How much impairment is required for ADHD? No evidence of a discrete threshold

• Published in: Journal of child psychology and psychiatry 2022-02, Vol.63 (2), p.229-237
• Main Authors: Arildskov, Trine Wigh, Sonuga‐Barke, Edmund J. S., Thomsen, Per Hove, Virring, Anne, Østergaard, Søren D.
• Format: Article
• Language: English
• Subjects: Attention Deficit Disorder with Hyperactivity - diagnosis; Attention deficit hyperactivity disorder; Child & adolescent psychiatry; Comorbidity; continuity; diagnosis; Functional impairment; Humans; Medical diagnosis; Rating Scales; Regression (Statistics); schoolchildren; Severity; Socioeconomic status; symptomatology; Symptoms (Individual Disorders)
• ISSN: 0021-9630; 1469-7610
• DOI: 10.1111/jcpp.13440

Background A diagnosis of attention‐deficit/hyperactivity disorder (ADHD) requires the presence of impairment alongside symptoms above a specific frequency and severity threshold. However, the question of whether that symptom threshold represents anything more than an arbitrary cutoff on a continuum of impairment requires further empirical study. Therefore, we present the first study investigating if the relationship between ADHD symptom severity and functional impairment is nonlinear in a way that suggests a discrete, nonarbitrary symptom level threshold associated with a marked step increase in impairment. Methods Parent reports on the ADHD‐Rating Scale (ADHD‐RS‐IV), the Weiss Functional Impairment Rating Scale (WFIRS‐P), and the Strengths and Difficulties Questionnaire were collected in a general population sample of 1st, 2nd, and 3rd graders (N = 1,914–2,044). Results Piecewise linear regression analyses and nonlinear regression modeling both demonstrated that the relationship between symptom severity (ADHD‐RS‐IV total score) and impairment (WFIRS‐P mean score) was characterized by a gradual linear increase in impairment with higher symptom severity and no apparent step increase or changing rate of increase in impairment at a certain high ADHD‐RS‐IV total score level. Controlling for socioeconomic status, sex, and co‐occurring conduct and emotional symptoms did not alter these results, though comorbid symptoms had a significant effect on impairment. Conclusions There was no clear evidence for a discrete, nonarbitrary symptom severity threshold with regard to impairment. The results highlight the continued need to consider both symptoms and impairment in the diagnosis of ADHD.