Introduction
Autism spectrum disorder (ASD) is a neurodevelopmental disorder characterized by deficits in communication and relational skills, repetitive verbal and motor behaviors, restricted patterns of interest, a need for a predictable and stable environment 1. Individuals with ASD often exhibit sensory perceptual anomalies, described as hypo- or hyper-reactivity to sensory input 2, which profoundly impact their daily lives and social inclusion 3-5.
Although clinical symptoms are present from early childhood, it is not uncommon for this disorder to go undiagnosed until adulthood. This is especially true for high-functioning autism, previously known as Asperger disorder, now classified in DSM-5 as ASD Level 1 (requiring support). Misdiagnosis often occurs because these individuals have strong academic and intellectual abilities. Thus, parents or teachers see them as gifted, interpreting their social isolation and unusual behaviors as a byproduct of their intelligence. As a result, high-functioning individuals with ASD may reach adulthood without a proper diagnosis or support. This lack of diagnosis and support exposes them to a high risk of psychopathological comorbidities related to distress and bullying 6.
These comorbid conditions may prompt individuals with ASD to seek help, but unfortunately, psychiatrists and psychologists are not always adequately trained to screen for ASD 7.
Thus, this paper presents the conceptualization and preliminary drafting of a new tool designed to support clinicians in the early identification of autism spectrum traits both in the general population and in psychiatric patients. Rather than validating the instrument at this stage, our goal is to offer a theoretically grounded and practically oriented proposal that may serve as a foundation for future empirical studies.
Materials and methods
The instrument
The SIAHA (Screening Instrument for Adult with High-Functioning Autism) has been designed with the goal to provide psychiatrists, psychologists, and other clinicians, with a tool able to help them to make a diagnosis of autism. Indeed, each scale’s item employed in the SIAHA is based on ASD criteria defined by the DSM-5. Symptoms and cut-off levels are aligned with DSM-5 criteria.
Figure 1 and Figure 2 presents the scale, including both the original Italian version and the English translation. This figure also provides a description of the clinical use and scoring method for the instrument.
Procedure
Data were collected by expert clinicians in two groups of patients: a sample of autistic patients and a sample of non-autistic patients with other psychiatric disorders. Psychiatric disorders included in the evaluation of non-autistic patients were: Brief Psychotic Disorder, Schizophrenia, Schizoaffective Disorder, Social Phobia, Avoidant Personality Disorder, Schizoid Personality Disorder, and Schizotypal Personality Disorder.
These disorders are often misdiagnosed as ASD. Patients with ASD involved in this study did not present any of these comorbidities.
In both groups of patients, the psychiatrist completed the SIAHA scale during a routine psychiatric visit. A control group including healthy subjects were also tested. Neuro-typical development group (TD) was composed by healthy volunteers recruited through random sampling of students at the University of Turin and health services operators, employing a snowball sampling technique 14. All participants were native Italian speakers and signed the written informed consent before taking part in the experiment. The experimental procedure was conducted in accordance with the Declaration of Helsinki 15.
Tool development process
The creation of the SIAHA tool followed four sequential phases to guarantee its validity and appropriateness for its intended use. These phases included: (i) Conceptualization, (ii) Item Generation, (iii) Questionnaire Development, and (iv) Evaluation 16,17.
v) Conceptualization
The first stage focused on identifying and defining the tool’s purpose and underlying constructs. The tool aimed to assess the presence of Autism Spectrum Disorder in adulthood and support clinicians in the diagnostic process. Moreover, SIAHA scale is developed to disentangle autistic traits from psychiatric symptoms and avoid misdiagnoses. The characteristics examined in the questionnaire closely align with the diagnostic conceptualization of autism as outlined in the DSM-5, with a narrow focus on deficits in social communication and interaction, alongside restricted, repetitive patterns of behavior, interests, or activities and their persistence over time 1.
vi) Items generation
In the second stage, expert clinicians developed a pool of potential items based on the predefined objectives and constructs. To enhance the reliability of the items, we grounded their creation in an extensive review of the literature on Autism Spectrum Disorder diagnostic criteria 1 and the latest guidelines from the ISS (i.e., Istituto Superiore di Sanità) 18. Ultimately, we produced a draft of the questionnaire, shown in Figure 1.
vii) Questionnaire development
This stage focused on psychometric testing, where each generated item was evaluated for its content and face validity by expert clinicians.
viii) Evaluation
The final stage consisted of a pilot test to examine the tool’s feasibility, assess its preliminary performance, and estimate its reliability. This step aimed to gather initial data on the tool’s functionality and provide insights into its potential effectiveness.
