BACKGROUND: Health-related quality of life (HRQOL) of haemodialysis (HD) and peritoneal dialysis (PD) patients has been assessed with health profiles and health preferences methods. Few studies have used both types of HRQOL instruments. The main objective of this study was to assess the relationship between information from the two types of HRQOL instruments in dialysis patients. METHODS: We interviewed 135 patients, using two health profiles (Short Form 36 and EuroQol/EQ-5D) and two health preferences methods (Standard Gamble and Time Trade Off). Socio-demographic, clinical, and treatment-related background data were collected from patient charts and during the interview. Relationships between the outcome measures were assessed with Pearson correlation coefficients. Multiple regression models were used to study the relationship of HRQOL outcomes to background variables. RESULTS: The HRQOL of dialysis patients as measured with health profiles was severely impaired. The health preferences scores were higher (0.82-0.89) than scores previously reported in the literature. Correlations between health profiles and health preferences were poor to modest. HRQOL outcomes were poorly explained by background characteristics. Differences between HD and PD groups could not be demonstrated. CONCLUSIONS: Health profiles and health preferences represent different aspects of HRQOL. An impaired health status may not be reflected in the preference scores. Coping strategies and other attitudes towards health may affect the preference scores more than they influence health profile outcomes. The added value of health preferences methods in clinical research is limited.
BACKGROUND: Health-related quality of life (HRQOL) of haemodialysis (HD) and peritoneal dialysis (PD) patients has been assessed with health profiles and health preferences methods. Few studies have used both types of HRQOL instruments. The main objective of this study was to assess the relationship between information from the two types of HRQOL instruments in dialysis patients. METHODS: We interviewed 135 patients, using two health profiles (Short Form 36 and EuroQol/EQ-5D) and two health preferences methods (Standard Gamble and Time Trade Off). Socio-demographic, clinical, and treatment-related background data were collected from patient charts and during the interview. Relationships between the outcome measures were assessed with Pearson correlation coefficients. Multiple regression models were used to study the relationship of HRQOL outcomes to background variables. RESULTS: The HRQOL of dialysis patients as measured with health profiles was severely impaired. The health preferences scores were higher (0.82-0.89) than scores previously reported in the literature. Correlations between health profiles and health preferences were poor to modest. HRQOL outcomes were poorly explained by background characteristics. Differences between HD and PD groups could not be demonstrated. CONCLUSIONS: Health profiles and health preferences represent different aspects of HRQOL. An impaired health status may not be reflected in the preference scores. Coping strategies and other attitudes towards health may affect the preference scores more than they influence health profile outcomes. The added value of health preferences methods in clinical research is limited.