Question
A case-control study was conducted to assess whether coffee consumption (high vs. low) is associated with peptic ulcer disease. The results, stratified on smoking status, are summarized below:
Non-Smokers
|
Smokers
| |||||
High
Coffee
|
Low Coffee
|
High Coffee
|
Low Coffee
| |||
Ulcer
|
40
|
65
|
Ulcer
|
150
|
45
| |
No Ulcer
|
45
|
150
|
No Ulcer
|
65
|
40
| |
a. Calculate the stratum-specific odds ratios and the crude odds ratio.
b. Based upon these calculations, is smoking status a confounder in these data? Justify your answer.
c. Is there evidence of interaction?
d. Assess whether:
i. smoking is related to coffee consumption among those without peptic ulcer disease
Ii. smoking is related to peptic ulcer disease among subjects with low coffee consumption.
e. What do your answers in part d tell you about whether smoking is a confounder in this study?
f. Is your conclusion in part e about whether smoking status is a confounder the same as in part a?
Source: 250B Problem Set
