1. Alba LA, Salinas G. Parent Ratings and Clinician Observational Tools in Autism: A Pilot Study in Spanish-Speaking Families. J Autism Dev Disord. 2026.

PURPOSE: Research shows modest concordance between parent-reported rating scales (e.g., ASRS, SRS-2) and diagnostic outcomes for autistic children. Yet parents frequently provide detailed autistic behaviors during interviews that align closely with clinician observations, indicating that measurement context, such as environment, expectations, and observation setting, differentially influences parent ratings versus direct clinical observation. However, this research has been predominantly conducted with English-speaking populations, leaving measurement concordance underexplored among linguistically diverse populations, such as Spanish-speaking families. This pilot study estimates associations between parent-reported and clinician-administered autism measures in a U.S. Spanish-speaking sample. METHODS: Eighteen Spanish-speaking parents completed the ASRS and SRS-2 during their child’s initial autism evaluation. Clinicians administered the CARS2-ST and ADOS-2. RESULTS: Spearman’s rho and Kendall’s tau rank correlations showed positive associations between social communication and interaction scores on the ASRS and SRS-2, and the ADOS-2 Calibrated Severity Score (CSS). Only the ASRS social communication and interaction scores showed an association with CARS2-ST t-scores. Confidence intervals for parent-reported RRBs were not significantly associated with ADOS-2 CSS and CARS2-ST total score. CONCLUSION: This pilot study estimates associations between autism-specific rating scales and clinician-administered observational tools among Spanish-speaking families. Findings are presented as preliminary point estimates with confidence intervals to establish a foundation for future research on culturally and linguistically responsive assessment practices and to highlight areas warranting further investigation.

Lien vers le texte intégral (Open Access ou abonnement)

2. Albores-Gallo L, Varela-Orozco KL, Roldán-Ceballos O, List-Hilton C, Maurer AP. Mexican version of the Autism Behavior Checklist: validity and reliability. Bol Med Hosp Infant Mex. 2026; 83(4): 252-60.

BACKGROUND: The Autism Behavior Checklist (ABC) evaluates symptoms in individuals with Autism Spectrum Disorders (ASD) within a school setting. The purpose of this study was to evaluate the validity of the ABC Mexican version. METHODS: Participants were children (n = 133, aged 2-17). All parents were interviewed with the autism diagnostic interview-revised (ADI-R) to confirm an ASD diagnosis and then answered the ABC checklist. The sample for the ABC test-retest analysis consisted of 19 parents with unaffected typically developing children between 2 and 17 years old. RESULTS: Children and adolescents (n = 133) with a mean age of 6.9 years (standard deviation [SD] 3.7), 83.5% were males. The ABC total mean score was 69.3 (SD 25). The Internal consistency through the Cronbach Alfa coefficient was α = 0.83 < p = 0.001 for the 57 ABC items. The 10-day test-retest reliability showed a Pearson correlation coefficient of r = 0.98, p < 0.001. The Spearman correlation coefficients between the ABC subscales and the ADI-R ranged from (rs = 0.494) to (rs = 0.816). Criterion validity with a cutoff of 30 resulted in a sensitivity of 87% and a specificity of 37%. The best Kappa coefficient was 0.285 between the ABC and the ADI-R. CONCLUSIONS: The Mexican ABC has good psychometric properties. Further studies should investigate its value in educational settings.

Lien vers le texte intégral (Open Access ou abonnement)

3. Boldsen S, Taipale J. Contiguity and the Genesis of Autistic Experience: What Psychoanalysis Contributes to Phenomenology. Psychopathology. 2026: 1.

