Autism Diagnostic Observation Schedule Module Research Articles

Quantitative trait locus analysis for endophenotypes reveals genetic substrates of core symptom domains and neurocognitive function in autism spectrum disorder

Autism spectrum disorder (ASD) represents a heterogeneous group of neurodevelopmental disorders and is largely attributable to genetic risk factors. Phenotypic and genetic heterogeneity of ASD have been well-recognized; however, genetic substrates for endophenotypes that constitute phenotypic heterogeneity are not yet known. In the present study, we compiled data from the Autism Genetic Resource Exchange, which contains the demographic and detailed phenotype information of 11,961 individuals. Notably, the whole-genome sequencing data available from MSSNG and iHART for 3833 individuals in this dataset was used to perform an endophenotype-wide association study. Using a linear mixed model, genome-wide association analyses were performed for 29 endophenotype scores and 0.58 million common variants with variant allele frequency ≥ 5%. We discovered significant associations between 9 genetic variants and 6 endophenotype scores comprising neurocognitive development and severity scores for core symptoms of ASD at a significance threshold of p < 5 × 10–7. Of note, the Stereotyped Behaviors and Restricted Interests total score in Autism Diagnostic Observation Schedule Module 3 was significantly associated with multiple variants in the VPS13B gene, a causal gene for Cohen syndrome and a candidate gene for syndromic ASD. Our findings yielded loci with small effect sizes due to the moderate sample size and, thus, require validation in another cohort. Nonetheless, our endophenotype-wide association analysis extends previous candidate gene discovery in the context of genotype and endophenotype association. As a result, these candidate genes may be responsible for specific traits that constitute core symptoms and neurocognitive function of ASD rather than the disorder itself.

Using machine learning to improve diagnostic assessment of ASD in the light of specific differential and co-occurring diagnoses.

Diagnostic assessment of ASD requires substantial clinical experience and is particularly difficult in the context of other disorders with behavioral symptoms in the domain of social interaction and communication. Observation measures such as the Autism Diagnostic Observation Schedule (ADOS) do not take into account such co-occurring disorders. We used a well-characterized clinical sample of individuals (n = 1,251) that had received detailed outpatient evaluation for the presence of an ASD diagnosis (n = 481) and covered a range of additional overlapping diagnoses, including anxiety-related disorders (ANX, n = 122), ADHD (n = 439), and conduct disorder (CD, n = 194). We focused on ADOS module 3, covering the age range with particular high prevalence of such differential diagnoses. We used machine learning (ML) and trained random forest models on ADOS single item scores to predict a clinical best-estimate diagnosis of ASD in the context of these differential diagnoses (ASD vs. ANX, ASD vs. ADHD, ASD vs. CD), in the context of co-occurring ADHD, and an unspecific model using all available data. We employed nested cross-validation for an unbiased estimate of classification performance and made available a Webapp to showcase the results and feasibility for translation into clinical practice. We obtained very good overall sensitivity (0.89-0.94) and specificity (0.87-0.89). In particular for individuals with less severe symptoms, our models showed increases of up to 35% in sensitivity or specificity. Furthermore, we analyzed item importance profiles of the ANX, ADHD, and CD models in comparison with the unspecific model revealing distinct patterns of importance for specific ADOS items with respect to differential diagnoses. ML-based diagnostic classification may improve clinical decisions by utilizing the full range of information from detailed diagnostic observation instruments such as the ADOS. Importantly, this strategy might be of particular relevance for older children with less severe symptoms for whom the diagnostic decision is often particularly difficult.

How Does an Observational Assessment Adapted for Online Delivery Perform Compared to an In-Person Assessment? Learning From the National Autism Service for Adults

