HW6A: Confounding

Please keep the following Scenarios available while you answer the questions:

Scenario A: In a case-control study of the association between low birth weight and subsequent delayed development after birth, researchers fail to consider whether children born to mothers who consumed alcohol during pregnancy might have differences in both birth weight and later development compared to children whose mothers did not drink during pregnancy.

Scenario B: Researchers are interested in studying the relationship between an individual’s dietary intake (exposure) and stroke (outcome). They are trying to figure out whether gender meets the a priori criteria for a confounder to determine if they should control for it in their analysis.  Those criteria are listed here in no particular order:

Criteria 1. The confounder must be a risk factor for the disease

Criteria 2. The confounder must be associated with exposure

Scenario C: A prospective cohort study was conducted to determine the risk of heart attack among men with varying levels of baldness. Third-year dermatology residents conducted visual assessment of baldness of all participants at the start of the study.  Each participant was categorized according to four levels of baldness: none, minimal, moderate, and severe.  When the results are obtained, the risk of heart attack is found to be higher among men who had moderate or severe baldness, compared to men with no baldness.

Scenario D: Data from a study of the association between pancreatic cancer (outcome) and coffee drinking (exposure) (last slide of confounding lecture).