Background While contemporary phenomenologists have highlighted the central role of sensory differences in autism, such differences have rarely been phenomenologically interrogated in their own right. The psychoanalytic approach to autism offers an underexplored resource for addressing this gap. Focusing on Thomas Ogden’s concept of contiguity, this paper investigates how autistic sensory experience may be fundamentally structured through a « touch-like » mode of relating to self and world. Summary Drawing on Ogden’s suggestion that autism is characterized by contiguity as a norm for the experience of self and world, we argue that contiguity presents as a mode of generating sensory experience in autism across a broad range of phenomena commonly grouped under the heading of sensory differences. Illustrating this point through an integrative reading of autistic autobiographical writings, we emphasize how autistic sensory experiences and behaviors may often serve to manage a felt porousness of bodily boundaries. As we show, this does not render autism pathological but accounts for autistic experience in terms of an accentuation of an indispensable dimension of human experience. In conclusion, we briefly discuss how Ogden’s psychoanalytic theorization may pave the way for a genetic phenomenology of autism. Key points We argue that Ogden’s concept of contiguity allows for a radical rearticulation of autistic experience in terms of a felt permeability to the world and others rather than an increased detachment from them. In doing so, we highlight the value of bringing psychoanalysis into dialogue with phenomenology for interdisciplinary autism research.

Lien vers le texte intégral (Open Access ou abonnement)

4. Casteel C, Prajapati AP, Pitzer C, Sprengel R, Eltokhi A. SHANK2A overexpression in forebrain neurons induces early-life communication deficits in mice. Behav Brain Res. 2026; 512: 116295.

SHANK2 is a postsynaptic scaffolding protein critical for excitatory synapse organization, and alterations in SHANK2 dosage are strongly associated with autism spectrum disorder (ASD). While SHANK2 loss-of-function rodent models have been extensively studied, the impact of increased SHANK2 expression on early-life communication remains unclear. Here, we examined ultrasonic vocalizations (USVs) in forebrain-specific SHANK2A-overexpressing (SH-WT) mice using the pup isolation paradigm at P8 and P12, a sensitive developmental window for assessing early social communication. SH-WT pups exhibited a transient increase in vocal output at P8, characterized by elevated call production and altered temporal organization, while fundamental acoustic features remained largely unchanged. By P12, these quantitative differences were no longer evident, indicating developmental normalization. However, SH-WT pups showed subtle alterations in call structure and composition, along with persistent changes in vocal syntax, reflected by more centralized and less diverse transition networks. Together, these findings indicate that SHANK2A overexpression transiently enhances early vocal output while inducing lasting alterations in the organization and flexibility of vocal behavior, suggesting disrupted maturation of communication-related neural circuits relevant to ASD.

Lien vers le texte intégral (Open Access ou abonnement)

5. Castro K, Duarte CK, Silva E, Hoffmann L, Borchardt JL, Pohl L, Luçardo J, Lapschies M, Cascaes AM, Freitas-Vilela AA, Cenci A, Vaz JS. Quality of Life in Autism: A Systematic Review and Meta-Analysis of Patients and Caregivers. Autism Res. 2026: e70348.

Autism spectrum disorder (ASD) is a lifelong neurodevelopmental condition associated with significant challenges in daily functioning and well-being. Caregivers often experience substantial psychological and physical burden. However, comprehensive synthesis of quality of life (QoL) outcomes for both individuals with ASD and their caregivers remains limited. This study evaluated QoL in individuals with ASD and their caregivers and examined key psychological, social, and economic determinants. A systematic review and meta-analysis were conducted following PRISMA guidelines (PROSPERO: CRD42024538872). PubMed, Scopus, Embase, PsycINFO, and preprint repositories were searched for studies published between January 2013 and July 2025. Eligible studies assessed QoL using validated instruments in individuals with ASD aged ≤ 19 years and/or their caregivers. Random-effects meta-analysis was used to estimate pooled mean differences. Of 14,368 records identified, 216 studies were included. Quantitative meta-analyses included up to 173 studies, depending on the outcome and availability of comparable data. Individuals with ASD scored 21.7 points lower than peers (95% CI: -26.9 to -16.5; p < 0.0001). Caregivers scored 13.64 points lower than controls (95% CI: -19.21 to -8.06; p < 0.0001). Emotional and social domains were most affected. Higher Human Development Index was associated with better QoL. QoL is markedly reduced in individuals with ASD and their caregivers, underscoring the need for multidimensional and equitable interventions.

