Evaluating disease management program effectiveness: an introduction to survival analysis

Currently, the most widely used method in the disease management industry for evaluating program effectiveness is the "total population approach." This model is a pretest-posttest design, with the most basic limitation being that without a control group, there may be sources of bias and/or competing extraneous confounding factors that offer plausible rationale explaining the change from baseline. Survival analysis allows for the inclusion of data from censored cases, those subjects who either "survived" the program without experiencing the event (e.g., achievement of target clinical levels, hospitalization) or left the program prematurely, due to disenrollement from the health plan or program, or were lost to follow-up. Additionally, independent variables may be included in the model to help explain the variability in the outcome measure. In order to maximize the potential of this statistical method, validity of the model and research design must be assured. This paper reviews survival analysis as an alternative, and more appropriate, approach to evaluating DM program effectiveness than the current total population approach.