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

BACKGROUND: 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. METHOD: 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. RESULTS: 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. CONCLUSIONS: 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.

Mapping Research Domain Criteria using a transdiagnostic mini-RDoC assessment in mental disorders: a confirmatory factor analysis.

This study aimed to build on the relationship of well-established self-report and behavioral assessments to the latent constructs positive (PVS) and negative valence systems (NVS), cognitive systems (CS), and social processes (SP) of the Research Domain Criteria (RDoC) framework in a large transnosological population which cuts across DSM/ICD-10 disorder criteria categories. One thousand four hundred and thirty one participants (42.1% suffering from anxiety/fear-related, 18.2% from depressive, 7.9% from schizophrenia spectrum, 7.5% from bipolar, 3.4% from autism spectrum, 2.2% from other disorders, 18.4% healthy controls, and 0.2% with no diagnosis specified) recruited in studies within the German research network for mental disorders for the Phenotypic, Diagnostic and Clinical Domain Assessment Network Germany (PD-CAN) were examined with a Mini-RDoC-Assessment including behavioral and self-report measures. The respective data was analyzed with confirmatory factor analysis (CFA) to delineate the underlying latent RDoC-structure. A revised four-factor model reflecting the core domains positive and negative valence systems as well as cognitive systems and social processes showed a good fit across this sample and showed significantly better fit compared to a one factor solution. The connections between the domains PVS, NVS and SP could be substantiated, indicating a universal latent structure spanning across known nosological entities. This study is the first to give an impression on the latent structure and intercorrelations between four core Research Domain Criteria in a transnosological sample. We emphasize the possibility of using already existing and well validated self-report and behavioral measurements to capture aspects of the latent structure informed by the RDoC matrix.

Social and Nonsocial Autism Symptom Domains in Children and Adolescents with Autism Spectrum Disorder and Attention-Deficit/Hyperactivity Disorder: Insights into Their Symptomatological Interplay.

INTRODUCTION: Autism spectrum disorder (ASD) and attention-deficit/hyperactivity disorder (ADHD) share overlapping symptomatology, particularly with regard to social impairments (including peer relationship difficulties), and they frequently co-occur. However, the nature of their co-occurrence remains unclear. Therefore, the current study aimed to examine the nature of the transdiagnostic link between ASD and ADHD from a symptomatological point of view measured with the Autism Diagnostic Observation Schedule (ADOS Module 3) and the Autism Diagnostic Interview-Revised (ADI-R). METHODS: We analyzed the social and nonsocial ASD symptom domain scores from both diagnostic instruments in 4 clinically referred groups (i.e., ASD, ADHD, ASD + ADHD, and no psychiatric diagnosis) without other co-occurring mental disorders using a two-by-two full-factorial MANOVA design with the factors ASD (yes/no) and ADHD (yes/no). RESULTS: We found no ASD by ADHD interaction effects across all symptom domain scores of ADOS and ADI-R, except for ADOS imagination/creativity. There were only main effects of the factor ASD but no main effects of ADHD. Follow-up contrasts showed that exclusively, ASD had an impact on the measured symptomatology in case of co-occurring ASD + ADHD. CONCLUSION: Overall, the results support an additive model of the symptomatology across areas of communication, social interaction, and stereotyped behaviors and restricted interests in case of the co-occurrence of ASD and ADHD when assessed with ADOS/ADI-R. Thus, one can assume that the phenotypic overlap of ASD + ADHD may be less complicated than suspected - at least with regard to ASD symptomatology - and that in the presence of ADHD, ASD symptomatology is generally well measurable with best-practice diagnostic instruments.

Conceptualization of the latent structure of autism: further evidence and discussion of dimensional and hybrid models.

Autism spectrum disorder (ASD) might be conceptualized as an essentially dimensional, categorical, or hybrid model. Yet, current empirical studies are inconclusive and the latent structure of ASD has explicitly been examined only in a few studies. The aim of our study was to identify and discuss the latent model structure of behavioral symptoms related to ASD and to address the question of whether categories and/or dimensions best represent ASD symptoms. We included data of 2920 participants (1-72 years of age), evaluated with the Autism Diagnostic Observation Schedule (Modules 1-4). We applied latent class analysis, confirmatory factor analysis, and factor mixture modeling and evaluated the model fit by a combination of criteria. Based on the model selection criteria, the model fits, the interpretability as well as the clinical utility we conclude that the hybrid model serves best for conceptualization and assessment of ASD symptoms. It is both grounded in empirical evidence and in clinical usefulness, is in line with the current classification system (DSM-5) and has the potential of being more specific than the dimensional approach (decreasing false positive diagnoses).

