Health-related quality of life in schizophrenia can be assessed by direct patient response or by a rating based on a structured interview. This study compares both types of instruments using a series of five standards: (1) sensitivity to change over time, (2) sensitivity to treatment effect, (3) correlation with symptom severity, (4) correlation with global clinical ratings, and (5) correlation with other measures of health-related quality of life. Four hundred and twenty-three inpatients with schizophrenia participating in a clinical trial comparing clozapine and haloperidol (VA Cooperative Study in Health Services #17) were evaluated using multiple measures of health-related quality of life (Lehman Quality of Life Interview; Heinrichs-Carpenter-Hanlon Quality of Life Scale; Strauss-Carpenter Level of Function scale, and clinical response.) The Quality of Life Interview showed less sensitivity to change and treatment effect, as well as lower correlations with all other measures than the Quality of Life Scale and the Level of Function scale. The latter scales showed high sensitivity to both change and treatment effect, and moderate-high correlations with other measures and with each other. The Quality of Life Scale and the Level of Function scale rater assessments appeared to be substantially more sensitive to subtle change and treatment effects than the patient-reported Quality of Life Interview for clinical trials. Health-related quality of life in schizophrenia is a more heterogeneous concept than previously appreciated.