The validity of subjective quality of life measures in psychotic patients with severe psychopathology and cognitive deficits: an item response model analysis.

PURPOSE: Subjective quality of life (SQOL) is an established patient-reported outcome in the evaluation of treatments for psychosis. The use of SQOL measures in the presence of psychiatric symptoms and cognitive deficits has been questioned. However, there is little evidence on whether items function differently as indicators of SQOL in psychotic patients with different levels of symptoms and deficits. Substantial differential item functioning (DIF) would, indeed, challenge the validity of established measures. We aimed to investigate the validity of a widely used measure of subjective quality of life (SQOL), i.e., the Lancashire Quality of Life Profile (LQOLP), in the presence of cognitive deficits and psychiatric symptoms in patients with severe and enduring psychosis.
METHOD: We analysed SQOL ratings of 690 psychotic patients on the LQOLP using item response modelling to detect differential item functioning (DIF) attributable to psychiatric symptoms and cognitive deficits.
RESULTS: Patients with more severe general psychopathology were less likely to rate their 'personal safety' positively (OR .96, 95% CI .93-.99). More severely depressed patients were less likely to endorse positive 'life as a whole' (OR .93, 95% CI .89-.98) and 'mental health' (OR .93, 95% CI .91-.97) ratings. There was no DIF attributable to cognitive deficits.
CONCLUSIONS: The findings suggest that the validity of the LQOLP in psychotic patients may be impaired by DIF due to psychopathology, although the magnitude of effects is unlikely to be of clinical significance. The validity appears not to be compromised by cognitive deficits.