[Pathological Demand Avoidance: Current State of Research and Critical Discussion]. / Pathological Demand Avoidance ­ aktueller Forschungsstand und kritische Diskussion.

Pathological Demand Avoidance: Current State of Research and Critical Discussion Abstract: Pathological demand avoidance (PDA) describes children who obsessively avoid any demand to a clinically relevant extent and is presently the subject of controversial discussion. Their behavior may be interpreted as an attempt to reduce anxiety by establishing security and predictability through rigid control of the environment as well as the demands and expectations of others. The symptoms are described in the context of autism spectrum disorder. This article reviews the current state of research and discusses the questionable validity of pathological demand avoidance as an independent diagnostic entity. It also addresses the impact of the behavior profile on development and treatment. This paper concludes that PDA is not a diagnostic entity nor a subtype of autism; rather, it is a behavior profile that can be associated with adverse illness progression and unfavorable outcomes. PDA is one feature in a complex model. We must consider not only the patient's characteristics but also those of the caregiver and their psychopathology. The reactions of the interaction partners as well as the treatment decisions play a key role play for the affected individuals. Substantial research is needed concerning the occurrence of the behavior profile PDA in diverse disorders, treatment options, and treatment responses.

[Autism spectrum disorder in childhood and adults: diagnosis and differential diagnoses]. / Autismus-Spektrum-Störungen im Kindes- und Erwachsenenalter: Diagnose und Differenzialdiagnosen.

Autistic disorders are summarized in DSM­5 under the term autism spectrum disorder (ASD) and are severe, lifelong, pervasive neurodevelopmental disorders. Core features manifested even in childhood are impairments in social interaction and communication as well as restricted and repetitive behavior. The intensity of symptoms, language and cognitive impairments vary but the majority of affected individuals have below average intelligence and 80% have at least one comorbid disorder. The diverse pathology and heterogeneity in phenotypes are caused by a complex genetic etiology, which is associated with a reduced synaptic plasticity of neural networks. The disorder is associated with a clearly reduced quality of life as well as a high familial burden. The differential diagnostics have a high relevance and the diagnosis should be carried out by specialized institutions. Behavioral therapeutic interventions are indicated.

[Blessing or curse? The World Wide Web as information source for autism and Asperger Syndrome]. / Segen oder Fluch? Das Internet als Informationsquelle über Autismus und Asperger-Syndrom.

Blessing or curse? The World Wide Web as information source for autism and Asperger Syndrome Abstract. Objective: The World Wide Web is today one of the most common methods used for obtaining health-related information, though the quality of the information is sometimes questionable. The present study addresses the quality of the information source internet and the resulting implications and discusses examples related to autism spectrum disorder. Method: We systematically evaluated 96 German websites, with the aim of estimating specific characteristic features, reliability of publications, presentation of information as well as overall website quality. We also analyzed the clinical implications of the presentations. Results: Only 39 % of the websites provided references to scientifically well-founded information, whereas advertisements were found on 53 % of websites. The greatest percentage of false information (17 %) was disseminated concerning therapy. 75 % of the websites provided incomplete information. Only 10 % of websites discussed the impairment or familial burden. The quality of information was insufficient on 30 %, poor on 41 %, and good on only 6 % of the websites. Conclusions: Similar to results available for English-language websites, the quality of German websites providing health-related information can be considered low. Implications concerning confirmation bias, stigma, overidentification, in-group/outgroup, contrast and snowball effects are discussed.

Complementary and alternative medicine use in adults with autism spectrum disorder in Germany: results from a multi-center survey.

