Comparing the quality-adjusted life-year output of two treatment arms in a randomized trial

The quality-adjusted life year (QALY) is an outcome measure with two dimensions: wellness and time. It is difficult to compare the QALY output of two groups when subjects are observed for varying time periods, because both elements of the two-dimensional outcome measure vary in all subjects. This paper evaluates the strengths and weaknesses of five approaches for making these comparisons. These approaches are: a subject-based approach, the elimination of time, calculation of the mean of available data, use of quality-adjusted survival analysis, and calculation of a maximum likelihood estimation. It is concluded that the maximum likelihood estimation offers significant advantages over the other methods.