Participants
The total study sample comprised 150 participants (50 with ASD, 50 with Psychiatric condition, and 50 with Neurotypical Development). In addition to the SIAHA scale score filled by clinicians, background information for each participant such as age, sex, educational level, and psychiatric diagnosis were anonymously collected. All the participants, or their legal guardians, signed a written informed consent, authorizing data collection and data processing as well.
ASD group (ASD)
Patients with ASD were evaluated at the Regional Center for Autism in Adulthood. For this study we randomly selected 50 individuals with Level 1 ASD diagnosis, according to DSM-5 criteria. Evaluations were conducted following standard guidelines and included medical history collection, psychiatric examination, cognitive assessment using the WAIS-IV, RAADS tests, and ADI-R administered to family members 8.
In the ASD group, 56% (n = 28) of individuals were male, and 44% (n = 22) were female. Ages ranged from 19 to 60 years old (M = 30.8; SD = 9.54). Educational levels ranged from 8 and 18 years (M = 13; SD = 2.25).
Psychopathological group (PC)
Patients with psychiatric disorders were evaluated at the Psychiatric Units within the Mental Health Department.
For this study we randomly selected 50 individuals with: Brief Psychotic Disorder (n = 16), Schizophrenia (n = 9), Schizoaffective Disorder (n = 5), Social Phobia (n = 10), Avoidant (n = 2), Schizoid (n = 4) and Schizotypal Personality Disorders (n = 4). In the PC group, 66% (n = 33) of individuals were male, and 34% (n = 17) were female. Ages ranged from 18 and 70 years old (M = 38.2; SD = 13.7). Educational levels ranged from 5 and 18 years (M = 10.8; SD = 3.34).
Neuro-typical development group (TD)
For this study, we also recruited 50 individuals who did not exhibit psychiatric or autistic symptoms to serve as a control group. These individuals were assessed using SIAHA by the same team from the autism center. In the TD group, 42% (n = 21) of individuals were male, and 58% (n = 29) were female. Ages ranged from 22 and 65 years old (M = 31; SD = 10.2). Educational levels ranged from 8 and 18 years (M = 15.8; SD = 2.65).
Results
Demographic
First, descriptive statistics were performed using Jamovi (version 1.6), reporting means and standard deviations for demographic variables (Tab. I). Results from the Chi-Square test revealed the absence of any differences in the male/female ratio in each groups X2 (5, N = 150) = 7.12, p = 0.212. However, significant differences were observed between groups in age [F (2, 96.1) = 5.68, p = 0.005] and educational level [F(2, 95.7) = 35.87, p < 0.001]. Games-Howell post-hoc comparisons (assuming unequal variance) revealed a significant age difference only between the ASD (M = 30.8, SD = 9.54) and PC (M = 38.2, SD = 13.7) groups. Regarding educational level, significant differences were found between all groups: ASD (M = 13, SD = 2.25), PC (M = 10.8, SD = 3.34), and TD (M = 15.8, SD = 2.65), with PC group showing the lowest educational level.
The SIAHA scale
Each of the SIAHA items was examined to test whether significant differences were observed among the 3 groups (ASD, PC and TD). We found significant differences (p < 0.05) among groups for all items (Tab. II) except 1.3.d (Does the partner complain about communication and relationship difficulties due to communicative inadequacy, e.g., cold, closed communicative style?) and 2.2.c (Are there any verbal rituals, e.g., a sequence of questions requiring an answer, which, if unmet, cause strong discomfort or behavioral disturbance? Were they present in the first eight years of life?). Also, we computed eta squared (η2) to examine the effect size and include it the same table. Additionally, we calculated Cronbach’s Alpha to assess the internal consistency of the scale, which yielded a value of 0.889 suggesting good internal consistency. Finally, post-hoc comparisons using Tukey’s HSD test were conducted to examine pairwise differences between Groups. Two items (1.3.d and 2.2.c) did not show significant group differences and were therefore excluded from further pairwise comparisons (Tab. III).
