By Molly A. Nikolas, Paul S. Marshall & James B. Hoelzle
Abstract
Despite widespread recognition that attention-deficit/hyperactivity disorder (ADHD) is a lifelong neurodevelopmental disorder, optimal methods of diagnosis among adults remain elusive. Substantial overlap between ADHD symptoms and cognitive symptoms of other mental health conditions, such as depression and anxiety, and concerns about validity in symptom reporting have made the use of neuropsychological tests in ADHD diagnostic assessment appealing. However, past work exploring the potential diagnostic utility of neuropsychological tests among adults has often relied on a relatively small subset of tests, has failed to include symptom and performance validity measures, and often does not include comparison groups of participants with commonly comorbid disorders, such as depression. The current study examined the utility of an extensive neuropsychological measure battery for diagnosing ADHD among adults. Two hundred forty-six participants (109 ADHD, 52 depressed, 85 non-disordered controls) completed a multistage screening and assessment process, which included a clinical interview, self, and informant report on behaviour rating scales, performance and symptom validity measures, and an extensive neuropsychological testing battery. Results indicated that measures of working memory, sustained attention, response speed, and variability best discriminated ADHD and non-ADHD participants. While single test measures provided performed poorly in identifying ADHD participants, analyses revealed that a combined approach using self and informant symptom ratings, positive family history of ADHD, and a reaction time (RT) variability measure correctly classified 87% of cases. Findings suggest that neuropsychological test measures used in conjunction with other clinical assessments may enhance prediction of adult ADHD diagnoses.
Short comment by Marios Adamou, psychiatrist
Several validated tools are available to facilitate the diagnostic assessment of ADHD in adulthood. The most well-known and frequently used include the Conners Adult ADHD Diagnostic Interview for DSM-IV (CAADID) which is a validated semi-structured interview and the DIVA-5 interview (Diagnostic Interview for ADHD in adults) which also takes the form of a semi-structured interview. These assessment tools have been found to have good sensitivity and reasonable specificity for the diagnosis of Adult ADHD. The process of diagnosing Adult ADHD remains complex however and diagnostic overshadowing remains an ever-present challenge for clinician’s tasked with differentiating between the neuro-cognitive deficits attributable to a neurodevelopmental cause versus those that might arise from any number of environmental causes, lifestyle factors or other psychiatric phenomena.
It is recognized that inattentiveness and dysexecutive functioning can arise from several causes which might not necessarily indicate the presence of ADHD. Current diagnostic assessments of ADHD tend to be weighted towards high levels of sensitivity (they can detect the presence of inattention and dysexecutive functioning that might represent ADHD) but do not have the same capacity for specificity (they cannot easily discount other factors which might explain inattention and dysexecutive functioning). The challenge for diagnostician’s therefore relates to how the latter can be maximized.
Neuro-cognitive assessments provide one potential avenue for increasing the specificity of diagnostic assessments of adult ADHD because they allow for direct measurement and quantification of the core features of the condition. Moreover, within neuro-psychiatric assessments it is possible to measure and control for confounding variables which might unduly influence the diagnostic outcome e.g., the application of effort. Despite their obvious promise neurocognitive assessments are often restricted for use as a screening tool. Examples of this include the use of continuous performance tests such as the Qb (Qb tech) and the Conner’s Continuous Performance Test (Conners CPT 3).
This study is unique in its breadth, it examines the performance of adults on all major neurocognitive domains and on performance validity tests whilst comparing performance with data derived from behaviour rating scales from participants and collateral information. The aim of this study is to determine the combination of the sources of data yielding the highest diagnostic accuracy. The large sample in this study (N=256) comparative to other studies of this kind is a strength. Performance on neurocognitive assessment was compared between three groups comprising; adult ADHD, unipolar depression and normal control. The inclusion of a unipolar depression clinical comparison group is important given that this is a commonly occurring psychiatric co-morbidity with ADHD. Nevertheless, the non-inclusion of other common co-morbid psychiatric conditions is a limitation.
This study provides four main findings. First, a combined approach including self and informant report, family history, and neurocognitive tests was optimal yielding a diagnostic accuracy of 87% when distinguishing between adults with and without ADHD. Second, the importance of neurocognitive tests in this process was inconclusive. Domains of working memory, inhibition, response speed, and sustained attention corresponded to the largest group difference and the Test of Variables of Attention (TOVA) proved one of the most promising. Third, cut off scores on neurocognitive assessments mirrored normal cut offs which may indicate a bias in the participant pool (it contained a large number of higher educated subjects). Finally, there was little evidence on performance validity tests to suggest sub optimal effort was a significant problem in the participant pool. In summary, the results suggest that the use of neurocognitive assessments might optimise the sensitivity and specificity of diagnostic assessments when combined with semi-structured interview, informant report and self-report measures but should not be relied upon solely when making a diagnostic judgement.
With thanks to Dr Tim Fullen Consultant Clinical Psychologist