AimsThe National Autism Service for Adults receives over 600 referrals annually and with an extensive waitlist, COVID-19 restrictions on in-person assessments were a challenge for service delivery. We aimed to adapt the Autism Diagnostic Observation Schedule (ADOS) for online delivery and investigate whether it is comparable to the in-person ADOS in predicting Autism Spectrum Disorder (ASD) diagnostic outcome. We also aimed to obtain qualitative feedback from service users and clinicians regarding experiences of the online ADOS.MethodsA working group of staff who administer ADOS and representatives from psychiatry, psychology and management reached consensus that an online version of ADOS module 4 was feasible based on experience that a lot information required for coding is obtained verbally and some tasks were adaptable for online delivery. After the pilot, it was agreed all algorithm items could be coded except ‘unusual eye-contact’. Subsequently, 163 service users attended an online ADOS between August 2020 and February 2021. A matched-comparison group consisted of 198 service users seen for an in-person ADOS between May 2014 and February 2020. Algorithm scores were recorded and ASD diagnosis was made by a trained clinician. Qualitative feedback regarding the online ADOS was collected from 46 service users and 11 clinicians.ResultsThe working group agreed the online and in-person ADOS were closely matched regarding administration and coding. Mean scores for service users who received an ASD diagnosis were comparable for the online and in-person ADOS groups (7 and 8 respectively). This was also shown for those who were not diagnosed with ASD (3 and 4 respectively). A two-sample t-test showed no significant difference in total scores between the online and in-person ADOS (p = 0.38). Qualitative feedback suggested good service user and clinician satisfaction; only 27% of service users indicated they would have preferred an in-person assessment; 88% of clinicians reported there were gains from offering an online alternative. Although the online and in-person ADOS perform similarly, clinicians reported relying more on qualitative reports over scores from the online version to inform diagnostic decision.ConclusionTo our knowledge, this is the first study to examine using an online ADOS within an adult diagnostic service. Due to its comparable performance, the online-ADOS is a viable alternative option for service delivery when in-person assessments are not possible. As this clinic group has high rates of comorbid mental health difficulties, the applicability of online assessments could generalise to other services and have an impact beyond the pandemic.

Subdimensions of social-communication behavior in autism-A replication study.

In order to identify more refined dimensions of social-communication impairments in autism spectrum disorder (ASD) a previous study applied exploratory and confirmatory factor analyses to diagnostic algorithm scores of the autism diagnostic observation schedule (ADOS), Module 3. A three-factor model consisting of repetitive behaviors, impairments in 'Basic Social-Communication' and in 'Interaction quality' (IQ) was established and confirmed. The current study aimed to replicate this model in an independent sample. To advance our understanding of the latent structure of social communication deficits, previous work was complemented by a probabilistic approach. Participants (N=1363) included verbally fluent children and young adults, diagnosed as ASD or non-ASD based on "gold standard" best-estimate clinical diagnosis. Confirmatory factor analysis examined the factor structure of algorithm items from the ADOS Module 3 and correlations with individual characteristics (cognitive abilities, age) were analyzed. Linear Regressions were used to test the contribution of each latent factor to the prediction of an ASD diagnosis. To tackle large inter-correlations of the latent factors, a Bayesian exploratory factor analysis (BEFA) was applied. Results confirmed the previously reported observation of three latent dimensions in the ADOS algorithm reflecting 'Restricted, Repetitive Behaviors', 'Basic Social-Communication' behaviors and 'Interaction Quality'. All three dimensions contributed independently and additively to the prediction of an ASD diagnosis. By replicating previous findings in a large clinical sample our results contribute to further conceptualize the social-communication impairments in ASD as two dimensional.

Identifying predictive features of autism spectrum disorders in a clinical sample of adolescents and adults using machine learning

Diagnosing autism spectrum disorders (ASD) is a complicated, time-consuming process which is particularly challenging in older individuals. One of the most widely used behavioral diagnostic tools is the Autism Diagnostic Observation Schedule (ADOS). Previous work using machine learning techniques suggested that ASD detection in children can be achieved with substantially fewer items than the original ADOS. Here, we expand on this work with a specific focus on adolescents and adults as assessed with the ADOS Module 4. We used a machine learning algorithm (support vector machine) to examine whether ASD detection can be improved by identifying a subset of behavioral features from the ADOS Module 4 in a routine clinical sample of N = 673 high-functioning adolescents and adults with ASD (n = 385) and individuals with suspected ASD but other best-estimate or no psychiatric diagnoses (n = 288). We identified reduced subsets of 5 behavioral features for the whole sample as well as age subgroups (adolescents vs. adults) that showed good specificity and sensitivity and reached performance close to that of the existing ADOS algorithm and the full ADOS, with no significant differences in overall performance. These results may help to improve the complicated diagnostic process of ASD by encouraging future efforts to develop novel diagnostic instruments for ASD detection based on the identified constructs as well as aiding clinicians in the difficult question of differential diagnosis.

