Aim: The assessment of quality of life with a single index, which is necessary for cost-utility analyses, causes a loss of information. Therefore the suitability of such indices for mental health and non-mental health conditions was analysed and their ability to detect effects of medical interventions was examined.
Method: Quality of life of 18 schizophrenic patients and 26 patients with diabetes mellitus was rated three times using disease specific instruments (BELP-KF, DMLQ), visual analog scale (VAS), EQ-5D and time trade-off (TTO).
Results: Effects of medical interventions were only detected with the disease specific instruments and EQ-5D. Using TTO and EQ-5D, there were significant, but reverse differences in the quality of life of patients with diabetes and patients with schizophrenia. EQ-5D correlated highly with DMLQ, but lowly with BELP-KF.
Conclusions: TTO and VAS were insensitive to effects of medical interventions. EQ-5D detected treatment-induced changes, but was inappropriate as a measure of quality of life in schizophrenic patients.