Lien vers le texte intégral (Open Access ou abonnement)

6. Cavalier HM, Volk HE, Kivumbi A, Liu M, Lyall K, Afanasyeva Y, Chen Y, Wang Y, Kannan K, Li Z, Sherris AR, Croen LA, Schmidt RJ, Bennett DH, Ames JL, Breton CV, Bakian AV, Bastain TM, Alkhouri NB, Eick SM, Goodrich JM, Schantz SL, O’Connor TG, Karagas MR, Herbstman JB, Trasande L, Ghassabian A. Prenatal Organophosphate Exposure and Autism-Related Traits in Children. JAMA Pediatr. 2026.

IMPORTANCE: Organophosphate pesticides (OPs) are established neurotoxicants; prenatal OP exposure is widespread in the US population, primarily through diet. Evidence linking prenatal OP exposure to autism spectrum disorder (ASD) and autism-related traits remains inconsistent. OBJECTIVE: To examine the association between prenatal OP exposure and autism-related traits in childhood and to assess potential differences in the association by child sex and maternal diet. DESIGN, SETTING, AND PARTICIPANTS: This pooled prospective cohort study used harmonized data from mother-child pairs from 20 study sites in the Environmental influences on Child Health Outcomes (ECHO) cohort. Participants were recruited from January 1997 through December 2019, and autism-related outcomes were assessed in children ages 2 to 19 years. Data analysis was performed from April 2024 and November 2025. EXPOSURE: Prenatal OP exposure using urinary concentrations of 3,5,6-trichloro-2-pyridinol (TCPy) and the sum of 6 dialkyl phosphate (ΣDAP) metabolites. MAIN OUTCOMES AND MEASURES: Parents reported autism-related traits in children using the Social Responsiveness Scale (SRS). Multivariable linear, quantile, and logistic regression models were used to estimate associations between biomarkers and SRS T scores overall and by child sex, adjusting for maternal sociodemographic and behavioral factors. We further adjusted for dietary factors in additional models and tested effect modification by maternal diet quality in exploratory analyses. RESULTS: Among 3339 mother-child pairs examined, mean (SD) maternal age was 31 (6) years, 1665 children (49.9%) were female, and the SRS was administered at a mean (SD) age of 6.4 (3.87) years. Crude or minimally adjusted models suggested inverse associations between prenatal TCPy or ΣDAP levels and SRS scores, but these associations attenuated and lost statistical significance after full adjustment (β per doubling of TCPy concentration [ng/mL] = -0.13; 95% CI, -0.57 to 0.31; β per doubling of ΣDAP [nmol/L] = -0.42; 95% CI, -1.03 to 0.19). Results were consistent across sex-stratified models, secondary analyses using quantile and logistic regressions, various sensitivity analyses, and after adjustments for dietary confounders. There was no evidence of effect modification by diet quality. CONCLUSIONS AND RELEVANCE: In this large, prospective cohort study among the ECHO cohort, prenatal OP exposure was not associated with autism-related traits in childhood. Given the known adverse impacts of pesticides in other contexts, further research should consider interactions with other factors, genetic susceptibility, higher exposures, or combined chemical mixtures.

Lien vers le texte intégral (Open Access ou abonnement)

7. Gardinal M, Rodacki ALF, Pavão SL. The Impact of Sensory Manipulation on the Limit of Stability of Children and Adolescents with Autism Spectrum Disorder. Phys Occup Ther Pediatr. 2026: 1-18.

