J Psychiatr Res. 2026 Feb 12;197:107-117. doi: 10.1016/j.jpsychires.2026.02.025. Online ahead of print.
ABSTRACT
PURPOSE: Increased social demands during adolescence and early adulthood pose specific challenges for individuals with autism spectrum disorder (ASD) without accompanying intellectual impairment. Targeted group-based social competence trainings are effective interventions. This work evaluates the Zurich social competence training (KOMPASS), a manualized program for adolescents and young adults (12-25 years) with ASD.
METHODS: In this quasi-experimental longitudinal study with a naturalistic waiting period and follow-up, we assessed participants of the KOMPASS intervention (N = 108) at our clinic through parent-, teacher- and self-reports along with a computerized task. A subset of participants (N = 65) was additionally assessed during a waiting period prior to intervention, allowing within-subject comparison of change before and during treatment. One-year follow-up data for the KOMPASS intervention group were collected (N = 62). Furthermore, we analyzed pre- and post-data from 35 individuals who completed KOMPASS at external clinics using the KOMPASS- manual without prior introduction to the method. Critically, we used Bayesian statistics to show evidence for stability of effects and comparable outcomes across settings.
RESULTS: Compared to the waitlist period, the KOMPASS training period showed a significant decrease in social and behavioral difficulties associated with ASD and improvements in social competences, which were sustained one year after training. General psychiatric symptoms also improved post-intervention in the KOMPASS training period. Participants from external clinics demonstrated similar significant improvements, affirming the manual’s effectiveness in external settings. No significant gender or age moderation effects were observed for the primary outcomes. Participants and caregivers expressed satisfaction with the training.
CONCLUSION: KOMPASS demonstrates significant effects in improving social functioning and reducing ASD-related difficulties in adolescents and young adults. The study benefits from a large sample, a naturalistic waiting-period comparison, a one-year follow-up, multi-informant assessment, and external validation. Limitations include the quasi-experimental design, the lack of formal fidelity monitoring, and reliance on some older assessment measures.
PMID:41774970 | DOI:10.1016/j.jpsychires.2026.02.025
