Name:   KEY --
 
 

This is the first paragraph of a story that appeared under the heading "Science in Brief" in the LA Times in May, 1997.

"A new study in the May 1 New England Journal of Medicine provides some of the strongest evidence yet that regular exercise helps protect women from breast cancer. The research, conducted in Norway, found that women who exercise at least four hours a week have a breast cancer risk about one-third lower than usual."

1. (5 pts) Do you think this was an observational or a controlled study? Explain. Make sure your indication indicates that you know what an observational and controlled study are.

The test to determine whether a study is observational or controlled is this:  did the researchers assign subjects to treatment groups?  If yes, then its a controlled study, if no then its observational.  By this definition, it is possible for a study without placebos, double blinds, or randomization to be a controlled study. It might not be a good study, but still it is a controlled study.  Also, note that even observational studies can have control groups.

It is very unlikely, probably unethical, and almost certainly impossible, to force some women to exercise long enough to observe breast-cancer rates.  So we can assume that the women themselves chose which treatment group they would be in.  (Note that there are at least two groups: those who exercise more than 4 hours, and those who exercise less.  Those who exercise less might be thought of as belonging to the "control" group, although probably the term "comparison group" is a little mroe accurate.)
 

2. (5 pts) Do you think the evidence produced by this study justifies the conclusion that regular exercise helps protect women from breast cancer? If so, say why. If not, give an example of a confounding factor.

No, the evidence suggests that this is merely a possibility, but not by any means a certainty.  One possible confounding factor that would explain away the relationship might be called "lifestyle."  Women who are concerned with health probably exercise frequently and eat a careful diet.  Possibly, this diet helps protect them from breast cancer.

Note that the key here is to explain how the evidence could have come out the way it did (women who exercise more have less breast cancer) and NOT be due to exercise.  For this to be the case, your explanation must do two things: it must explain why it causes some women to exercise, and it must explain why it makes these women less susceptible to breast cancer.

Several people noted that because the study was conducted in Norway, this was reason enough to negate it.  However, this is only reason enough to prevent us from generalizing to a population larger than Norway.  The question still remains, does exercise prevent breast cancer in Norway?  Because if it does it there, from what we know about biology, it probably does it everywhere.