Questions 1-5: Researchers were interested in investigating the association between tobacco use and head and neck cancer. They obtained alcohol and tobacco use self-reported data in 1410 individuals with head and neck cancer and 3488 individuals without head and neck cancer. They hypothesized that alcohol is an effect modifier of the relationship between tobacco use and head and neck cancer. Below is the data that they obtained: Table 1. Head and neck cancer overall Individuals with head and neck cancer, n Individuals without head and neck cancer, n Alcohol use alone 202 800 Tobacco use alone 408 868 Tobacco and alcohol use 650 830 Total 1410 3488 Question 1. Name the specific type of study design and the appropriate measure of association. Question 2. Based on the information provided, identify and explain a potential source of information bias in the study. Question 3. Evaluate the association between smoking and head and neck cancer, and interpret the result. Question 4. Based on the above data and the study design in Question 1, is alcohol use a confounder or an effect modifier of the smoking-head and neck cancer association on the multiplicative scale and additive scale, using both homogeneity-heterogeneity and observed- expected values approaches? Show how you assessed this. Question 5. How would you report the association of smoking with head and neck cancer? Why?

References

Szklo,M.,Nieto,F.J.(2014).Epidemiology: Beyond the Basics.United States:Jones & Bartlett Learning., Chapter 6