Ans. is 'd' i.e., The attributable risk of breast cancer resulting from the pill may be directly measured AR calculation requires incidence which can be obtained from only a coho study (Not from a case control study) o In a Case Control Study, 'Cases' are diseased and 'Controls' are healthy o Controls should be similar to Cases in all respects (for ensuring comparability) o Cases should be matched with controls for all factors 'EXCEPT for the (risk) factor under study' (otherwise the etiological role of risk factor under study, would be eliminated from the study, since both groups are exactly similar in all respects); So if controls do not exclude women known to be taking the pill at the time of the survey, both groups will become similar in respect to risk factor (contraceptive pill) under study and no relationship can be established with breast cancer.