As reported in the table above, ANOVAs were conducted on all 20 items. Of these, 18 items showed a significant main effect of group (p < . 05). For those items, post-hoc comparisons using Tukey’s HSD were conducted to examine pairwise group differences.
After analyzing the scores on each individual scale’s item, we performed a one-way ANOVA comparing the 3 groups on the total scores. As reported in Table 3, the differences in total scores across the groups were all statistically significant (p < 0.001). Also, we computed eta squared (η2) to examine the effect size and include it the same table. Finally, post-hoc comparisons using Tukey’s HSD test were conducted to examine pairwise differences between Groups (Tab. V).
Discussion
ASD are frequently misdiagnosed in adulthood, especially in cases of ASD DSM-5 Level 1. Therefore, a reliable screening scale could help clinicians to identify ASD patients in clinical setting. The results of our study confirm the utility of the questionnaire in identifying central clinical differences among individuals with ASD, those with severe psychopathological disorders, and those with a neurotypical development.
For what concern the demographic data, no significant differences in gender composition emerged among the groups, indicating a balanced ratio between males and females. However, significant differences were found in age and level of education among the groups. Individuals with ASD had a lower average age compared to patients with severe psychopathological disorders. Additionally, participants with ASD and those in the neurotypical control group (TD) had higher levels of education compared to patients with severe psychopathological disorders. These findings are consistent with previous studies suggesting that individuals with Autism Spectrum Disorders often exhibit elevated intellectual functioning compared to those with severe psychopathological disorders 9,10.
Concerning the SIAHA scale scores, our results revealed significant differences among the 3 groups in almost all items providing compelling evidence of a systematic difference in item-level performance across the three experimental groups. Individuals with ASD showed greater difficulties in understanding jokes, metaphors, and modulating voice compared to the other groups, reflecting a distinctive characteristic of communication difficulties in autism 10.
Individuals with severe psychopathological disorders, on the other hand, tended to exhibit symptoms such as delusions, hallucinations, and negative symptoms (e.g., flattened affect, alogia), which were not prevalent in individuals with ASD 11,12. These results are consistent with previous literature highlighting how individuals with schizophrenia and other psychotic disorders manifest distinct symptoms compared to those with ASD, despite potential symptomatic overlaps 13.
From the results obtained by comparing the SIAHA total scores across the different groups, it emerges that this tool could assist psychiatrists in better identifying the presence of basic autism criteria during initial screenings, potentially reducing the misdiagnosis often attributed to adult patients with autism. This is particularly supported by the Multistep Network model for assessing participants with ASD 8.
Although the present study provides valuable insights into the tool’s effectiveness, it is important to note that specific statistical analyses to directly validate the tool assessing reliability, sensitivity, and specificity were not conducted. As a result, while these constructs may have been indirectly explored through the observed differences between groups, their formal evaluation was not part of this analysis. This represents a limitation of the study, as further research should aim to explicitly assess these psychometric properties to better understand the tool’s overall performance. Moreover, our findings suggest that while the overall trend is robust, the sensitivity of some items to group differences may vary. Future work might explore whether these items could be refined or whether they tap into dimensions that are more resistant to change.
In summary, our results underscore the importance of using precise instruments, such as SIAHA questionnaire, to properly recognizing the specific characteristics of autism spectrum disorders and severe psychopathological disorders in clinical and diagnostic settings. Understanding these differences can enhance diagnostic accuracy and inform more targeted and effective therapeutic interventions. Despite symptomatic overlaps, clinical manifestations of ASD and psychotic disorders remain distinctly identifiable and require differentiated therapeutic approaches 10-13.