BACKGROUND: Complementary and Alternative Medicine (CAM) is widely used both in the general population and for the treatment of somatic and psychiatric disorders. Studies on CAM use among patients with autism spectrum disorder (ASD) have so far only focused on children and adolescents. The aim of this study was to investigate patterns of CAM use among adults with ASD. METHODS: A questionnaire survey concerning current and lifetime use of CAM was distributed to adults with ASD between November 2015 and June 2016. Participants diagnosed by experienced clinicians using the current diagnostic gold standard were recruited from four ASD outpatient clinics in Germany. Questionnaire data was then linked to supplementary clinical data. RESULTS: The final sample consisted of 192 adults (response: 26.8%) with a mean age of 31.5 years (80% male; diagnoses: Asperger's syndrome (58%), childhood autism (27%), atypical autism (12%)). 45% of the respondents stated that they were currently using or had used at least one CAM modality in their life. Among the participants with lifetime CAM use, almost half had used two or more different types of CAM. Alternative medical systems (e.g. homeopathy, acupuncture) were most frequently used, followed by mind-body interventions (e.g. yoga, biofeedback, animal assisted therapy). Overall, 20% of respondents stated that they would like to try at least one listed CAM modality in the future. CONCLUSIONS: This is the first study on CAM use in adults with ASD, demonstrating considerable CAM use in this population. Given the popularity of CAM, patients should be informed about the effectiveness and potentially dangerous side effects of CAM treatments, as evidence for the majority of CAM methods in ASD is still limited.

Study protocol of the ASD-Net, the German research consortium for the study of Autism Spectrum Disorder across the lifespan: from a better etiological understanding, through valid diagnosis, to more effective health care.

BACKGROUND: Autism Spectrum Disorder (ASD) is a severe, lifelong neurodevelopmental disorder with early onset that places a heavy burden on affected individuals and their families. Due to the need for highly specialized health, educational and vocational services, ASD is a cost-intensive disorder, and strain on health care systems increases with increasing age of the affected individual. METHODS: The ASD-Net will study Germany's largest cohort of patients with ASD over the lifespan. By combining methodological expertise from all levels of clinical research, the ASD-Net will follow a translational approach necessary to identify neurobiological pathways of different phenotypes and their appropriate identification and treatment. The work of the ASD-Net will be organized into three clusters concentrating on diagnostics, therapy and health economics. In the diagnostic cluster, data from a large, well-characterized sample (N = 2568) will be analyzed to improve the efficiency of diagnostic procedures. Pattern classification methods (machine learning) will be used to identify algorithms for screening purposes. In a second step, the developed algorithm will be tested in an independent sample. In the therapy cluster, we will unravel how an ASD-specific social skills training with concomitant oxytocin administration can modulate behavior through neurobiological pathways. For the first time, we will characterize long-term effects of a social skills training combined with oxytocin treatment on behavioral and neurobiological phenotypes. Also acute effects of oxytocin will be investigated to delineate general and specific effects of additional oxytocin treatment in order to develop biologically plausible models for symptoms and successful therapeutic interventions in ASD. Finally, in the health economics cluster, we will assess service utilization and ASD-related costs in order to identify potential needs and cost savings specifically tailored to Germany. The ASD-Net has been established as part of the German Research Network for Mental Disorders, funded by the BMBF (German Federal Ministry of Education and Research). DISCUSSION: The highly integrated structure of the ASD-Net guarantees sustained collaboration of clinicians and researchers to alleviate individual distress, harm, and social disability of patients with ASD and reduce costs to the German health care system. TRIAL REGISTRATION: Both clinical trials of the ASD-Net are registered in the German Clinical Trials Register: DRKS00008952 (registered on August 4, 2015) and DRKS00010053 (registered on April 8, 2016).

Facial affect recognition performance and event-related potentials in violent and non-violent schizophrenia patients.

We investigated whether male inpatients with schizophrenia and a history of hands-on violent offences (forensic schizophrenic, FOS) are more impaired in emotion recognition than matched schizophrenia patients without any history of violence (general psychiatric schizophrenic, GPS). This should become apparent in performance in psychometry and in scalp event-related brain potentials (ERPs) evoked by pictures of facial affect. FOS and GPS (each n = 19) were matched concerning age, intelligence, comorbid addiction, medication and illness duration. FOS revealed significantly poorer affect recognition (AR) performance, especially of neutral and fear stimuli. Analysis of ERPs revealed a significant interaction of hemisphere, electrode position and group of the N250 component. Post hoc analysis of group effect showed significantly larger amplitudes in FOS at FC3. These results support the hypothesis that in FOS emotional faces are more salient and evoke higher arousal. Larger impairment in AR performance combined with higher salience and arousal may contribute to the occurrence of violent acts in schizophrenia patients.