L’ADI-R et l’ADOS face au diagnostic différentiel des troubles du spectre autistique : intérêts, limites et ouvertures

IntroductionCurrent developments in international public health are leading qualified authorities to release clinical practice guidelines for Autism Spectrum Disorders (ASD) assessment and diagnosis. Such documents incorporate procedures that rely on the Autism Diagnostic Interview-Revised (ADI-R) and the Autism Diagnostic Observation Schedule (ADOS) which are considered to be the “gold standard” assessment measures in the evaluation of ASD. Although these tools do prove their effectiveness in the evaluation of autistic symptomatology, they nevertheless stumble whenever the situation becomes more complex and reveal differential diagnostic issues in infantile and adult autism cases. This differential diagnostic issue remains significant in the clinical practice of daily life and has strong implications for the course of therapeutic treatment. ObjectivesOur objective is to underline and nuance the metrological qualities of the ADI-R and the ADOS in the differential diagnosis of autistic disorders by presenting a synthesis of recent studies, thus supporting the interest of maintaining an open debate on diagnostic practices. MethodsOur selective review of the recent literature focuses on studies that confront the ADI-R and the ADOS — used either independently or in a combination — with various differential issues in adulthood and childhood autism in order to highlight their qualities and limits. ResultsThe ADI-R is a semi-structured interview applied by trained examiners and applied to relatives and/or caregivers who collect developmental information about the patient's first years of life on a variety of behaviors and skills. It is therefore relatively dependent on the availability and personal bias of the interviewed third parties. Metric features highlighted by the reviewed studies straightforwardly reveal a certain effectiveness of the tool and a good discrimination of childhood disorders. At the same time, the tool's discriminatory capacity seems insufficient when applied to toddlers, very young children, and adults alike. The latter seems particularly true when it comes to differentiating between autistic and schizophrenic spectrum disorders. The ADOS is a semi-structured standardized observation assessment tool that has experienced several successive developments, such as optimizations concerning the sensitivity of its cutoffs. Many works have contributed to building such a tool with reliable and solid metric qualities which nevertheless retain important biases such as the subjectivity of the caregiver or the evaluator during the scoring process. For assessments of autistic children the tool still has a good diagnostic validity but seems to retain cases of incorrect diagnosis of ASD (false positives). In other words, disorders or developmental disabilities of some children and adolescents could not be distinguished from ASD when relying on this test alone. The ADOS Module 4, designed for the diagnosis of adolescents and adults with fluent speech, has undergone less updating. This revisited algorithm has metrological qualities useful for clinicians and remains one of the few available tools for this population. Unfortunately, its diagnostic accuracy is lower when applied to women, the elderly, people with personality disorders or higher intellectual abilities, or for the discrimination between ASD and schizophrenia. Overall, scores from these two instruments bring strong evidence of their usefulness in the diagnostic process of ASD, provided that they are used with caution and a critical clinical perspective, and only as a secondary technical support. Their use in combination is effective since they are complementary and compensate for each other's limitations. However, their globalized hegemony as “gold-standard” tools constitutes a setback insofar as it constrains the diagnosis of ASD to a set of stereotyped items. The latter in turn sets a normative model of autism that excludes other phenotypic forms, especially in the case of women and the elderly. Finally, the discrimination between autism and psychosis for children seems to remain an insoluble task even for the ADI-R/ADOS combination. ConclusionsThe problematics of differential diagnosis remain critical for clinical approaches to autism. Therefore, formalizations of the diagnostic procedures must be able to remain open-minded and accompanied by a creative clinical approach, especially in the case of complex situations that are not soluble by means of conventional diagnostic tools. One possibility may lie in the deepening of the phenomenological approach to autism as an attempt to model the subjective phenomena of autistic subjects and thus operationalize elements that serve the diagnostic process. In the same way, a psychodynamic epistemology could help clinicians to go beyond the consideration of observable behaviors and scores, introducing a psychoanalytic point of view that interfaces objective behaviors with the individual's dynamic intrapsychic functioning. This project could be articulated with projective methodologies — notably the Rorschach test — which respects the needs for standardization and quantification of conventionally used diagnostic tools.

Autism Spectrum Symptomatology in Children with Williams Syndrome Who Have Phrase Speech or Fluent Language.

To characterize autism spectrum-related symptomatology in children with Williams syndrome (WS) with phrase speech or fluent language, the Autism Diagnostic Observation Schedule Module 2 or 3 was administered. The cutoff for autism spectrum was met by 35% (14/40) who completed Module 2 and 30% (18/60) who completed Module 3. Similarities and differences in socio-communicative strengths and weaknesses as a function of language ability were identified. Symptom severity was negatively associated with IQ for participants with phrase speech but not for those with fluent language. The findings suggest an elevated risk of ASD for individuals with WS relative to the general population and contribute to a more nuanced sense of the socio-communicative functioning of children with WS.

