Anxiety is one of the most common psychiatric comorbidities associated with ADHD, and can affect up to 47% of adults with ADHD. There is symptomatic overlap between ADHD and anxiety (e.g. restlessness, concentration difficulties, decreased attention, increased distractibility, mood swings and anger outbursts), which may lead to misdiagnosis. The aim of this study was to assess the ability of the widely used CAARS* and STAI† self-report rating scales to differentiate between patients with ADHD and/or anxiety and a control population, and also between patient subgroups (ADHD vs anxiety vs ADHD plus anxiety). This study also aimed to assess overlapping response patterns across the CAARS and STAI subscales in patients with ADHD and/or anxiety.
In this cross-sectional study, a clinical sample of 52 adults with ADHD and/or anxiety (mean age 30.9 years) and an additional 74 adults without ADHD or anxiety (mean age 27.6 years) were recruited. The clinical sample consisted of 22 patients with ADHD (mean age 30.6 years), 14 patients with anxiety (mean age 31.1 years) and 16 patients with ADHD and anxiety (mean age 31.0 years). Participants were asked to complete the CAARS and STAI rating scales via an online platform. Sensitivity and specificity rates were calculated for each subscale of the CAARS and STAI for the whole sample, and specificity rates for the clinical sample were also obtained. A comparison of mean scores indicated that the clinical sample scored higher on both the CAARS and STAI subscales than the control sample (p<0.005). A cluster analysis was used to assess the pattern of responses of all participants on the CAARS and STAI self-report measures in order to assess overlapping symptoms. The different clusters that were assessed included: ‘inattention/memory’, ‘emotions/well-being’ and ‘hyperactivity, impulsivity and anxiety’. For each of the three cluster groups, observation of the means showed that the clinical sample scored significantly higher compared with the control sample (p<0.001).
Results from the study showed that, overall, the specificity rates of both the CAARS and STAI rating scales were good. However, poorer discriminant ability was observed in the clinical sample versus the whole sample for both rating scales, with specificity rates in the clinical sample falling below the 70% threshold for the CAARS inattention/memory problems subscale (50.00%), the CAARS impulsivity/emotional lability subscale (64.29%), the CAARS inattention symptoms subscale (42.86%) and the STAI trait subscale (63.64%). The clinical sample scored significantly higher on all subscales versus controls (p<0.005) for both the CAARS and STAI scales. There were significant differences between clinical subgroups observed for CAARS inattention/memory (p=0.001), CAARS problems with self-concept (p=0.039), CAARS inattentive symptoms (p=0.001) and STAI trait (p=0.025).
The study also aimed to propose changes to be made to the current versions of the CAARS DSM total symptoms and CAARS ADHD index, and STAI scales for differential diagnostic use between ADHD and/or anxiety based on results from the cluster analysis. This analysis showed a significant difference (p<0.001) between clinical subgroups for the ‘inattention/memory’ and ‘emotions/well-being’ clusters, and for the ‘hyperactivity, impulsivity and anxiety’ cluster. Therefore, this article suggests that these clusters may contain items that distinguish between the three clinical subgroups.
Limitations of this study included the small sample size and participant diagnoses not being confirmed using a semi-structured clinical interview. Moreover, a different analysis method (e.g. two-step clustering using larger samples) could have provided further insight.
The authors concluded that clinicians should be made aware of the limitations of the CAARS and STAI scales in differentiating between symptoms of ADHD and anxiety, and an alternative approach should be considered when making a differential diagnosis of ADHD in the context of anxiety. Recommendations for modified CAARS and STAI rating scales included use of inattentive items only for CAARS and exclusion of state anxiety-present items on STAI.
*The Conners’ Adult ADHD Diagnostic Interview for DSM-IV (CAARS) is made up of 66 items which can be subdivided into eight subscales: inattention/memory problems, hyperactivity/restlessness, impulsivity/emotional lability, problems with self-conception, DSM-IV inattentive symptoms, DSM-IV hyperactive/impulsive symptoms, DSM-IV ADHD total symptoms and ADHD index. Ratings are given on a 4-point scale (responses include: “Not at all, never”, “Just a little, once in a while”, “Pretty much, often” and “Very much, very frequently”), with individuals scoring T>70 likely to meet the diagnosis for ADHD
†The State Trait Anxiety Inventory (STAI) is a 40-item questionnaire relating to current and general symptoms of anxiety. All items are rated on a 4-point scale (responses include: “Almost never”, “Sometimes”, “Often” and “Almost always”). T>70 is a positive diagnosis for anxiety
Grogan K, Gormley CI, Rooney B, et al. Differential diagnosis and comorbidity of ADHD and anxiety in adults. Br J Clin Psychol 2017; Epub ahead of print.