Longitudinal Assessment of Quality of Life in Acute Psychiatric Inpatients: Reliability and Validity

This study examined the reliability, validity, and responsiveness of Lehman's Quality of Life Interview (QOLI) as an outcome measure on 981 acutely ill psychiatric inpatients assessed longitudinally at admission and discharge. Patients were stratified into five diagnostic (DX) (depressed bipolar, depressed unipolar, schizophrenia, mania, and other diagnoses) and two substance use disorder (SA) strata (with and without concurrent substance abuse/dependence) based on DSM-III-R criteria. There was good replication of the factor structure, excellent internal consistency, overall and within DX and SA groups. Intercorrelations showed that the functional and satisfaction indices measure unique aspects of the quality of life. The construct consistency of the QOLI was dependent upon psychiatric diagnosis and life domain. Intercorrelations of functional and satisfaction indices for patients with depression were greater than for manic patients. We demonstrated strong consistency of construct validity for family and social relation domains, but not safety or leisure activities. Construct validity was shown to hold longitudinally. Analyses of DX and SA group differences on satisfaction and functional indices of the 8 life domains supported discriminative validityDepressed patients reported the most dissatisfaction, followed by schizophrenic patients, and manic patients reported the greatest satisfaction in most life domains. Patients with concurrent substance abuse generally reported less satisfaction and lower quality of life than patients without a dual diagnosis. Examination of longitudinal changes in satisfaction indicated the QOLI is responsive to changes in global life, leisure activities, living situation, and social relations from hospital admission to discharge (an average of 2 weeks). This study supports the use of the QOLI as an outcome measure to assess quality of life in acutely ill hospitalized psychiatric patients.