Skip to main content
Kudos to members of one of the University of Minnesota’s residency programs, who describe in their Innovation a plan to speed diagnosis and treatment initiation for ADHD (and depression, complex psychiatric conditions, and substance use disorders) through a new process within their family medicine clinic. It is all of our work to refine processes, reduce backlog, and improve patient satisfaction, and like those involved in the Innovation I too believe that family physicians are best-positioned to diagnose and manage ADHD – no other specialty sees the impact of impoverished executive function across the lifespan, nor could another specialty as usefully address all the varied derangements that could together impair cognition.
Perhaps the brief format of the Innovation article undermines their argument that their process meets the need, however; perhaps they left out important detail. As written, the process is described as using a single office visit – of “standard length”, presumably 15 to 30 minutes – to administer and act upon screens for depression, bipolar disorder, generalized anxiety, substance abuse disorders, sleep apnea, and ADHD. Guidelines now under development by the American Professional Society for ADHD and Related Diseases and expert opinion believe that this work usually takes 60 to 90 minutes; this teaching clinic teaches its residents haste, risking quality. The attempt to reduce diagnosis of these potentially challenging and widely varied conditions to use of a screening tool misuses those tools, which are not of themselves diagnostic; the Innovation risks instructing its users to conflate a screen with insight.
As importantly, the initial encounter fails to screen for significant contributants to problematic attention. There is no critical assessment of the medication list; the quality of special senses isn’t considered (I once cured “ADHD” with a cochlear implant). The families burdened with ADHD are those same families managing eating disorders; where is a SCOFF, or a BEDS-7? Iron deficiency isn’t sought; estrogen variation or the “brain fog” of menopause is missed. Learning disorders are nearly as often comorbid with ADHD as is anxiety – but the Innovation doesn’t reach for a Colorado Learning Difficulties Questionnaire or other screen for these. And, hugely, there is no review of trauma; if that patient’s ADHD was not named in childhood, was that because of a double-digit ACE score, counseling around which would be needed to optimize function in even in late adulthood?
After an initial encounter that finishes with an ASRS screener – a test, like the other screening tools, meant to be used within a clinical conversation, and which has a mediocre 11.5% positive predictive value meaning that 86–90% of people identified as having probable ADHD with this tool are unlikely to have ADHD – the patient is sent home to complete a DIVA questionnaire. The DIVA, like the aforementioned tools, is pointedly not meant to be used alone; it is instead a clinical interviewer’s guide to a semi-structed conversation, ideally completed in the presence of a family member who can provide collateral information about its’ content. Completed alone and without corroboration, the Innovators review the patients’ annotation of it and conclude that 80% of the patients self-referred meet diagnostic criteria for ADHD – a condition which is thought present in 2.5 to 5% of adults in the United States, and for which the diagnostic thresholds were intended to capture function several standard deviations from norms. This outcome mocks the vast importance of the label.
Family medicine residents are in desperate need of exposure to nuanced primary care psychiatry and are right to address ADHD in adults within their clinic; I applaud Heltemes et al for identifying the issue and directing their attention to it. But the Innovation as described threatens to confuse learners with its implications, to muddle patients’ diagnoses lists with imprudent labels, and to miss numerous factors which if treated could benefit patient cognition.