Autism Spectrum Disorder and IQ - A Complex Interplay.

Autism spectrum disorder (ASD) is characterized as a very heterogeneous child-onset disorder, whose heterogeneity is partly determined by differences in intelligence quotient (IQ). Older epidemiological studies suggested that the IQ-related spectrum tends to be skewed to the left, i.e., a larger proportion of individuals with ASD have below average intelligence, while only few individuals with ASD may have an IQ above average. This picture changed over time with broadening the spectrum view. Within the present perspective article, we discuss discrepancies in IQ profiles between epidemiological and clinical studies and identify potential underlying aspects, for example, the influence of external factors such as sample biases or differences in availability of autism health services. Additionally, we discuss the validity and reciprocal influences of ASD diagnostics and IQ measurement. We put the impact of these factors for diagnostic as well as care and support situations of patients into perspective and want to encourage further research to contribute to the conceptualization of "autism" more comprehensively including the IQ as well as to examine broader (life) circumstances, interacting factors and diagnostic requirements of given diagnoses in childhood as compared to adulthood.

Corrigendum: Abilities and Disabilities-Applying Machine Learning to Disentangle the Role of Intelligence in Diagnosing Autism Spectrum Disorders.

[This corrects the article DOI: 10.3389/fpsyt.2022.826043.].

Abilities and Disabilities-Applying Machine Learning to Disentangle the Role of Intelligence in Diagnosing Autism Spectrum Disorders.

Objective: Although autism spectrum disorder (ASD) is a relatively common, well-known but heterogeneous neuropsychiatric disorder, specific knowledge about characteristics of this heterogeneity is scarce. There is consensus that IQ contributes to this heterogeneity as well as complicates diagnostics and treatment planning. In this study, we assessed the accuracy of the Autism Diagnostic Observation Schedule (ADOS/2) in the whole and IQ-defined subsamples, and analyzed if the ADOS/2 accuracy may be increased by the application of machine learning (ML) algorithms that processed additional information including the IQ level. Methods: The study included 1,084 individuals: 440 individuals with ASD (with a mean IQ level of 3.3 ± 1.5) and 644 individuals without ASD (with a mean IQ level of 3.2 ± 1.2). We applied and analyzed Random Forest (RF) and Decision Tree (DT) to the ADOS/2 data, compared their accuracy to ADOS/2 cutoff algorithms, and examined most relevant items to distinguish between ASD and Non-ASD. In sum, we included 49 individual features, independently of the applied ADOS module. Results: In DT analyses, we observed that for the decision ASD/Non-ASD, solely one to four items are sufficient to differentiate between groups with high accuracy. In addition, in sub-cohorts of individuals with (a) below (IQ level ≥4)/ID and (b) above average intelligence (IQ level ≤ 2), the ADOS/2 cutoff showed reduced accuracy. This reduced accuracy results in (a) a three times higher risk of false-positive diagnoses or (b) a 1.7 higher risk for false-negative diagnoses; both errors could be significantly decreased by the application of the alternative ML algorithms. Conclusions: Using ML algorithms showed that a small set of ADOS/2 items could help clinicians to more accurately detect ASD in clinical practice across all IQ levels and to increase diagnostic accuracy especially in individuals with below and above average IQ level.

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

Introduction: 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. Methods: 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: 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. Conclusion: By replicating previous findings in a large clinical sample our results contribute to further conceptualize the social-communication impairments in ASD as two dimensional.

Phenotypic differences between female and male individuals with suspicion of autism spectrum disorder.