AIMS: To compare the limit of stability (LoS) in a group of children and adolescents with autism spectrum disorder (ASDG) and with typical development (TDG), under different sensory conditions. METHODS: Fifty-nine participants (ASDG = 28; TDG = 31) performed LoS test under four sensory conditions: eyes_open/stable; eyes_closed/stable; eyes_open/unstable; eyes_closed/unstable. Maximum CoP displacements and velocities were tracked in the anterior/posterior and medial/lateral directions. Statistical analysis included a repeated-measures ANOVA (p < 0.05). RESULTS: The groups showed similar CoP displacement in the anterior and mediolateral directions. The ASDG showed greater CoP velocities in anterior-posterior and mediolateral directions. The groups showed greater anterior and mediolateral CoP displacement and faster oscillation with an unstable base of support than with a stable one, regardless of the visual manipulation. The absence of visual information associated with an unstable base of support increased CoP velocity and reduced the anterior displacement. Under stable conditions, the absence of visual information reduced CoP during anterior displacement. CONCLUSION: The ASDG exhibited reduced postural control during the test, as evidenced by greater AP and ML CoP velocities. Both groups were similarly affected when visual and somatosensory information was manipulated. Regardless of the visual information, the unstable surface resulted in increased displacement and velocity of CoP.

Lien vers le texte intégral (Open Access ou abonnement)

8. Hasanain RS, Al-Kuraishy HM, Shokr MM, Kafy S, Alshaikh ABA, Mogharbel H, Kashgari FA, Sebghatallah A, Batiha GE. Prenatal paracetamol exposure and autism spectrum disorder: A critical review of proposed mechanisms, epidemiological evidence, and causal inference. Neurotoxicology. 2026; 116: 103555.

Autism Spectrum Disorder (ASD) is a complex neurodevelopmental condition characterized by persistent deficits in social communication and restricted, repetitive patterns of behavior. Recent claims linking prenatal exposure to paracetamol (acetaminophen), the most commonly analgesic during pregnancy, to increased risk of ASD. This review aims to critically evaluate the validity of these associations, clarify potential biological mechanisms, and provide a balanced, evidence-based perspective to inform clinical practice. Although several observational cohort studies report statistical associations between in utero paracetamol exposure and ASD, these findings are often limited by confounding. Maternal conditions necessitating pain treatment, such as infection or fever, are themselves established risk factors for adverse neurodevelopmental outcomes. Greater emphasis here is therefore placed on robust study designs, particularly sibling-comparison analyses, which account for shared genetic and environmental influences. These epidemiological studies control for unmeasured confounders and consistently demonstrate attenuation or absence of previously reported associations, suggesting that paracetamol exposure is unlikely to be causative. This review also examines proposed mechanistic pathways, including mitochondrial dysfunction and inhibition of ribonucleotide reductase, but finds insufficient evidence to support a clinically meaningful effect in humans. Given the known risks of alternative therapies, particularly non-steroidal anti-inflammatory drugs during pregnancy, paracetamol remains the recommended first-line treatment for pain and fever. Overall, current evidence does not support a significant increase in ASD risk, and clinical guidelines should remain unchanged.

Lien vers le texte intégral (Open Access ou abonnement)

9. Hosseinpoor S, Zali H, Zohrevand H, Mirmotalebisohi SA, Khodagholi F, Bazrgar M, Asadi S, Ahmadiani A. An integrated systems biology and machine learning framework for identifying potential biomarkers and pathways in autism spectrum disorder. PLoS One. 2026; 21(8): e0355984.

BACKGROUND: Autism spectrum disorders (ASD) are a group of neurodevelopmental disorders whose underlying molecular mechanisms and biological processes remain incompletely understood. In this study, we used a multi-layered systems biology approach to prioritize candidate genes and regulatory factors associated with ASD. METHOD: Gene expression data from peripheral blood samples were obtained from the Gene Expression Omnibus (GEO) database (GSE18123). Using analyses performed in R software, differentially expressed genes (DEGs) in patients with ASD were identified (p-value < 0.05 and |log2FC| > 0.5). These DEGs were used to perform weighted gene co-expression network analysis (WGCNA) and construct a protein-protein interaction (PPI) network. By integrating the results of these network analyses with feature selection techniques (LASSO and random forest feature importance), candidate genes associated with ASD were prioritized and evaluated using qRT-PCR in the valproic acid (VPA)-induced rat model of autism. Furthermore, a gene regulatory network (GRN) was constructed to identify the regulatory factors associated with DEGs. RESULT: TLR8 and CASP4 were prioritized as candidate genes that may be associated with ASD, because they were located within the co-expression module that showed the strongest correlation with ASD, were identified as key nodes of the PPI network, and were selected by feature selection algorithms. Our experimental validation showed increased expression of TLR8 and CASP4 in the autism model compared with controls; TLR8 was upregulated in both the hippocampus and peripheral blood, whereas CASP4 was upregulated only in the hippocampus. Furthermore, GRN analysis identified miR-891b and miR-627-3p as potential regulators of TLR8, and miR-26b-5p as associated with CASP4. CONCLUSION: These findings indicate that CASP4 and TLR8, together with their associated regulatory miRNAs, may represent promising biomarkers and potential therapeutic targets for future ASD research and contribute to a better understanding of the pathophysiological mechanisms underlying ASD.