Sparsifying machine learning models identify stable subsets of predictive features for behavioral detection of autism

BackgroundAutism spectrum disorder (ASD) diagnosis can be delayed due in part to the time required for administration of standard exams, such as the Autism Diagnostic Observation Schedule (ADOS). Shorter and potentially mobilized approaches would help to alleviate bottlenecks in the healthcare system. Previous work using machine learning suggested that a subset of the behaviors measured by ADOS can achieve clinically acceptable levels of accuracy. Here we expand on this initial work to build sparse models that have higher potential to generalize to the clinical population.MethodsWe assembled a collection of score sheets for two ADOS modules, one for children with phrased speech (Module 2; 1319 ASD cases, 70 controls) and the other for children with verbal fluency (Module 3; 2870 ASD cases, 273 controls). We used sparsity/parsimony enforcing regularization techniques in a nested cross validation grid search to select features for 17 unique supervised learning models, encoding missing values as additional indicator features. We augmented our feature sets with gender and age to train minimal and interpretable classifiers capable of robust detection of ASD from non-ASD.ResultsBy applying 17 unique supervised learning methods across 5 classification families tuned for sparse use of features and to be within 1 standard error of the optimal model, we find reduced sets of 10 and 5 features used in a majority of models. We tested the performance of the most interpretable of these sparse models, including Logistic Regression with L2 regularization or Linear SVM with L1 regularization. We obtained an area under the ROC curve of 0.95 for ADOS Module 3 and 0.93 for ADOS Module 2 with less than or equal to 10 features.ConclusionsThe resulting models provide improved stability over previous machine learning efforts to minimize the time complexity of autism detection due to regularization and a small parameter space. These robustness techniques yield classifiers that are sparse, interpretable and that have potential to generalize to alternative modes of autism screening, diagnosis and monitoring, possibly including analysis of short home videos.

Coherence and content of relating emotions to life events in autism spectrum disorder and typical development: a cross-sectional age study.

Understanding one's own emotions is an important part of social-emotional development in early childhood. Few studies have looked at the ability of children with autism spectrum disorder (ASD) to relate their own emotions to previous life events. Our previous study showed that the description of events that elicited specific emotions is qualitatively and quantitatively different in ASD in comparison to typically developing (TD) pre-adolescents. The current study evaluated differences in coherence and content of responses to questions on emotions in ASD and TD in two age groups. The evaluation was based on the section on Emotions of the Autism Diagnostic Observation Schedule Module 3 test. The study included 96 boys, 48 diagnosed with ASD (IQ≥85) and 48 TD children, divided into younger (6:0-8:0y) and older (8:2-11:0y) groups. Young TD children were able to give coherent responses to questions on experiences that evoked basic emotions. Children with ASD gave fewer coherent responses and more 'no response' and 'odd' responses across the examined age range. Only in the TD group was the level of vocabulary associated with the number of coherent statements. TD children gave more responses with content related to interpersonal relationships, self-awareness and social events than children with ASD. Deficits in coherence and content of responses to questions on emotions related to previous life events derive from the core deficits of ASD. The significant quantitative and qualitative gap that exists between ASD and TD may be useful during the diagnostic process of ASD in childhood.

Understanding One's Own Emotions in Cognitively-Able Preadolescents with Autism Spectrum Disorder.

There are still no straightforward answers as to whether understanding one's own emotions is impaired in autism spectrum disorder (ASD). This study evaluated the perception of one's own different emotions, based on the relevant section of the Autism Diagnostic Observation Schedule Module 3 test. Forty boys, aged 8-11years, 20 diagnosed with ASD (IQ≥85) and 20 typically developing children were included. Description of events that elicited specific emotions in ASD was characterized by more 'odd' statements and 'no responses' and less use of content related to 'social situations', 'interpersonal' and 'self-awareness'. More 'no responses' and odd statements were associated with the severity of ASD symptoms. Clinicians should be aware of these differentiating factors during the diagnostic process of ASD.

The Autism Diagnostic Observation Schedule, Module 4: Application of the Revised Algorithms in an Independent, Well-Defined, Dutch Sample (n = 93).