BACKGROUND: Although autism spectrum disorder (ASD) is a common developmental disorder, our knowledge about a behavioral and neurobiological female phenotype is still scarce. As the conceptualization and understanding of ASD are mainly based on the investigation of male individuals, females with ASD may not be adequately identified by routine clinical diagnostics. The present machine learning approach aimed to identify diagnostic information from the Autism Diagnostic Observation Schedule (ADOS) that discriminates best between ASD and non-ASD in females and males. METHODS: Random forests (RF) were used to discover patterns of symptoms in diagnostic data from the ADOS (modules 3 and 4) in 1057 participants with ASD (18.1% female) and 1230 participants with non-ASD (17.9% % female). Predictive performances of reduced feature models were explored and compared between females and males without intellectual disabilities. RESULTS: Reduced feature models relied on considerably fewer features from the ADOS in females compared to males, while still yielding similar classification performance (e.g., sensitivity, specificity). LIMITATIONS: As in previous studies, the current sample of females with ASD is smaller than the male sample and thus, females may still be underrepresented, limiting the statistical power to detect small to moderate effects. CONCLUSION: Our results do not suggest the need for new or altered diagnostic algorithms for females with ASD. Although we identified some phenotypic differences between females and males, the existing diagnostic tools seem to sufficiently capture the core autistic features in both groups.

Differentiation of autism spectrum disorder and mood or anxiety disorder.

LAY ABSTRACT: Symptoms of mood and anxiety disorders overlap with symptoms of autism spectrum disorder, making the diagnostic process challenging. This study found that a combination of communicational deficits and unusual and/or inappropriate social overtures facilitates differentiation between autism spectrum disorder and mood and anxiety disorders. Furthermore, the results confirm the essential need of a behavioral observation with the Autism Diagnostic Observation Schedule in combination with a full Autism Diagnostic Interview-Revised to support diagnostic decisions.

How Do Adults with Autism Spectrum Disorder Participate in the Labor Market? A German Multi-center Survey.

International studies show disadvantages for adults with autism spectrum disorder (ASD) in the labor market. Data about their participation in the German labor market are scarce. The aim of this study was to examine the integration of adults with ASD in the German labor market in terms of education, employment and type of occupation by means of a cross-sectional-study, using a postal questionnaire. Findings show above average levels of education for adults with ASD compared to the general population of Germany and simultaneously, below average rates of employment and high rates of financial dependency. That indicates a poor integration of adults with ASD in the German labor market and emphasizes the need for vocational support policies for adults with ASD.

A data driven machine learning approach to differentiate between autism spectrum disorder and attention-deficit/hyperactivity disorder based on the best-practice diagnostic instruments for autism.

Autism spectrum disorder (ASD) and attention-deficit/hyperactivity disorder (ADHD) are two frequently co-occurring neurodevelopmental conditions that share certain symptomatology, including social difficulties. This presents practitioners with challenging (differential) diagnostic considerations, particularly in clinically more complex cases with co-occurring ASD and ADHD. Therefore, the primary aim of the current study was to apply a data-driven machine learning approach (support vector machine) to determine whether and which items from the best-practice clinical instruments for diagnosing ASD (ADOS, ADI-R) would best differentiate between four groups of individuals referred to specialized ASD clinics (i.e., ASD, ADHD, ASD + ADHD, ND = no diagnosis). We found that a subset of five features from both ADOS (clinical observation) and ADI-R (parental interview) reliably differentiated between ASD groups (ASD & ASD + ADHD) and non-ASD groups (ADHD & ND), and these features corresponded to the social-communication but also restrictive and repetitive behavior domains. In conclusion, the results of the current study support the idea that detecting ASD in individuals with suspected signs of the diagnosis, including those with co-occurring ADHD, is possible with considerably fewer items relative to the original ADOS/2 and ADI-R algorithms (i.e., 92% item reduction) while preserving relatively high diagnostic accuracy. Clinical implications and study limitations are discussed.

Health Services Use and Costs in Individuals with Autism Spectrum Disorder in Germany: Results from a Survey in ASD Outpatient Clinics.

Autism spectrum disorders (ASD) are associated with high services use, but European data on costs are scarce. Utilisation and annual costs of 385 individuals with ASD (aged 4-67 years; 18.2% females; 37.4% IQ < 85) from German outpatient clinics were assessed. Average annual costs per person were 3287 EUR, with psychiatric inpatient care (19.8%), pharmacotherapy (11.1%), and occupational therapy (11.1%) being the largest cost components. Females incurred higher costs than males (4864 EUR vs. 2936 EUR). In a regression model, female sex (Cost Ratio: 1.65), lower IQ (1.90), and Asperger syndrome (1.54) were associated with higher costs. In conclusion, ASD-related health costs are comparable to those of schizophrenia, thus underlining its public health relevance. Higher costs in females demand further research.