In case of positive screening for ASD, it is essential to gather accurate data on childhood behavior and utilize cognitive and ASD diagnostic tests.
Conclusions
The SIAHA scale could be a valuable tool for clinicians in screening for ASD Level 1. A positive results from the SIAHA screening should be followed by a comprehensive clinical evaluation and specific diagnostic tests conducted by expert and trained clinicians. Additionally, translating the SIAHA test into different languages and validating it in various countries could enhance its utility globally.
Acknowledgements
We thank the people who participated in this study, as well as to their families and the professionals involved.
Funding
This research received no specific grant from any funding agency, commercial or not-for-profit sectors.
Conflict of interest statement
The Authors declare no conflict of interest.
Authors’ contributions
RK: He invented the instrument, conceptualized the research, supervised the research, and reviewed the article. GC: He conducted data collection and statistical analysis and drafted the article. EN: She collaborated on the introduction and discussion section of the paper.
EB: She conducted data collection, clinical record analysis, interviews, database construction and compilation. ED: She conducted data collection, clinical record analysis, interviews, database construction and compilation. She drafted the article. RB: He conducted statistical analysis. SB: She collaborated on the introduction section of the paper. CR, LQ, AC, CR, EC, AV: They participated in the interviews. FV: He supervised the research. FC & ODM: Reviewed the article.
Ethical consideration
All procedures performed in studies involving human participants were in accordance with the ethical standards of the institutional and/or national research committee and with the 1964 Helsinki Declaration and its later amendments or comparable ethical standards.
Figures and tables
FIGURE 1. Italian version of the SIAHA items related to Criterion 1.
FIGURE 2. English translation of the SIAHA items related to Criterion 1.
| Sex | Age | Education | |
|---|---|---|---|
| ASD | F: 22 | 29.5 (7.74) | 13.2 (2.11) |
| M: 28 | 31.8 (10.8) | 12.8 (2.36) | |
| PC | F: 17 | 39.8 (12.5) | 11.2 (3.78) |
| M: 33 | 37.4 (14.4) | 10.6 (3.13) | |
| TD | F: 29 | 30.4 (10.9) | 15.7 (2.55) |
| M: 21 | 31.8 (9.36) | 15.8 (2.84) | |
| ASD = Autism Spectrum Disorder; PC = Psychiatric condition; TD = Neuro-typical Development. | |||
| Items | ASD | PC | TD | |||||
|---|---|---|---|---|---|---|---|---|
| F | M | F | M | F | M | p | η2 | |
| c1.2a | 0.63 (0.49) | 0.89 (0.31) | 0.17 (0.39) | 048 (0.5) | 0.10 (0.31) | 0.14 (0.35) | < 0.01 | .301 |
| c1.2b | 0.72 (0.45) | 0.82 (0.39) | 0.29 (0.47) | 0.48 (0.5) | 0 (0) | 0 (0) | < 0.01 | .423 |
| c1.2c | 0.4 (0.5) | 0.71 (0.46) | 0.17 (0.39) | 0.33 (0.47) | 0.03 (0.18) | 0 (0) | < 0.01 | .253 |