Lien vers le texte intégral (Open Access ou abonnement)

10. Khalil W, Al-Dalabeeh EA, Zihlif M. Association of HTR2A (rs6313) gene polymorphism with autism spectrum disorder in Jordanian children: a case-control study. Drug Metab Pers Ther. 2026.

OBJECTIVES: Autism spectrum disorder (ASD) is a complex neurodevelopmental condition with a significant genetic component, often linked to disruptions in the serotonergic system. The HTR2A gene, specifically the rs6313 (102T>C) polymorphism, is a primary candidate for investigating ASD susceptibility. The aim of this study is to investigate the association of rs6313 polymorphism with susceptibility to ASD in the Jordanian population. METHODS: In this case-control study, 99 Jordanian children with ASD and 109 neurotypical controls were genotyped using PCR-RFLP. Genotype and allele frequencies were analyzed under multiple genetic models. RESULTS: No statistically significant differences were found between cases and controls regarding genotype (p=0.54) or allele frequencies (p=0.3284). The distribution adhered to Hardy-Weinberg equilibrium in both groups. CONCLUSIONS: Our findings suggest no significant association between the HTR2A rs6313 and ASD susceptibility in the Jordanian population. These results emphasize the need for larger, multi-marker studies to account for regional genetic diversity.

Lien vers le texte intégral (Open Access ou abonnement)

11. Kim J, Yang S. Identifying Distinct Depression Trajectories in Korean Parents of Individuals With Intellectual Disability or Autism Spectrum Disorder: A Growth Mixture Modeling Analysis. J Autism Dev Disord. 2026.

PURPOSE: This study examined heterogeneous longitudinal patterns of depression among parents caring for individuals with intellectual disability (ID) or autism spectrum disorder (ASD), identifying distinct trajectory classes and their associated baseline predictors and distal psychosocial correlates. METHODS: Growth mixture modeling (GMM) was applied to five waves of annual data (2018 – 2022) from the Korean Disability and Life Dynamics Panel, comprising 680 parents of individuals with ID or ASD. RESULTS: Three distinct depression trajectory classes emerged, with moderate classification certainty (entropy = .58). Nearly one-third of parents (34.53%; High-Level Persistent Class) maintained persistently elevated self-reported depressive symptoms throughout the five-year period. An additional 15.44% (Moderate-Level Stable Class) showed moderately stable depression with no significant change. The largest group (50.03%; Low-Level Decreasing Class) exhibited low initial depression with a gradual decline over time. Trajectory class membership was significantly associated with child age, disability type, severe economic deprivation, and parental experience of discrimination. At Year 5, the High-Level Persistent Class demonstrated significantly less favorable distal psychosocial correlates across all domains, particularly in self-esteem and family functioning. CONCLUSION: These findings demonstrate that parental depression is not a uniform experience but follows qualitatively distinct developmental trajectories, each linked to a distinct pattern of baseline predictors and distal psychosocial correlates. A tiered support model addressing distinct risk profiles may be warranted, rather than a uniform approach applied across this heterogeneous population.

Lien vers le texte intégral (Open Access ou abonnement)

12. Knudsen LV, Vafaee MS, Michel TM. Letter to the Editor: Autism and Cortical Thickness Deviation From Neurotypical Controls: Evidence for a Spatial Association With Serotonin Receptors. Autism Res. 2026: e70358.

