Are tests of frontal lobe functions useful in the diagnosis of attention deficit disorders?

Abstract Nine neuropsychological tests presumed to assess various frontal lobe functions in children were evaluated for their accuracy in classifying children as having attention deficit disorders (ADD) with (ADD+H) and without hyperactivity (ADD-H). Four groups of children were evaluated: ADD+H (n=12), ADD-H (n=12), non-ADD learning disabled (LD) children (n=11), and a normal community control group (n=12). Results indicated low levels of sensitivity for the ADD groups. Positive predictive power (PPP) for the diagnosis of ADD+H was poor for eight of the tests reaching levels exceeding 50% only for the CPT commission score. Combining both ADD groups resulted in a PPP that approached or exceeded reasonably acceptable levels (90% or higher) for two tests: the CPT scores and the F-A-S version of the Controlled Oral Word Association Test. This led to a marked reduction in negative predictive power for these same tests, however. Hence, abnormal scores on these two tests may be predictive of ADDs, though not of which subtype, while normal scores are not indicative of an absence of ADD and should go uninterpreted. The remaining tests do not seem helpful for clinical diagnostic purposes with ADD children at this time.