| c1.2d | 0.59 (0.5) | 0.75 (0.44) | 0.23 (0.43) | 0.48 (0.5) | 0.06 (0.25) | 0 (0) | < 0.01 | .293 |
| c1.2e | 0.27 (0.45) | 0.42 (0.5) | 0.05 (0.24) | 0.27 (0.45) | 0.03 (0.18) | 0 (0) | < 0.01 | .124 |
| c.1.3a | 1.00 (0) | 0.78 (0.41) | 0.88 (0.33) | 0.84 (0.36) | 0.20 (0.41) | 0.09 (0.30) | < 0.01 | .483 |
| c.1.3b | 0.77 (0.42) | 1.00 (0) | 0.94 (0.24) | 0.81 (0.39) | 0.06 (0.25) | 0 (0) | < 0.01 | .654 |
| c.1.3c | 0.77 (0.42) | 0.64 (0.48) | 0.64 (0.49) | 0.45 (0.50) | 0.03 (0.18) | 0.14 (0.35) | < 0.01 | .276 |
| c1.3d | 0.18 (0.39) | 0.07 (0.26) | 0.23 (0.43) | 0.24 (0.43) | 0.10 (0.31) | 0.04 (0.21) | .063 (ns) | .037 |
| c2.1a | 0.36 (0.49) | 0.57 (0.5) | 0 (0) | 0.15 (0.36) | 0.10 (0.31) | 0.04 (0.21) | < 0.01 | .197 |
| c2.1b | 0.04 (0.21) | 0.39 (0.49) | 0.05 (0.24) | 0.18 (0.39) | 0 (0) | 0.04 (0.21) | 0.01 | .070 |
| c2.1c | 0.09 (0.29) | 0.46 (0.5) | 0 (0) | 0.12 (0.33) | 0.03 (0.18) | 0 (0) | < 0.01 | .125 |
| c.2.2a | 0.86 (0.35) | 0.85 (0.35) | 0.58 (0.50) | 0.42 (0.50) | 0.06 (0.25) | 0 (0) | < 0.01 | .452 |
| c2.2b | 0.18 (0.39) | 0.57 (0.5) | 0.41 (0.50) | 0.30 (0.46) | 0.06 (0.25) | 0 (0) | < 0.01 | .129 |
| c2.2c | 0.09 (0.29) | 0.21 (0.41) | 0 (0) | 0.09 (0.29) | 0.06 (0.25) | 0 (0) | 0.33 (ns) | .035 |
| c2.4a | 0.77 (0.39) | 0.6 (0.49) | 0.05 (0.24) | 0.15 (0.36) | 0 (0) | 0.04 (0.21) | < 0.01 | .446 |
| c2.4b | 0.5 (0.51) | 0.25 (0.44) | 0 (0) | 0 (0) | 0 (0) | 0 (0) | < 0.01 | .273 |
| c2.4c | 0.68 (0.47) | 0.71 (0.46) | 0.17 (0.39) | 0.12 (0.33) | 0.06 (0.25) | 0 (0) | < 0.01 | .407 |
| c2.4d | 0.22 (0.42) | 0.28 (0.46) | 0.05 (0.24) | 0.03 (0.17) | 0 (0) | 0 (0) | < 0.01 | .145 |
| c2.4e | 0.18 (0.39) | 0.28 (0.46) | 0 (0) | 0 (0) | 0.03 (0.18) | 0.04 (0.21) | < 0.01 | .130 |
| ASD = Autism Spectrum Disorder; PC = Psychiatric condition; TD = Neuro-typical Development. Mean scores and (SDs) are reported for F = Female participants and M = Male participants. The last two columns of the table reports the statistical results about the group comparison on each item (n.s. = not significant) and the effect size values. | ||||||||
| Items | Group 1 vs 2 | Std. Error | p | Group 1 vs 3 | Std. Error | p | Group 2 vs 3 | Std. Error | p |
|---|---|---|---|---|---|---|---|---|---|
| c1.2a | .660* | .084 | < .001 | .400* | .084 | < .001 | -.260* | .084 | .006 |
| c1.2b | .780* | .075 | < .001 | .360* | .075 | < .001 | -.420* | .075 | < .001 |
| c1.2c | .560* | .080 | < .001 | .300* | .080 | < .001 | -.260* | .080 | .004 |
| c1.2d | .640* | .082 | < .001 | .280* | .082 | .002 | -.360* | .082 | < .001 |
| c1.2e | .340* | .075 | < .001 | .160 | .075 | .085 | -.180* | .075 | .045 |
| c.1.3a | .720* | .070 | < .001 | .020 | .070 | .956 | -.700* | .070 | < .001 |
| c.1.3b | 860* | .058 | . < .001 | .040 | .058 | .771 | -.820* | .058 | < .001 |
