|
Abstract:
|
As prevalence of an outcome changes, binary measures of differences between rates of experiencing or avoiding it tend to change as well. As an outcome like mortality declines in prevalence, relative differences between rates of experiencing it tend to increase while relative differences between rates of avoiding it tend to decline. Absolute differences between rates and odds ratios also tend to change, though in a more complicated manner. As the outcome changes from being nearly universal to being rare, absolute differences tend to increase for a time and then decline, while odds ratios decrease for a time and then increase. This session would explain these tendencies and the obstacles they create for appraising the size of group differences in settings of different prevalences, and explore two approaches to measuring group differences that avoid those obstacles.
|
