The tobacco industry has everyone obsecessing about whether the IARC study reaches "statistical significance" at the 95% level. Another way of looking at this issue is whether the 95% confidence interval for the true risk includes 1. The 95% confidemce interval for the study is for exposure to spousal ETS was 1.16 (95% confidence interval [CI] 0.93-1.44); the OR for exposure to workplace ETS was 1.17 (CL 0.94-1.45). As the industry notes, these intervals include 1.
It is important to note that the true risk could be ANYWHERE inside the confidence interval (with 95% confidence). This means that we can be just as confident that the true risk is 1.45 as that it is 1.
Most governments use the upper 95% bound in the confidence interval as the measure of risk for policy purposes in order to be health protective.
It is time that tobacco control started to do the same thing. The issue should be how HIGH the risk could be (the upper end of the interval), not how low (the bottom, which the tobacco industry has everyone talking about).
Comments on this posting?
Click here to post a public comment on the Trash Talk Bulletin Board.
Click here to send a private comment to the Junkman.