Differentiation of autism spectrum disorder and mood or anxiety disorder

Wittkopf, Sarah; Stroth, Sanna; Langmann, Anika; Wolff, Nicole; Roessner, Veit; Röepke, Stefan; Poustka, Luise; Kamp‐Becker, Inge

doi:10.1177/13623613211039673

Cited by 13 publications

(19 citation statements)

References 55 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…Interestingly, the model with the anxiety group showed the largest increase in specificity for an ASD diagnosis, whereas the improvement for the models with the CD and ADHD samples was more related to increased sensitivity. This finding resonates well with other studies demonstrating lower specificity of the ADOS for Mood and Anxiety disorders (Wittkopf et al, 2021) and a high amount of symptom overlap (Hartley & Sikora, 2009;Tyson & Cruess, 2012;van Steensel et al, 2013). Thus, several individuals with anxiety-related disorders may be misclassified as having ASD when relying too much on the ADOS-cut off scores of module 3.…”

Section: Asd Classification In the Context Of Differential Diagnosessupporting

confidence: 90%

“…To a lesser degree, items such as Comments on Others' Emotions/Empathy, Shared Enjoyment in Interaction, Reporting of Events, and Asks for Information were also more important for the classification. Interestingly, the item Stereotyped/Ideosyncratic Use of Words or Phrases from the RBB domain was the item with the highest importance for the ANX vs. ASD model (see also (Wittkopf et al, 2021). The RBB domain might be particularly important for the distinction between ASD and anxiety disorders.…”

Section: Anx-model Item Importance Profilementioning

confidence: 97%

“…Diagnostic decisions in ASD are particularly challenging because difficulties in social interaction and communication are common also for a range of other conditions and behaviors including, for example, affective and anxiety disorders (Tyson & Cruess, 2012;van Steensel et al, 2013;Wittkopf et al, 2021), attention deficit hyperactivity disorder (ADHD, Ros & Graziano, 2018;Willis et al, 2019), and conduct disorder (CD, Gilmour et al, 2004;Milledge et al, 2019).…”

Section: Introductionmentioning

confidence: 99%

“…Recent work has explored the potential of machine learning (ML) methods to support diagnostic procedures (see Hyde et al, 2019 for a review). Most of these studies aimed at streamlining the diagnostic procedures for ASD by identifying a reduced set of most discriminative items from clinician-coded examinations such as ADOS-2 and ADI-R (Bone et al, 2015(Bone et al, , 2016Duda et al, 2014;Duda, Daniels, et al, 2016;Kamp-Becker, n.d.;Kosmicki et al, 2015;Küpper et al, 2020;Levy et al, 2017;Stroth, n.d.;Wall, Dally, et al, 2012;Wall, Kosmicki, et al, 2012;Wittkopf et al, 2021, see Bone et al, 2015 for a critical review). ML models built upon such selected items often retain or even exceed the diagnostic accuracy of the original ADOS-2 algorithm.…”

Section: Introductionmentioning

confidence: 99%

See 3 more Smart Citations

Using machine learning to improve diagnostic assessment of ASD in the light of specific differential diagnosis

Schulte‐Rüther

Kulvičius

Stroth

et al. 2021

Preprint

Self Cite

View full text Add to dashboard Cite

BackgroundDiagnostic assessment of ASD requires substantial clinical experience and is particular 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 comorbid and differential disorders.MethodWe used a well-characterized clinical sample of individuals (n=1262) that had received detailed outpatient evaluation for the presence of an ASD diagnosis (n=481) and covered a range of additional differential or overlapping diagnoses, including anxiety related disorders (ANX, n=100), ADHD (n=440), and conduct disorder (CD, n=192). 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) and an unspecific model using all available data. We employed nested cross-validation for an unbiased estimate of classification performance (ASD vs. non-ASD).ResultsWe obtained very good overall sensitivity (0.89-0.94) and specificity (0.87-0.89) for the classification of ASD vs. non-ASD. In particular for individuals with less severe symptoms (around the ADOS cut-off) our models showed increases of up to 20% in sensitivity or specificity. Furthermore, we analyzed item importance profiles of the ANX-, ADHD- and CD-models in comparison to the unspecific model. These analyses revealed distinct patterns of importance for specific ADOS-items with respect to differential diagnoses.ConclusionUsing ML-based diagnostic classification may improve clinical decisions by utilizing the full range of information from comprehensive and detailed diagnostic observation such as the ADOS. Importantly, this strategy might be of particular relevance for individuals with less severe symptoms that typically present a very difficult decision for the clinician.

show abstract

Section: Asd Classification In the Context Of Differential Diagnosessupporting

confidence: 90%

Section: Anx-model Item Importance Profilementioning

confidence: 97%

Section: Introductionmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

See 2 more Smart Citations

Using machine learning to improve diagnostic assessment of ASD in the light of specific differential diagnosis

Schulte‐Rüther

Kulvičius

Stroth

et al. 2021

Preprint

Self Cite

View full text Add to dashboard Cite

show abstract

“…The current data suggest anxiety and mood disorders are particularly important to understanding who may require more comprehensive in-person testing. Research suggests that ASD shares many symptoms with other psychiatric conditions, including anxiety and depression, making the differential diagnosis challenging (e.g., Greene et al, 2021 ; Wittkopf et al, 2021 ). Specifically, it can be difficult to accurately assess social communication deficits that may differentiate ASD and anxiety and mood disorders (Wittkopf et al, 2021 ).…”

Section: Discussionmentioning

confidence: 99%

Identifying Patient Characteristics to Understand Which Children May Receive Diagnostic Clarity in a Virtual Autism Spectrum Disorder Evaluation

Phelps

Sample

Greene

et al. 2022

J Autism Dev Disord

View full text Add to dashboard Cite

Given long waitlists for autism spectrum disorder (ASD) evaluation coupled with the COVID-19 pandemic, it is crucial to triage patients to services they are likely to receive diagnostic clarity (i.e., virtual, in-person evaluation). Participants attended a virtual ASD assessment. A subset also attended in-person evaluation. Results suggest younger children with educational services for ASD may benefit from virtual assessment while older patients with a history of psychiatric conditions may benefit from in-person evaluation. An ASD symptom severity tool related to virtual and in-person diagnostic clarity. Family history of ASD related to in-person diagnosis while other variables (e.g., age, educational services) did not. The study suggests patient characteristics may be used to determine for whom virtual ASD assessment may be appropriate.

show abstract