TESTING THE TEST FOR BREAST CANCER
Let's begin with a deceptively simple question: Do mammograms work? Obviously, to answer this question someone has to decide what counts as "working." At one end of the spectrum, the question might be: Do mammograms find small cancers? At the other end of the spectrum the question would be: Does undergoing regular mammography lower a woman's overall chance of death (e.g., dying within the next 10 years)?
Although I'll argue that this last question is one of relevant (and the one that patients hope doctors can give the answer to), let's pinpoint the slightly narrower question that the major studies have focused on: Does undergoing regular mammography lower the chance of death from breast cancer?
In the early 1960s, the Health Insurance Plan of Greater New York carried out the first randomized trial of mammography, known today as the HIP study. Because all the mammography studies I'm going to discuss are randomized trials, I need to review exactly what that means.
In its simplest form, a randomized trial is an experiment in which participants are assigned to 1 of 2 groups e.g., mammography versus no mammography solely by chance. (Although the approach is essentially like flipping a coin, the groups are in fact determined on the basis of random numbers generated by a computer.)1 One group undergoes the "intervention" the therapy or test being studied, in this case mammography. The other group, called the control group, does not no mammography.
Randomization is the easiest method to construct two groups of subjects that are similar in every way except one whether they experience the intervention. Thus, groups selected at random are similar not only with respect to factors we know put people at risk for breast cancer death (e.g., age, menstrual history, family history of breast cancer), but also with respect to factors we have yet to learn about (e.g., the yet-to-be-discovered "cancer promoter gene" on chromosome 13). Which means that any differences observed after the study has begun must be the result of the intervention. A randomized trial, in short, is the only way to make sure that a screening test works.
Good news
In the HIP study, 62,000 women were selected from the health plan patient list and were randomized into one of two groups by investigators. (The women were never aware of the randomization.) The intervention group was invited to have annual clinical breast exams usually performed by a surgeon and annual mammograms. The control group remained unaware that they were part of a study and continued to receive their usual medical care (during the time, no annual clinical exam and no mammogram). After 10 years of follow-up, the breast cancer death rate in the intervention group was 23% lower than that in the control group.
Almost everybody assumes that this reduction was due to mammography. But as you can tell from the design, the study really tested the combined effect of the clinical exam and mammography; indeed, it is impossible to discern the independent effect of either. It may be that mammography saved all the lives and the clinical exam added nothing, or the reverse could be true. No-one can tell. All we can say is that women receiving the clinical breast exam and mammography did better than women receiving standard 1960s medical care, when the breast was largely ignored.
In 1979, 45,000 women entered a trial in Edinburgh, Scotland, that, like the HIP study, evaluated the effectiveness of mammography plus clinical exam versus usual care. Unfortunately, instead of randomizing the 45,000 women, the researchers randomized the 84 general practices that cared for them. This failed to produce equivalent groups of women. Women in the control group were poorer and 20% more likely to die from reasons other than breast cancer.
After seven years of followup, women cared for in practices that provided clinical exams and mammograms were reportedly 17% less likely to die from breast cancer. However, because the women were so different otherwise, one could argue that the true effect of these screenings was considerably smaller or considerably larger. In the long run, all we can say is that clinical breast exam and mammography might be better than the standard medical care. Given these uncertainties, I won't consider the Edinburgh study further.
At just comparable time, four randomized trials of mammography involving 280,000 women were carried out in Sweden: one in two rural provinces and the other three in the cities of Stockholm, Malmö, and Göteborg. As in the HIP study, women were unaware of their inclusion in a randomized trial. The intervention group was invited to undergo mammography (every 18 to 33 months); the other group continued usual medical care. After five to 13 years of follow-up, women in the mammography group had a 24% reduction in the breast cancer death rate.
This time there was no confusion: it was mammography versus nothing. The sound bite was clear. "Mammography works."
Our website is not responsible for the information contained by this article. Webworldarticles.com is a free articles resource thus practically any visitor can submit an article. However if you notice any copyrighted material, please contact us and we will remove the article(s) in discussion right away.
This article was sent to us by:
Fiona Landman at
08152010
1. URINARY SEPTICAEMIA: TREATMENT
All articles in this directory are property of their respective authors. Additionally, read our Privacy Policy
© 2010 WebWorldarticles.com - All Rights Reserved. Partners: Gunblade Saga