Review of: Berger I et al, Arch Clin Neuropsychol 2017;32:81–93.

Diagnosing ADHD continues to be difficult, time-consuming, and expensive. Having an “objective,” valid test for ADHD would be of value to clinicians and families alike. Continuous performance tests (CPT) are gaining popularity as a complement to clinical examination, rating scales, and interviews with parents and teachers. Depending on the specific test, the CPT can generate information about impulsivity, focus, distractibility, and sometimes hyperactivity. But do these quick, kid-friendly computer tests actually aid in diagnosis, or are they just another data point without real clinical benefit? Researchers evaluated the ability of the MOXO-CPT (NeuroTech Solutions Ltd)—a computer-based test that assesses several variables, including “correct attention,” “correct timing,” “impulsive commission,” and “hyperactive commission”—to detect the presence or absence of ADHD.

Researchers recruited 459 drug-naïve children with ADHD from outpatient pediatric clinics at a university hospital neurocognitive center. Children were included if they met DSM-IV-TR criteria for ADHD and if they did not have a comorbid disorder. Controls were normally developing children recruited from regular school classrooms who scored below the cutoff for ADHD or any developmental disorder on all Conners subscales and DSM-IV scales. Altogether, 493 boys and 305 girls aged 7–12 participated. The two groups did not differ in gender ­distributions.

For all subjects, parents and teachers were asked to fill out the Conners Rating Scale or ADHD/DSM-IV scales. Participants took the MOXO-CPT, a simple 15-minute computer test in which images appear on screen for 0.5, 1, or 3 seconds, followed by a blank screen for an equal amount of time. Short animated clips (3–15 seconds) with audio and visual features appear randomly and in close proximity to the test images, to simulate the everyday distractions faced by those with ADHD. In this study, participants were instructed to give a timely, accurate response by pressing the space bar once whenever the designated image appeared. Errors were counted and coded to four performance indices: delay (timing), failure to respond (attention), pressing the space bar excessively or pressing other keys (hyperactivity), or response to the wrong image (impulsivity). The sum of the four indices provided a total score. Presumably, children with ADHD who were on medication did not take it when completing the MOXO-CPT, but the authors did not address medication beyond excluding “chronic” use.

Compared to controls, children with ADHD achieved statistically significant lower scores on all four indices, with the total score being the most robust measure. These findings held true when the subjects were stratified by age. Using different post-hoc cutoffs for the MOXO-CPT total scores for each age, the authors calculated sensitivity ranging from 81% to 91% and specificity ranging from 85% to 89%. The authors concluded that “integration of CPT indices improves the diagnostic capacity of ADHD.”

CCPR’s take
The authors demonstrated that, under the simplest condition (ie, distinguishing children who have only ADHD from those with no developmental issues), the MOXO-CPT performed well. Even so, it still did less well than the gold standard of a clinical diagnosis by a skilled clinician. Moreover, the authors’ decision to use only children with ADHD who were not taking medication chronically and who had no common comorbid conditions excluded such a large proportion of the ADHD population that it is impossible to judge how helpful this version of the CPT would be in real-world clinical situations.

Practical Implications
Clinicians who are seeking “hard data” to boost their diagnosis of ADHD may wish to check out this CPT, along with the many others on the market. However, this is clearly not a shortcut that can replace a thorough assessment, especially when other conditions also are present.