Age-Related Changes in Autistic Traits: A Survey for the Adults with Currently High Autistic Traits with and without Autism Spectrum Disorder.

Background: Recent studies have showed that adults with autism spectrum disorder (ASD) exhibit symptomatic overlaps with other psychiatric disorders such as schizophrenia spectrum disorder and borderline personality disorder. These results indicate that adults with non-ASD disorders also present ASD traits and, especially in adults, a diagnostic boundary between ASD and non-ASD is blurry. However, little is known about whether or not developmental trajectories in autistic traits show overlaps between adult ASD and non-ASD with currently high autistic traits. Do developmental trajectories in autistic traits between adult ASD and non-ASD show similar/different patterns, especially when they all exhibit high ASD traits and are difficult to differentiate? Thus, we investigate overlaps or differences of age-related changes in autistic traits among adults with and without ASD who present currently high autistic traits.

Objectives: We aim to reveal the overlaps or differences in age-related changes in autistic traits between adult ASDs and non-ASDs, both of whom currently presents high ASD traits.

Methods: We recruited 26 adults (6 female; mean age = 28.31, SD = 8.16) with possible ASD who consulted our specialists for the detailed assessment of ASD through two or three weeks of hospitalization. Our diagnostic process included approximately ten hours of interviews for clients and their parents during their admission. After the clinical interview, the psychiatrist diagnosed the clients based on the consensus of the psychiatrists and psychologists according to the Diagnostic and Statistical Manual of Mental Disorders-5 (DSM-5) criteria. We used Parent-interview ASD rating Scale (PARS) score to examine age-related changes in autistic traits. This scale is a semi-structured interview for clients’ parents and evaluates ASD traits in their childhood, adolescent, and adulthood. As a self-rating scale, we also analyzed Autism-Spectrum Quotient (AQ) scores to confirm currently high ASD traits among participants.

Results: Detailed examination through weeks of hospitalization resulted in the clinical diagnosis of ASD (n = 16) and non-ASD (n= 10). A t-test showed that AQ total scores of ASD group was not statistically different from those of non-ASD group (t = 0.44, p = 0.66), which confirmed that both groups similarly presented high autistic traits. We analyzed early childhood peak scores and current adolescent/adult scores of PARS as the index of age-related changes in autistic traits. A two-way ANOVA (Group x Period) showed the significant interaction between Group and Period (F = 7.30, p = 0.01). Post-hoc t-tests indicated that the early childhood peak scores of ASD group were significantly higher than those of non-ASD group (t = 2.82, p = 0.01), while the current adolescent/adult scores of ASD group was not statistically different from those of non-ASD group (t = 0.60, p = 0.55).

Conclusions: Our preliminary results indicate that age-related changes in autistic traits could exhibit different patterns between adult ASD and non-ASD groups even though both groups presently show similarly high ASD features. This finding could be helpful for clinicians to appropriately diagnose adults with high autistic traits and thus giving them efficient service.