Comparison of health-related quality of life preferences between physicians and cirrhotic patients: implications for cost-utility analyses in chronic liver disease.

Accurate assessment of utilities to calculate quality-adjusted life expectancy for medical interventions is needed in cirrhosis. To date, limited data exist in cirrhotics and are generally physician-assigned. Therefore, our aim was to determine utilities for six clinical scenarios in cirrhosis and to define if differences exist in utilities assigned by physicians versus patients. We administered a questionnaire to 83 physicians and 114 cirrhotics to obtain utilities using the time trade-off method for (1) compensated cirrhosis, (2) decompensated cirrhosis, (3) encephalopathy, (4) spontaneous bacterial peritonitis, (5) variceal bleeding, and (5) hepatocellular carcinoma. On a scale from 0 (death) to 1 (perfect health), mean utilities of physicians and patients were compared using the Student t test. One-way analysis of variance was used to compare the utilities between patients according to Child-Pugh class. Statistical significance was defined as a P value <0.05. The mean age of the physicians was 42 +/- 11, with 52% being male. The mean age of the patients was 52 +/- 9; with 59% male. The mean Child-Pugh score was 8 +/- 2 and HCV was the most common etiology (54%). The mean utilities for physicians and patients were as follows: CC, 0.78 vs. 0.88; DC, 0.55 vs. 0.74; E, 0.38 vs. 0.55; SBP, 0.33 vs. 0.45; VB, 0.27 vs. 0.40; and HCC, 0.19 vs. 0.30. All comparisons were statistically significant. Although physicians and patients assigned similar relative rankings to each health state, physicians assigned utilities were significantly different from those assigned by patients. These results suggest that studies that have used physician-assigned utilities do not accurately reflect patient preferences.