In his overview on cognitive effects of antipsychotics in schizophrenia Sharma (Reference Sharma1999) stresses a relationship between cognitive function in schizophrenia and quality of life as an outcome measure. I think that Sharma's use of the concept ‘quality of life’ has to be clarified to prevent a number of rather common biases. He quotes two studies that are said to support a relationship between cognitive function in schizophrenia and quality of life (Reference Davidson and KeefeDavidson & Keefe, 1995; Reference GreenGreen, 1996). The term quality of life is not operationalised in the first study. In the second study, which is in fact an overview of other studies, it is reported by Heinrichs' Quality of Life Scale (Reference Heinrichs, Hanlon and CarpenterHeinrichs et al, 1984). Like most other instruments which have been used to detect the effect of atypical neuroleptics on quality of life in schizophrenia (Reference Priebe, Oliver and KaiserPriebe et al, 1999) the Quality of Life Scale (subtitled “An instrument for rating the schizophrenia deficit syndrome”) assesses clinical judgements of negative symptoms of schizophrenia rather than subjective appraisals of quality of life made by the patient. As it seems reasonable to assume at least a moderate relationship of negative symptoms and cognitive functions in schizophrenia, it is not surprising that a relationship is found between cognitive functioning and quality of life when the quality of life measures seem to be confounded to a considerable extent by psychiatric symptomatology.

We think that it is necessary to make a distinction between quality of life as an evaluation criterion for illness-related phenomena (negative symptoms), and quality of life as a subjective assessment by the patient as a “ subjective evaluation of oneself and one's social and material world” (Reference Orley, Saxena and HermanOrley et al, 1998) - that is, subjective quality of life, not as a disease but as a generic concept. Since there are some studies that show that cognitive functioning in schizophrenia may predict social outcome, and since objective social outcome is moderately (although surprisingly weakly) associated with generic subjective quality of life, some association between cognitive functioning and subjective quality of life is conceivable, but has not yet been supported by empirical evidence.

In a validation study of a German short form of the Lancashire Quality of Life Profile (Reference Kaiser, Isermann and HoffmannKaiser et al, 1999), equivalent to the English short form of the instrument MANSA (see Reference Priebe, Oliver and KaiserPriebe et al, 1999), we did not find any significant correlation between any of the categories of the Wisconsin Card Sorting Test (WCST; Reference Heaton, Chelune and TalleyHeaton et al, 1993) (number of categories, perserverative errors and responses, etc.) and the mean value of all satisfactions ratings, satisfaction with life as a whole and with satisfaction with mental health in a carefully selected sample of out-patients with DSM-III-R schizophrenia (American Psychiatric Association, 1987; n=36; mean age=47 years; mean illness duration=19 years). Our conclusion so far is that whether or not subjective quality of life is related to cognitive deficits in schizophrenia (in attention or memory, besides deficits in executive functioning, which are seen on a variety of tasks, most notably the WCST) remains unclear and so far is only a hypothesis, although it is widespread as an advertising slogan for atypical antipsychotic medication.