Lien vers le texte intégral (Open Access ou abonnement)

13. Kunze M, Dueñas AD, Miranda M, Muñoz Lavanderos A, Orendain Soto C. Culturally Responsive Adaptations: A Virtual Caregiver-mediated Intervention for Spanish-Speaking Families and Children with Autism. Behav Modif. 2026: 1454455261473252.

Using a surface adaptation, this single-case experimental study evaluated Promoting Reciprocal Relationships with Flexibility, Coaching, and Teaching (PRRFCT Match) for Spanish-speaking families whose children have an autism diagnosis. PRRFCT Match is a caregiver-mediated, naturalistic developmental intervention package that has been previously tested in 2 single-case experiments involving a total of 16 families. This iteration culturally and linguistically adapted the intervention package for use with four transborder, Spanish-speaking families, who were matched with Spanish-speaking coaches. The Spanish-speaking caregivers represent families whose home language or preferred language is Spanish. Using a multiple baseline design, data from caregiver and child behavior change were analyzed via visual analysis, non-overlap, and standard mean difference. This study shows promise for surface adaptation of the PRRFCT Match intervention package. Recommendations and considerations for cultural and linguistic adaptations of interventions that meet the needs of individual families are discussed. Important Considerations When Changing Intervention Programs Based on Culture and Language: An Online Program for Spanish-speaking Families and Their Children with AutismThis study adapted a program called Promoting Reciprocal Relationships with Flexibility, Coaching, and Teaching (PRRFCT Match), which has been used with 16 English-speaking families. The goal of this study was to redesign the PRRFCT Match program for use with Spanish-speaking families who have a child diagnosed with autism. The redesign of PRRFCT Match resulted in a program in which the caregiver and child interact while the caregiver receives coaching in Spanish from a trained therapist via an online communication system. The redesigned PRRFCT Match-Spanish better fits the cultural and language needs of families who speak Spanish. The researchers tested this with four families using a method called multiple baseline design, which helps track changes in caregiver and child behavior over time. The program showed potential to be effective when adapted for these families. The study includes suggestions for adjusting similar programs in the future to better meet families’ unique needs. eng.

Lien vers le texte intégral (Open Access ou abonnement)

14. Wagner VA, Maples S, Thapa R, Wimberly J, Pettijohn E, Varallo A, Rais M, Sutley-Koury S, Hickmott P, Ethell IM. Hippocampal Astrocytes Impact Postnatal Development of Inhibitory Connections, Parvalbumin Levels, Social, and Spatial Navigation Behaviors in a Mouse Model of Fragile X Syndrome. J Neurochem. 2026; 170(8): e70541.

Fragile X Syndrome (FXS) is a leading genetic cause of autism-like symptoms and intellectual disability, resulting from epigenetic silencing of the Fragile X messenger ribonucleoprotein (Fmr1) gene. Recent observations in FXS models suggest abnormal GABAergic signaling and excitation/inhibition imbalance may underlie the pathophysiology of FXS. As most studies have focused on neuronal mechanisms, the role of astrocytes in mediating defective inhibition in FXS is largely unknown. Our previous study showed the effects of astrocyte-specific Fmr1 conditional knockout (cKO) on cortical inhibitory circuit development using EEGs that were attributed to excess GABA synthesis by Fmr1 KO astrocytes. As the hippocampus plays an important role in spatial learning and social behaviors that are altered in FXS, in this study we focused on dissecting the mechanism of abnormal inhibition in the CA1 hippocampus using slice electrophysiology. While we observed a reduction in the expression of synaptic GABA(A) receptor subunits and overall density of perisomatic GABAergic synapses in cKO, the amplitude of spontaneous inhibitory postsynaptic currents (sIPSCs) was enhanced in pyramidal cells. In contrast to changes in phasic inhibition, astrocyte-specific cKO did not affect tonic inhibition in pyramidal cells or the expression of extrasynaptic GABA(A) receptors, which were both impaired in global KO. Our study suggests that elevated levels of extracellular GABA due to abnormal GABA transport in cKO astrocytes may contribute to the enhanced power of sIPSCs and potentially affect parvalbumin (PV) cell activity. Acute inhibition of GABA transport in astrocytes enhanced PV expression and improved spatial memory and socialization in cKO mice. Our work supports astrocytes as key players in the development and regulation of hippocampal inhibitory circuits in FXS, potentially through GABA transport.

