We read Grisham et al's paper ^{Reference Grisham, Anderson, Poulton, Moffitt and Andrews1} with some concerns. Without doubt, the study – using longitudinal data instead of cross-sectional designs as in previous studies – adds positively to the subject area, which has not been well researched and the results of which are far from conclusive. However, we have a number of concerns about the reported results.

This study represents only one type of population and also, owing to the small number of obsessive–compulsive disorder (OCD) cases (only 13/700) found in this study, the authors' statement of ‘individuals with OCD have premorbid impairment in visuospatial abilities and some forms of executive functioning, consistent with biological models of OCD’ will be considered an overstatement and cannot be generalised to other population subsets.

We know that a previous study of OCD with the same birth cohort at age 18 found that the OCD group did not differ significantly on any of the neuropsychological tests at age 13, ^{Reference Douglass, Moffitt, Dar, McGee and Silva2} and it will be interesting to know whether there were any associations at ages 15, 21 and 26, ages at which Grisham et al's cohort was also assessed. Although participants in Grisham et al's study were assessed at 3, 5, 7, 9 and 11 years of age, according to their performance on neuropsychological tests there is some evidence to suggest that there is no cognitive impairment in children with OCD, and that OCD symptoms may not interfere with cognitive abilities early on in the illness. ^{Reference Beers, Rosenberg, Dick, Williams, O'Hearn and Birmaher3} However, disturbance of cognitive functions may become significant over time, as we know that psychotropic medications are the main pharmacological treatment that may also influence neuropsychological function. ^{Reference Hollander, Schiffman, Cohen, Rivera-Stein, Rosen and Gorman4} Neuroimaging studies suggest that the basal ganglia and ventral prefrontal cortex are most frequently implicated in OCD in adults. If brain dysfunction underlies OCD, decrements on neuropsychological tests should be found. ^{Reference Beers, Rosenberg, Dick, Williams, O'Hearn and Birmaher3} With this in mind, it is difficult to understand how people had neuropsychological deficits prior to developing OCD, when evidence suggests that children with OCD do not exhibit significant cognitive deficits early in the illness.

Evidence is in favour of executive dysfunction and auditory attention problems in late-onset OCD (age 13–17) rather than the early-onset (prior to 12 years) disorder. Performing poorly on the neuropsychological tests is not very conclusive as they may help to identify a dysfunction in a particular anatomical area, but provide little evidence on the actual cause leading to the pathology. Late-onset OCD is also associated with poorer visual memory relative to healthy comparison groups. Roth et al's findings ^{Reference Roth, Milovan, Baribeau and O'Connor5} suggest that early- and late-onset OCD may be the result of at least partially differing neurobiological mechanisms.

There is not much evidence at present to show the effects of therapeutic interventions on neuropsychological deficits in OCD, ^{Reference Bolton, Raven, Madronal-Luque and Marks6} and if any, are they curative in order to avoid the illness in future? The majority of people who had OCD also had comorbid illnesses – was it these illnesses that were the cause of neuropsychological deficits that later led to developing OCD (chemical abnormalities such as serotonin)? Perhaps studies on this aspect may be an area of interest for the authors.

The number of participants in the study is so small that no definitive statements should be made at this stage. We also wonder whether there are children and adolescents with neuropsychological deficits but not diagnosable psychiatric disorders and how we might compare them with individuals with conditions such as OCD.