Discriminant and convergent validity of a subjective quality-of-life instrument aimed at high content validity for schizophrenic persons

Abstract

To analyse whether a subjective quality-of-life (QoL) instrument (QLiS-Quality-of-Life in Schizophrenia), developed on the basis of a large number of open-ended interviews with schizophrenic patients, has sufficient discriminant and convergent validity to justify its application as a schizophrenia-specific QoL instrument. The discriminant and convergent validity of the QLiS (comprising 12 subscales) was analysed in a cross-sectional study. Schizophrenic persons (n = 135) from different care settings were surveyed using the QLiS, the WHOQOL-Bref, the SWN and 7-point satisfaction items. Partial correlational analyses and regression analyses controlling for general life satisfaction were conducted comparing the QLiS subscales with those of the other instruments. Positive correlation coefficients were found among all subscales of the QLiS and the other QoL instruments (WHOQOL-BREF from r = 0.29 to r = 0.72; SWN, r = 0.14 to r = 0.83; satisfaction scales, r = 0.18 to r = 0.69). One QLiS subscale (cognitive functioning) was shown to be empirically redundant (r>0.80) to the mental functioning subscale of the SWN. All other subscales proved to have unique variance. The non-QLiS QoL instruments only accounted for substantial amounts of variance (>20% after controlling for global life satisfaction) in the QLiS subscales leading a normal life, appreciation by others, appraisal of accommodation/housing and social contacts. Most of the QLiS subscales can be regarded as sufficiently distinct from other QoL instruments, and thus show evidence of discriminant and convergent validity. A subjective QoL questionnaire with high content validity can provide additional empirical information about schizophrenics' QoL not accounted for by other common QoL instruments.