2026 ECNP Congress

PS03-3072

All that glitters is not gold: shifting referral pathways and diagnostic yield in adult autism assessment

Maria Martinez Ramirez — J. Lugo Marin, I. Setien Ramos, J.A. Ramos-Quiroga, L. Gisbert Gustemps · Hospital Universitari Vall Hebron, Mental Health, Barcelona, Spain

2026 ECNP Congress

Background

Referral pathways to adult autism spectrum disorder (ASD) diagnostic services have undergone substantial changes in recent years, reflecting broader shifts in public awareness, access to information, and healthcare utilisation patterns. In particular, increasing recognition of ASD in adulthood and the growing visibility of autism-related content have contributed to a rise in self-referrals [1]. These changes may have important implications for both the diagnostic yield of services and the clinical and sociodemographic composition of referred populations [2]. Despite these developments, longitudinal evidence examining how referral source interacts with key predictors of ASD diagnosis—such as age, sex, and clinical presentation—remains limited. Understanding these dynamics is essential for optimising diagnostic pathways and resource allocation.

Methods

We conducted a retrospective cohort study of 775 adults referred for ASD assessment between 2017 and 2024 at a specialised multidisciplinary diagnostic unit. Of these individuals, 523 met diagnostic criteria for ASD, while 252 did not. All participants underwent a comprehensive clinical evaluation, including a structured psychiatric interview and administration of standardised diagnostic instruments: the Autism Diagnostic Observation Schedule–2 (ADOS-2) and, where available (n ≈ 537), the Autism Diagnostic Interview–Revised (ADI-R). Sociodemographic variables (e.g., sex, age, education, employment status, and living arrangements), referral source (self vs. professional), and clinical measures were systematically collected. Group differences were analysed using χ² tests for categorical variables and independent-samples t-tests for continuous measures. Binary logistic regression models were employed to identify predictors of both referral type and ASD diagnosis, incorporating temporal trends and interaction effects.

Results

Individuals diagnosed with ASD differed significantly from those not meeting diagnostic criteria across multiple sociodemographic domains. The ASD group had a higher proportion of males (59.5% vs. 36.1%), and differed significantly in living arrangements, educational attainment, and employment status (all p < 0.001). Referral source emerged as a strong predictor of diagnostic outcome: individuals referred by professionals were significantly more likely to receive an ASD diagnosis than those who self-referred (OR = 3.21). Over the study period, self-referral rates increased significantly (OR = 1.43 per year, 95% CI 1.22–1.68, p < 0.001). Referral patterns were also shaped by demographic factors, with women and younger individuals more likely to self-refer. Importantly, a significant interaction between year and referral source (OR = 1.40, p = 0.006) indicated diverging trends in diagnostic yield: while the probability of diagnosis among professionally referred individuals remained stable, it declined over time among self-referred cases. Clinically, participants with ASD demonstrated significantly higher scores across all ADOS-2 domains and ADI-R measures compared with non-ASD participants, with large effect sizes (Cohen’s d ranging from 1.15 to 1.55). The only exception was the restricted and repetitive behaviours domain of the ADOS-2, which did not show a statistically significant difference between groups.

Conclusions

Adult ASD diagnostic services are increasingly characterised by rising rates of self-referral alongside stable professional referral pathways. However, diagnostic yield remains substantially higher among professionally referred individuals. These findings emphasise the importance of referral source as a key determinant of pre-test probability and highlight the continued necessity of rigorous, standardised clinical assessment in ensuring accurate diagnosis in adult populations.

References

  1. Cumin, J., Pelaez, S., & Mottron, L. (2022). Positive and differential diagnosis of autism in verbal women of typical intelligence: A Delphi study. Autism : the international journal of research and practice, 26(5), 1153–1164. https://doi.org/10.1177/13623613211042719
  2. Rødgaard, E. M., Jensen, K., Vergnes, J. N., Soulières, I., & Mottron, L. (2019). Temporal Changes in Effect Sizes of Studies Comparing Individuals With and Without Autism: A Meta-analysis. JAMA psychiatry, 76(11), 1124–1132. https://doi.org/10.1001/jamapsychiatry.2019.1956