Analyzing Outcome Score Data

Clinical trial endpoints are often bounded outcome scores which take restricted values on finite intervals and achieve boundary values. Such end points are conceptually ordered categorical variables, but often analyzed as continuous data due to the large number of possible values. The continuous analysis approach may predict the data outside of the natural range and may be biased with skewed data. Several dedicated approaches have been proposed but many confusions exist on their validity, effectiveness, and comparisons. This talk will clarify the confusions, present the most recent advance, and show how to select the appropriate analysis method in practice