Question 1

Asbestos

Consider the results of a prospective cohort study that followed a group of cancer free quarry employees over 10 years in an asbestos mine. The employees were broken into two groups. The exposed group consisted of miners who were exposed to high levels of asbestos on a day to day basis. The unexposed group consisted of quarry administrators, office workers, drivers and maintenance employees who were never exposed to high levels of asbestos. At the end of the follow-up, the researchers recorded how many employees had developed lung cancer. The results are shown in the table below. The researchers want to know if asbestos is a risk factor for lung cancer.

Asbestos	Lung Cancer +	Lung Cancer -
Exposed +	44	330
Not Exposed -	11	543

The researchers are unsure whether to use relative risk (RR) or odds ratio (OR) to summarise the findings of their study. Which statistic is the most suitable and why?

A Relative risk (RR) is most appropriate because the study has followed employees over time and recorded how many went on to develop lung cancer. Therefore, risk of lung cancer can be estimated and compared between those exposed and not exposed to a risk factor.

B Odds ratio (OR) is most appropriate in this situation because the study compares the odds of being exposed to a risk factor between cases who have already developed lung cancer and controls without lung cancer.

C Both relative risk (RR) and odds ratio (OR) are equally appropriate as the researcher is looking at the association between a risk factor and a disease.

D Neither statistics are appropriate. The researcher should only use the Chi-square test of association. The Chi-square test of association looks at the relationship between two categorical variables.

Question 2

Calculate the appropriate statistic you selected in the previous question for the Asbestos example. (Answer correct to 2 decimal places)