| c.1.3c | .620* | .085 | < .001 | .180 | .085 | .091 | -.440* | .085 | < .001 |
| c2.1a | .400* | .075 | < .001 | .380* | .075 | < .001 | -.020 | .075 | .962 |
| c2.1b | .220* | .066 | .003 | .100 | .066 | .289 | -.120 | .066 | .169 |
| c2.1c | .280* | .064 | < .001 | .220* | .064 | .002 | -.060 | .064 | .620 |
| c.2.2a | .820* | .075 | < .001 | .380* | .075 | < .001 | -.440* | .075 | < .001 |
| c2.2b | .360* | .083 | < .001 | .060 | .083 | .749 | -.300* | .083 | .001 |
| c2.4a | .680* | .068 | < .001 | .580* | .068 | < .001 | -.100 | .068 | .303 |
| c2.4b | .360* | .056 | < .001 | .360* | .056 | < .001 | .000 | .056 | 1.000 |
| c2.4c | .660* | .071 | < .001 | .560* | .071 | < .001 | -.100 | .071 | .337 |
| c2.4d | .260* | .056 | < .001 | .220* | .056 | < .001 | -. 040 | .056 | .756 |
| c2.4e | .200* | .055 | .001 | .240* | .055 | < .001 | .040 | .055 | .746 |
| Group 1 = ASD, Group 2 = TD, Group 3 = PC. Values represent mean differences (MD). Significance levels: * p < . 05, ** p < . 01, *** p < . 001; bold = not significant. | |||||||||
| Items | F | gdl1 | gdl2 | p | η2 |
|---|---|---|---|---|---|
| K. Score C1.1 | 73.0 | 2 | 84.5 | < 0.001 | .409 |
| K. Score C1.2 | 114.2 | 2 | 77.6 | < 0.001 | .483 |
| K. Score C1.3 | 56.3 | 2 | 91.9 | < 0.001 | .600 |
| K. Score C2.1 | 22.2 | 2 | 95.9 | < 0.001 | .276 |
| K. Score C2.2 | 138.8 | 2 | 92.5 | < 0.001 | .493 |
| K. Score C2.3 | 69.2 | 2 | 96.3 | < 0.001 | .533 |
| K. Score C2.4 | 225.2 | 2 | 82.1 | < 0.001 | .633 |
| K Total C1 | 237.1 | 2 | 73.0 | < 0.001 | .698 |
| K Total C2 | 270.1 | 2 | 94.8 | < 0.001 | .721 |
| C1 = Criterion 1 and C2 = Criterion 2. The last column of the table reports the statistical results about group comparisons on the total scores and the effect size values. | |||||
| Items | Group 1 vs 2 | Std. Error | p | Group 1 vs 3 | Std. Error | p | Group 2 vs 3 | Std. Error | p |
|---|---|---|---|---|---|---|---|---|---|
| K. Score C1.1 | .760* | .076 | < .001 | .300* | .076 | < .001 | -.460* | .076 | < .001 |
| K. Score C1.2 | .840* | .072 | < .001 | .420* | .072 | < .001 | -.420* | .072 | < .001 |
| K. Score C1.3 | .740* | .056 | < .001 | .040 | .056 | .756 | -.700* | .056 | < .001 |
| K. Score C2.1 | .520* | .079 | < .001 | .500* | .079 | < .001 | -.020 | .079 | .965 |
| K. Score C2.2 | .860* | .072 | < .001 | .440* | .072 | < .001 | -.420* | .072 | < .001 |
| K. Score C2.3 | .740* | .065 | < .001 | .720* | .065 | < .001 | -.020 | .065 | .949 |
| K. Score C2.4 | .880* | .061 | < .001 | .780* | .061 | < .001 | -.100 | .061 | .228 |
| K Total C1 | 2.340* | .130 | < .001 | .760* | .130 | < .001 | -1580* | .130 | < .001 |
| K Total C2 | 2.960* | .161 | < .001 | 2.400* | .161 | < .001 | -.560* | .161 | .002 |
| Group 1 = ASD, Group 2 = TD, Group 3 = PC. Values represent mean differences (MD). Significance levels: * p < . 05, ** p < . 01, *** p < . 001; bold = not significant. | |||||||||
