Morgan et al have made a useful contribution in the area of intellectual disability/mental illness dual diagnosis. ^{Reference Morgan, Leonard, Bourke and Jablensky1} However, this study, like most in this area, is flawed by inadequate definition of terms. ‘Intellectual disability’, the current phrase of fashion for this population, is unsatisfactory because many individuals in the higher IQ ranges are not disabled. The American Association on Mental Retardation (now AAIDD) definition, probably the most widely used definition, is cited. It gets around the disability issue by requiring that individuals with intellectual disability must also have ‘limitations in adaptive behaviours and skills’. This confounds the intellectual disability and mental illness categories, as such limitations may well be due to mental illness. Perhaps a better term for studies to use would be ‘intellectual impairment’, which, like visual impairment, does not necessarily imply disability; then, all individuals in certain IQ ranges could be included. As it is, a certain proportion of individuals without mental illness are excluded by the definition. This may inflate the prevalence rates.

Additionally, there is a problem in lumping together all ranges of intellectual disability. As Morgan et al note, mental illness, particularly schizophrenia, is more likely to be diagnosed in the borderline group and pervasive developmental disorder is more likely to be diagnosed in the severe/profound group. Rather than a true reflection of incidence, this may reflect a nosological bias. A strict definition of schizophrenia is difficult to apply to a non-verbal person. Historically, pervasive developmental disorder and schizophrenia have sometimes been used interchangeably in apparently disturbed and non-verbal individuals; however, since the 1990s, at least in the USA, there has been a massive shift towards the diagnosis of pervasive developmental disorder sub-categories such as autism and Asperger syndrome. The diagnosis of schizophrenia has an additional stigma which some families find unacceptable. The authors found some trends distinguishing individuals with dual diagnosis from those with intellectual disability alone. Some of these trends also distinguished borderline from other levels of intellectual disability (e.g. fewer genetic causes, less Down syndrome, less epilepsy). To distinguish dual diagnosis from intellectual disability alone, results should probably be controlled for IQ level.

Morgan et al have considered patients with dual diagnosis to have more severe mental illness than other patients with mental illness as indicated by number of hospitalisations, length of hospitalisations, etc. Perhaps this just indicates that treatment and placement options for these patients are poorer. Future studies need to be done to clarify the unique aspects of this population.