• mood disorders
• post-traumatic stress disorder
• ruminative thinking
• broad autism phenotype

Background: A growing number of studies has stressed the higher vulnerability to trauma- and stress-related disorders among subjects with autistic traits (AT). In particular, subjects with AT seem to often show a higher risk to be exposed to traumatic events as well as reduced coping strategies, thus developing more frequently psychiatric symptoms after trauma-related experiences, even after a history of mild/moderate stressful events. Some authors have also hypothesized a possible role of ruminative thinking, a feature associated with AT, in increasing the risk for developing post-traumatic symptoms after the exposure to a traumatic experience, due to a negative effect of ruminations on processing the stressful event. Parallel, a possible link between AT and mood symptoms has also gained interest in literature.
In this framework, the aim of this work is to evaluate, in a sample of University students, the relationship among AT, post-traumatic symptoms, ruminative thinking and mood symptoms according to a spectrum approach, with a special focus on gender differences and on how AT interact with trauma-related symptomatology as well as with ruminations in predicting mood symptoms.
Methods: 178 students (93 males and 85 females) were recruited from three Universities of excellence in Italy. The presence of full-blown diagnoses of psychiatric disorders was evaluated by means of The Structured Clinical Interview for DSM-5 (SCID-5); in addition, we employed the Adult Autism Subthreshold Spectrum (AdAS Spectrum), the Ruminative Response Scale (RRS), the Trauma and Loss Spectrum, Self report (TALS-SR) and the Moods Spectrum, Self report (MOODS-SR) to evaluate, according to a dimensional approach, the wide spectrum of subthreshold and overthreshold symptomathology.
Results: Considering the quartiles of AdAS Spectrum score, the sample was split in two groups, with 133 subjects (74.7%) categorized as “low scorers” (below the 3rd quartile, Q3 = 65) and 45 subjects (25.3%) as “high scorers” (above the 3rd quartile, Q4 > 65). A higher prevalence of full-blown bipolar disorders as well as of suicidal ideation/attempts was found among high scorers, which reported also significantly higher scores on TALS-SR, MOOD-SR and RRS instruments. Despite no subject reported a full-blown Post-traumatic stress disorder (PTSD) symptomatology by means of the SCID-5, when assessing the presence of the disorder excluding the fulfilment of DSM-5 criterion A, we found a prevalence of PTSD symptoms of 17.1%, with significantly higher rates among AdAS Spectrum high scorers. No gender differences emerged for RRS scale. Among AdAS spectrum low scorers, females showed significantly higher scores than males in some TALS-SR domains (Loss events, Grief reaction, Re-experiencing), but no differences were found between males and females in the high scorer group. Significantly higher MOOD-SR depressive component and total scores were reported among females in all the groups; while higher manic symptoms among males were reported only in the whole sample and in the low scorer group. Finally, among high scorer group, females reported higher scores than males on AdAS Spectrum Childhood and adolescence domain. Statistically significant correlations were found, by means of Pearson's correlation, between AdAS Spectrum and TALS-SR, as well as between AdAS and RRS, with the strongest correlation emerging between AdAS Spectrum and MOODS-SR total scores. Performing a multiple linear regression, we found that TALS-SR, RRS and AdAS Spectrum total scores significantly predicted MOODS-SR total score, with the AdAS Spectrum demonstrating the highest standardized coefficient. Finally, performing a mediation analysis, results highlight that both total and direct effect of AdAS Spectrum total score on MOODS-SR total score were statistically significant, with AdAS Spectrum total scores showing also a significant indirect effect on MOODS-SR total score through TALS-SR and RRS.

Conclusions:
Results of this work highlight the presence of a significant relationship of AT with both full-blown mood disorders and mood spectrum symptoms, as well as with ruminative thinking and trauma/stress-related symptoms. In particular, the relationship between AT and mood symptoms may be partially mediated by the presence of ruminations and post-traumatic symptomatology. According to the complex Post-traumatic stress disorder (cPTSD) model, our data seem to corroborate the hypothesis of a possible development of PTSD-like symptomatology also in response to mild/moderate stressful experiences, especially in high-risk individuals.
Globally, this study suggests that the continuity between general and clinical population with respect to autistic core features might also apply to some clinically significant correlates of AT, like suicidal risk or mood disorder spectrum symptoms.