together, both groups improved across the course of the study on the ‘Eyes’ task (p = 0.049), SRS-2 (p = 0.03) and WFIRS-S (p = 0.053). However, there were no significant differences between the groups. There was some trend-level evidence that the social skills group improved more on the tests of social cognition (p ≤ 0.1 for group x time interaction). This was not the case for the SRS-2 (p = 0.25) and for the WFIRS-S there was a trend for the social interaction group to show greater improvements (p = 0.07).

Conclusions

This study did not show clear evidence of benefit from social skills training compared to facilitated social interaction, although larger studies are required to draw firm conclusions about efficacy.

Introduction

Interventions targeting social skills have been reported to be useful for children and adolescents with autism spectrum disorders (ASD) (Reichow, Steiner, & Volkmar, 2013). However, some individuals with ASD do not receive the diagnosis until adulthood and the social demands on adults differ from those on children. It is therefore important to establish whether comparable interventions are of use in an adult population.

A number of studies have reported details of social skills interventions in adults with ASD and generally report benefits (Bishop-Fitzpatrick, Minshew, & Eack, 2013). However, it is not always clear that the improvements noted in basic social cognitive skills in some studies generalise to benefits in ‘real-life’ social situations. Additionally, it is not known whether social skills interventions are effective due to the specific components of social skills training or whether there are non-specific benefits attributable to regular meetings with a supportive clinician and/or interacting with other people with ASD. Designing and running a social skills training programme may require more resources than a non-specific intervention therefore it is important to establish whether there is any increase in benefit from doing so.

We set out to pilot a group based social skills intervention, with specific social skills training, and compare its efficacy to an active control (social interaction) group. We aimed to use measures of social function including both social cognition tasks and measures capturing reported behaviour and function to see if improvements generalised beyond the research context. We hypothesised that social skills training would prove more effective than a general social interaction group for individuals with ASD.