Accurate assessment of adult ADHD: a key to better outcomes?

Cortese et al ¹ provide a thoughtful overview of the scientific and clinical aspects of attention‐deficit/hyperactivity disorder (ADHD) in adults. This is a field that has developed rapidly over the past few years, moving from the margins of most people's perception and awareness into the limelight. While this increased recognition of and attention to the needs of adults with ADHD is welcome, it has brought with it some additional demands and challenges. This is highlighted by the tension between the debate, played out publicly in the media, about whether ADHD is being “overdiagnosed”, and the message that adults with ADHD are having great difficulty accessing adequate services ² .

With reference to adult ADHD, it is more helpful to talk about missed diagnosis and misdiagnosis ³ . It is almost certainly the case that in many countries we are currently seeing a combination of the two. While there are still many adults with ADHD who do not receive a diagnosis or treatment, there are also people who do not meet the criteria for ADHD but have been given a diagnosis and are receiving treatment, most commonly with stimulant medication.

The most common route to this is a quick and poorly conducted assessment. By continuing to miss a diagnosis of ADHD, we are also missing the opportunity to facilitate access to the available effective pharmacological and non‐pharmacological treatment. But by diagnosing and treating people who do not meet criteria, we are changing the meaning of ADHD and moving beyond the evidence base. I understand the pressures on clinicians, particularly in private practice, when asked to conduct an “ADHD assessment” by someone who for whatever reason and by whatever route has already identified strongly or perhaps “self‐diagnosed”, but we must remain objective and assess properly.

Indeed, one of the most important issues facing clinicians working in this area is to ensure that assessments are accurate and comprehensive. This does not seem to always be the case. There have been many recent reports of services popping up that are offering rapid access to “specialist ADHD assessments”, often over telehealth and with a hefty price tag. Feedback from people who have been through such processes is that the assessment was short, often a single appointment of around 30 min, relied heavily on questionnaires rather than clinical interview, and did not include a developmental history, or a screen for other physical or mental health conditions, or the collection of collateral information from sources other than the person being assessed. A positive diagnosis is almost always made. While medication treatment is usually recommended, this is not offered as part of the service. Instead, a referral is made to a primary care physician who is asked to start, titrate and monitor medication and outcomes.

If this is not good practice, how should we be working? I strongly advise that, as recommended in evidence‐based guidelines, a structured clinical interview should always form the core of a clinical assessment ⁴ . As emphasized by Cortese et al, our work on screening questionnaires in children and young people highlighted problems with specificity, and are therefore associated with high rates of false positives ⁵ . While we have not yet completed our review of screeners in adults, it is likely that we will find a similar result. Why are current approaches to screening so disappointing? One major factor is that they focus on symptoms and do not assess impairment. Perhaps in the future we will be able to harness multistage screening and/or artificial intelligence to help with both screening and assessment, but this is not yet the case.

Another aspect of assessment that I would take a hard line on is the use of cognitive testing, including continuous performance tasks (CPTs) such as QbTest, to aid diagnosis. At the cognitive level, ADHD is highly heterogeneous. While people with this condition have differences across a broad range of cognitive functions, there is no definitive ADHD cognitive profile ⁶ . No cognitive deficit is shared by everyone with ADHD, and no cognitive deficit is unique to ADHD. Just because one performs poorly on one or more task(s), including QbTest, that does not mean that he/she has ADHD. On the other hand, when someone does not show problems with executive functioning or on a CPT, this does not rule out ADHD. Cognitive testing can help you understand a person's strengths and weaknesses, but it does not aid with diagnosis ⁷ .

This also helps to answer another question posed by Cortese et al: whether executive functioning should be considered a core feature of ADHD. The answer depends on how one is defining this cognitive domain. Performance on neuropsychological tests of executive functioning as part of a formal assessment is clearly not decisive. Less than half of those with ADHD perform poorly. On the other hand, scoring highly on the Behavior Rating Inventory of Executive Function (BRIEF) is very common. However, it is important to note that, while the BRIEF scores correlate very highly with ADHD symptom measures, their correlation with recognized tests of executive functioning is usually non‐significant. I firmly believe that executive functioning difficulties should be listed among the associated features of ADHD rather than regarded as a core feature.

One area where we clearly need more evidence is for females with ADHD. As highlighted by Cortese et al, the male to female ratio changes across development, with a preponderance of males in childhood but equality in adults. What is not clear is how and why this happens. Are more male children with ADHD remitting before adulthood? Are some females who are subsyndromal during childhood developing full ADHD as they reach adulthood? Is it a combination of these, or is there another explanation?

A close examination of data from the most recent epidemiological Australian survey of child and adolescent mental health is intriguing. For children aged 6 to 12 years, the prevalence of ADHD is 10.9% in males and 5.4% in females, a ratio of around 2:1. For adolescents aged 13 to 18 years, the prevalence in males is similar (9.8%), but the prevalence in females drops by around 50% to 2.7%, giving a male to female ratio of 3.6:1 ⁸ . This was a well‐conducted population‐based study. The assessment process, a structured research diagnostic interview, was similar across the age range. I believe that the results of this study pose serious questions about the validity and reliability of our clinical assessments for adolescent females. It seems very unlikely that there is an increase in the gap between males and females during adolescence and then an equalling out in adulthood. Is this problem with assessment limited to adolescents, or does it continue into adulthood? I am not suggesting here that we should have different criteria for males and females, although perhaps this argument could be made. What I argue is that we need to be sure that we are accurately identifying symptoms in females and applying the same standards across sexes and throughout development.

There are, of course, many other areas relevant to adult ADHD in which we could improve our performance. However, if we can at least get the assessment process right and make sure that we are diagnosing the right people, this will be a very significant first step.