Question

Epidemiology: confounding

(10) In the early 1940s, epidemiologists and public health officials set out to prove hat smoking causes lung cancer. Epidemiologists designed studies, analyzed data, and constructed logical arguments to demonstrate this relationship. Cigarette companies argued that association does not equal causation. Using the concept of confounding, explain how the cigarette companies could have been correct. (Make sure to explain what confounding is and use examples specific to this situation).

Answer 1

Confounding variable is associated with both the dependent variable and independent variable and causes spurious association between them. For example goitre and high altitude are significantly associated. Similarly, low iodine content in water and high altitude are associated. Here, high altitude acts as the confounding variable. Goitre is actually caused by low iodine content in water. Both of these variables are associated with high altitude which acts as a confounding variable.

Analytical studies such as case control study and cohort study illustrates the association between two variables. To establish the causation, randomised controlled trials should be done. Therefore, cigarette companies argued that association does not always mean causation.