Lien vers le texte intégral (Open Access ou abonnement)

15. Yiwen W, Junying Y, Dengna Z. Pathogenic Genetic Variants, Comorbid Autism and Adaptive Developmental Quotient as Independent Predictors of Intellectual Disability in Children With Global Developmental Delay: An Interpretable Machine Learning Model With Calibrated Risk Estimation. J Intellect Disabil Res. 2026.

BACKGROUND: Global developmental delay (GDD) frequently precedes intellectual disability (ID), but no validated multivariable prognostic tool exists to support individualised counselling during the initial diagnostic work-up. Existing risk indicators are typically considered in isolation, and their joint contribution within an interpretable predictive framework remains uncertain. METHODS: We retrospectively analysed 2453 children diagnosed with GDD between January 2014 and December 2023 at a provincial tertiary children’s rehabilitation centre, all followed to a minimum age of 60 months. Twenty-eight candidate predictors covering perinatal, developmental, neuroimaging, electrophysiological, genetic and comorbidity domains were retained after multiple imputation, multicollinearity screening with random-forest importance protection and standardisation. Five algorithms-L2- and L1-regularised logistic regression, random forest, XGBoost and LightGBM-were trained on a stratified 70% training partition with class-weight rebalancing; no synthetic minority over-sampling was applied. Probabilities from the L2 model were post hoc recalibrated by Platt scaling. We evaluated discrimination, calibration (slope and intercept after Platt scaling), Brier score and net benefit on the held-out 30% test set, with 1000 bootstrap confidence intervals. Sensitivity analyses excluded post-baseline candidate predictors, and we benchmarked the full model against parsimonious one-, three- and five-feature regressions. RESULTS: Of the cohort, 2037 children (83.0%) progressed to ID, reflecting the referral profile of a tertiary centre. The Platt-calibrated L2 logistic regression achieved an AUC of 0.783 (95% CI 0.735-0.828) with calibration slope 1.01 and intercept -0.005, and a Brier score of 0.113 (95% CI 0.097-0.130). All five algorithms performed within a 0.015 AUC band. The strongest independent risk factors were pathogenic genetic variant pathogenicity (OR 2.13, 95% CI 1.86-2.44), comorbid autism spectrum disorder (OR 1.69, 95% CI 1.41-2.03) and EEG epileptiform discharges (OR 1.43, 95% CI 1.18-1.74); higher Gesell adaptive developmental quotient was the strongest protective factor (OR 0.45 per standardised unit, 95% CI 0.34-0.60). At the Youden-optimal threshold of 0.85, sensitivity, specificity, positive and negative predictive values were 66.1%, 76.8%, 93.3% and 31.7%, respectively. Removal of early intensive intervention from the model lowered AUC by only 0.012 (95% CI - 0.002 to 0.026), indicating that retrospective treatment information was not the primary driver of model performance. A parsimonious five-feature model achieved AUC 0.762, recovering most of the discriminative signal. Discrimination was robust to strict exclusion of post-baseline predictors (AUC 0.769), to complete-case analysis (0.839) and to restriction to genetically tested children (0.864). CONCLUSIONS: Among children referred to a tertiary centre with GDD, an interpretable, well-calibrated logistic regression model integrating routinely available clinical, neurophysiological and genetic data stratifies the risk of progression to ID with moderate discrimination and high positive predictive value. The model is suitable for high-risk triage; its application to community populations will require refitting and external validation. The accompanying nomogram and SHAP explanations support clinician-facing risk communication.

Lien vers le texte intégral (Open Access ou abonnement)