Reply to the comment below

As I worked through the chapter questions, one of the most thought-provoking and challenging problems was Chapter 3, Question 6, which examined an outbreak of Legionnaires' disease during the 1976 American Legion convention. This question involved calculating and interpreting cumulative incidence, incidence ratio, incidence difference, and attributable proportionnot just numerically, but conceptually.

I initially had difficulty distinguishing the nuances between cumulative incidence ratio and attributable proportion, particularly in applying them in an outbreak investigation context. While cumulative incidence ratio (CIR) tells us how much more likely delegates were to get sick compared to nondelegates (in this case, over 17 times more likely), the attributable proportion gave deeper meaningit quantified that over 94% of the risk among delegates could be attributed to attending the convention.

This raised broader public health implications. For example, how do we interpret such measures to guide epidemic response, and how do we separate association from causation in real-time outbreaks? I also wondered how we handle potential biases, such as underreporting of illness among nondelegates or differences in exposure intensity.

I'm curious to hear how others interpret these risk measures beyond the mathespecially how you would communicate these findings to stakeholders like public officials, event organizers, or the general public.

Has anyone struggled with the difference between "cumulative incidence difference" and "attributable proportion"? Or have you encountered real-world examples where such epidemiologic data led to policy changes or interventions?

Looking forward to your perspectives!