The diagnosis of ADHD in children is fraught with ambiguity and bias with the consequence of poor epidemiological understanding and treatment decisions. More quantitative assessment can improve outcomes.
ADHD is a disorder characterized by symptoms of inattention, impulsivity and hyperactivity. Over recent years there has been considerable debate about where the line between normal behavior ends and ADHD begins, as well as discussion about the pharmacological treatment of behaviors which, in some instances, could be seen as imperfect rather than abnormal.
See related posts: ADHD and the theta/beta ratio and Neurofeedback and ADHD.
According to a recent survey from 2016, the prevalence of ADHD in children in the United States is estimated to be 9.4%, although diagnosis rates vary considerably from country to country. This makes ADHD, first described in 1957, to be the most commonly diagnosed childhood disorder in the USA and in an increasing number of other countries globally. How does one arrive at such an epidemiological number?
ADHD is typically diagnosed using standardized clinical diagnostic criteria such as the DSM-5 or the ICD-10. (See The Challenges of Mental Health Diagnosis for more discussion). There are also complementary questionnaires which have been developed to provide an additional perspective on the symptoms of childhood ADHD (e.g. Conners parent rating scale, ADHD Child Evaluation, Vanderbilt ADHD Diagnostic Rating Scales), although these aren’t considered to be sufficient alone to make an accurate diagnosis. But how easy is to accurately diagnose ADHD?
Remembering the past?
The DSM-5 is the latest version of structured clinical diagnosis in a long history of categorizing symptoms to determine the appropriate treatment. In relation to ADHD, it typically requires individuals to think about the past 6 months of their life and asks them questions such as “have you often had trouble staying focused on things like reading a book, following a conversation or doing household chores?”; “have you often finished people’s sentences or blurted out an answer before the other people finished asking the question?” or “have you often felt physically restless, especially when you had to stay put for a while?”. However, when we know that the human brain typically exhibits recency and primacy biases (memory biased towards recent experiences or first time experiences), how accurate are people at remembering what their, or their child’s, behavior was like over the past 6 months? For example, if you kept of diary of your attention levels for a week, and then a week later answered the question “how distracted or inattentive were you last week?” would they line up accurately? Probably not. Therefore, only relying on potentially inaccurate memory traces may confound the accurate diagnosis of ADHD.
Often or occasionally?
Similarly, commonly used questionnaires such as the Conners Parent Rating Scale provide a series of statements such as “Forgetful in daily activities” and “Has difficulty waiting in lines or awaiting turn in games or group situations” and requires the parent to choose whether, over the past month, they think the right answer is “Not true at all (never, seldom)”; “Just a little true (occasionally)”; “Pretty much true (often, quite a bit)” or “Very much true (very often, very frequent)”. This not only requires an accurate memory of the last month, but also a subjective assessment of how the frequency of the problem behavior equates to terms such as “occasionally” or “often”, across 80 different behavioral measures. Furthermore, the subtle inconsistency of response terminology used across different screening questionnaires (e.g. is “somewhat true” – a term used in the Child Behavior Checklist – the same as “just a little true” or “pretty much true”?) is a problem which extends beyond ADHD screening and diagnosis.
See related post The Difficulty of Diagnosing Depression
A negative skew on behavior
It also becomes very apparent when reviewing ADHD screening and diagnostic interviews that just about every question focuses on negative aspects of the child’s behavior such as “how much they avoid, dislike or are reluctant to engage in tasks that require sustained mental effort?”. Such framing will subtly direct the child or parent to recall only the negative incidents and not the positive ones Instead, if the questions had a positive framing (for example if the clinician asked “how much they approach, like or enjoy engaging in tasks that mentally challenge them?” instead of) it is quite possible that it would elicit a shift in the diagnosis profile. More generally it raises the question about whether focusing only on what is “wrong” with the child, without having a clear picture of what is “right” with them, is appropriate for making a balanced diagnosis of ADHD.
Outcomes and triggers
Critically the DSM-5 also asks a couple of questions at the end of the section about the impact of the symptoms on the individual’s daily life such as “How have your ADHD symptoms affected your work/school?”. This allows the clinician to make the all important judgement not only of the symptoms, but of how much those symptoms are actually causing a problem for the individual. In contrast, there is very little explicit questioning on the causes or triggers of the symptoms, or the wider context relating to the onset of those symptoms, beyond what the clinician themselves may choose to add. It is not uncommon for example, for a child to not pay attention when they are preoccupied by a stressful circumstance. Obtaining a more balanced perspective of symptoms, outcomes and causes, would potentially provide a broader view of the ADHD symptom trajectory, aiding not only the diagnosis of ADHD, but also the selection of an appropriate treatment plan.
Towards more quantitative diagnosis
In summary, the current approach of employing a subjective-style of questioning dependent on patient and care-giver memory, combined with the use of potentially ambiguous verbal labels and a negative skew in behavior assessment leads to tremendous ambiguity. This can result in poor epidemiological assessment and conclusions. Worse, it can lead to misdiagnosis and inappropriate treatment, detrimentally impacting the life of the child. Approaches that utilize more objective ways of tracking symptoms behavior over time can minimize some of these biases, offering an opportunity to both strengthen the accuracy of ADHD diagnosis and provide a better benchmark of assessment against which to evaluate progress.
Posted by Jennifer J. Newson