Aelia Jusa |
03-03-2007 05:44 PM |
Quote:
Originally posted by robertthebard:
So by this test, they can, in spite of the fact that not all of the other 1/2 of the test subjects developed the cancer, say that the only reason those women didn't get cancer was because of the vaccine? It seems to me that the number of women who did get the cells for cancer would have had to be a lot higher to take into account the fact that not all women get cervical cancer. To me, it seems a bit misleading to say that the only reason those women didn't get the cancer was because they were vaccinated. Maybe it works, maybe not, I don't know, but I don't think that one real study constitutes proof. How many studies did they do with Viox to prove it was safe, only to wind up in court because it wasn't? You'll notice, in the linked article, that stock prices went up after they released the study results.
|
Well, whether it is safe and whether it does the thing it is meant to are two different questions. Tests for Viox would have showed that it did the thing it was supposed to (sorry but I can't recall what it was for - arthritis?) but were clearly not sufficient to demonstrate long term side effects. Similarly the tests for this vaccine are aiming to demonstrate that it prevents cervical cancer by immunising against HPV. Not necessarily that it is safe in other respects.
If a large number of the control subjects developed cervical cancer then that would indicate that there was clearly something screwy with the way they recruited participants. The risk of developing cervical cancer is very low overall. So the small number of women developing the cancer in the control group would be representative of the population risk of cervical cancer. The point was not to show that lots of women get cervical cancer when they DON'T get vaccinated, but that no (or at least many fewer) women get it when they DO get the vaccine. Which was shown.
They have two very large groups which are essentially identical in every way - general health, age, risk of cancer, etc. This is achieved through random assignment to control and vaccine groups. Therefore you can assume that since everything else is equal, if neither of the groups had been given the vaccine then after a period of time there would be equal numbers of women from both groups developing cervical cancer - we can be confident in assuming this because the groups are the same in every significant way. So then if one group is given the vaccine, and the other isn't, and the control group has a significant number of women developing cancer and the vaccine group has no one with cancer then the likelihood of this result occurring purely by chance, i.e. it would have happened without the vaccine, is very low. Much lower than the likelihood of the result occurring because of the vaccine. Statistical analyses would have been run in this test to determine the probability of this result occurring by chance. For a result to be considered significant in studies like these, the probability of the result being due to chance not the vaccine would be extremely low (e.g., 1% or lower).
Of course, support for a result in a study like this is much stronger when it has been replicated in other studies. Although I think Bungleau indicated that a number of studies have been conducted showing the same pattern of results, not just one. However, provided that the methodology in this study was sufficiently rigorous (i.e. proper random assignment, appropriate pre-tests, etc.) then this result is very strong on its own. This is how all drug trials are conducted. It is definitely possible to be wary of this vaccine because there have not been sufficient studies to demonstrate possible long term side effects, or whether the effect lasts over a person's lifetime. However, you should not be wary of whether the drug does what it supposed to do, i.e. prevent HPV and therefore cervical cancer, because the tests have shown that it does.
|