QALY MODEL: Test of Utility Independence when health varies over time

Objectives: In recent years, economic evaluations of health care interventions are widely undertaken in order to inform policy makers in the health care sector about the cost-effectiveness of the old and new technologies. Regarding the benefits of these technologies, the QALY model is the most used measure for utility. It is an index that combines quality and quantity of health after the elicitation of people’s preferences over the interventions, which are going to be compared. For QALY to be a valid instrument when elicit people preferences, some fundamental principles should hold. One of them is Utility Independence. There is empirical evidence, which supports QALY in regards to utility independence for chronic health states. However, there is limited evidence when health varies over time. Method: This study consists of two parts. In the first part, three tests of utility independence are conducted by using the standard gamble method to elicit people’s utilities over various health profiles. The health states that were used to form the profiles were taken from EQ-5D-5L. After the elicitation of utilities we did two statistical tests, one paired t-test and one Wilcoxon test in respect to each utility independence test. The second part of our study consists of a questionnaire in order to investigate sequencing effects based on people’s general preferences over mild and severe health states. Results: In our first part of the study where utility independence was tested, statistical analysis showed that utility independence holds, since there were no statistically significant differences between the utilities for each test. Regarding the qualitative research through the questionnaire we created, we explored some sequences effects, such as adaptation to mild disability. We believe that this happened due to limitation in the procedure of elicitation. Discussion: Utility Independence is supported through our study; however, we should consider some limitations that don’t allow us to conclude over the validity of the instrument. A relative small sample (n=30), the method we elicited utilities (SG), the difficulty of our sample to involve in such a complex experiment and the puzzling results of our qualitative part where some sequences effects were found, lead us to the conclusion that more research should be done for the QALY in order to be certain of its validity, especially in the field of sequencing effects on people’s preferences.