This study examined the discriminative ability of the revised Autism Diagnostic Observation Schedule module 4 algorithm (Hus and Lord in J Autism Dev Disord 44(8):1996–2012, 2014) in 93 Dutch males with Autism Spectrum Disorder (ASD), schizophrenia, psychopathy or controls. Discriminative ability of the revised algorithm ASD cut-off resembled the original algorithm ASD cut-off: highly specific for psychopathy and controls, lower sensitivity than Hus and Lord (2014; i.e. ASD .61, AD .53). The revised algorithm AD cut-off improved sensitivity over the original algorithm. Discriminating ASD from schizophrenia was still challenging, but the better-balanced sensitivity (.53) and specificity (.78) of the revised algorithm AD cut-off may aide clinicians’ differential diagnosis. Findings support using the revised algorithm, being conceptually conform the other modules, thus improving comparability across the lifespan.

Female Autism Phenotypes Investigated at Different Levels of Language and Developmental Abilities.

This study investigated the differences in clinical symptoms between females and males with autism spectrum disorder (ASD) across three verbal ability groups (nonverbal, phrase and fluent speech), based on which Autism Diagnostic Observation Schedule module was administered to 5723 individuals in four research datasets. In the Simons Simplex Collection and Autism Treatment Network, females with ASD and phrase or fluent speech had lower cognitive, adaptive, and social abilities than males. In the Autism Genetics Resource Exchange and the Autism Consortium, females with phrase or fluent speech had similar or better adaptive and social abilities than males. Females who were nonverbal had similar cognitive, adaptive, and social abilities as males. Population-based longitudinal studies of verbally fluent females with ASD are needed.

Do women with eating disorders who have social and flexibility difficulties really have autism? A case series.

BackgroundMany women with eating disorders (EDs) have social impairments and difficulties with flexibility. It is unclear to what extent these are manifestations of an underlying autism spectrum disorder (ASD); or whether they are instead the consequence of starvation, anxiety, low mood or obsessive compulsive disorder, all of which are highly prevalent in EDs. The resolution of this clinically and theoretically important uncertainty will require the use of gold-standard ASD assessment measures. To date these have not been employed in ED research. This case series is the first report of a well-validated, direct-observational measure of ASD, the Autism Diagnostic Observation Schedule (ADOS), being administered to women with EDs. We aimed to learn about the feasibility of the ADOS in this population, and to contribute to debates about whether a sub-group with EDs really have ASD.MethodsTen women (mean age = 26.4 years, range = 19 to 38 years) who had a suspected ASD due to social and flexibility difficulties and were receiving treatment for ED (seven anorexia, two ED not otherwise specified, one bulimia) at a specialist service (four inpatient, six outpatient) received an ADOS Module 4 assessment.ResultsAll 10 participants completed all activities of the ADOS Module 4. Five scored in the ASD range on the ADOS diagnostic algorithm. An additional two were judged likely to have ASD, even though they scored below the ADOS’s diagnostic threshold. This was on the basis of clinical observation, participant self-report and parent report. The seven women who we estimated to have ASD all reported autistic difficulties prior to the onset of their ED. They commonly described longstanding non-autistic neurodevelopmental problems, including dyslexia, dyspraxia and epilepsy. Only one had a childhood diagnosis of ASD.ConclusionsA substantial proportion of women with EDs who present with social and flexibility difficulties may have an unrecognised ASD, indicated by a constellation of autistic difficulties that appears to predate the onset of their eating problems. The ADOS is a useful component of an ASD assessment for adult women with ED.

Overlap With the Autism Spectrum in Young Children With Williams Syndrome

The sociocommunicative abnormalities of young children with Williams syndrome (WS) with limited language were compared with those of children with clinical diagnoses of autism, Pervasive Developmental Disorder-Not Otherwise Specified (PDD-NOS), or nonspectrum developmental disability. Participants were 30 children with WS and individually matched groups of participants with autism (n = 28), PDD-NOS (n = 17), and mixed etiology nonspectrum developmental disabilities (n = 16). The autism, PDD-NOS, and mixed etiology groups were matched individually to the children with WS for age, sex, and developmental level. All participants were administered the Autism Diagnostic Observation Schedule Module 1 and the Mullen Scales of Early Learning. As a group, children with WS with limited language showed fewer sociocommunicative abnormalities than children with autism, about the same level as children with PDD-NOS, and more abnormalities in reciprocity social interaction than participants in the mixed etiology group. Examination of the subgroup of participants with WS matched and compared with children with PDD-NOS indicated that half of the children showed fewer abnormalities than their individual matches with PDD-NOS, whereas half of the children with WS showed more abnormalities than their matches with PDD-NOS. Sociocommunicative difficulties are present for many children with WS and overlap with the autism spectrum. The results of this investigation suggest that these abnormalities are not accounted for by developmental delay alone, and care should be taken to avoid diagnostic overshadowing in young children with WS.