The new recommendations on screening for breast and cervical cancer have been a hot topic here and elsewhere in the last few days. The whole subject of what is appropriate in screening is a tricky subject and the new recommendations are easily politicized and distorted precisely because most people have a hard time understanding the nature of the debate. I am not by any means expert in the area of either of those cancers, but I do have some knowledge in an analagous field and want to share some of that knowledge to try to help folks understand the subject a bit better.
The following is part of the text of an email I sent to a friend this morning after he had sent me something related to the new recommendations:
Neither of the groups that has issued these recommendation changes has been a governmental group and neither has been basing their recommendation on cost. These are groups of professionals basing their recommendation on what they believe is in the best interest of patients. That's not to say they are necessarily "right" or even that there necessarily is a "right" on issues like this, but it is absolutely true that more care is not always better care, and more screening is not always better screening.
To go into that last a little deeper: Every screening test has a certain rate of false positives - people who test positive but do not actually have the disease - and a certain rate of false negatives - people who have the disease but don't test positive. False positives are a problem because a positive test usually results in further testing and even if you ignore cost totally, which is not responsible in the real world, the follow-up tests tend to be invasive and like all invasive procedures they have a complication rate. Some complications are minor, and some can be major, depending on the test. One of the well established ways to reduce false positives is by selection of the population you screen.
I'll use an example from the field I know best. The standard screening test for Coronary Disease is the Treadmill Stress Test. In any population that you test there will be some false positives, but the lower the risk in the population the higher the percentage of the positives that will be false positives. If I stress a group of 60 year old, overweight males with high cholesterol, I'll get quite a few positives and relatively few of the positives will be false - though some still will. On the other hand, if I stress a group of apparently healthy college students, I will get far fewer positives, but nearly 100% of the positives will be false.
A positive treadmill usually leads to a cardiac catheterization. The rate of serious complications has been pushed down pretty low, but a certain predictable percentage of the people who have a cath will have a major bleeding episode or damage to a peripheral artery requiring surgical repair, some will suffer kidney damage, a very few will have a stroke or heart attack, and a tiny number will die, all as complications of the test. So in very low risk populations, it is irresponsible and actually harmful to do screenings that result in further testing. You will demonstrably hurt more people than you help.
That's not to say that I agree with the recommendations they are making, because I don't know enough about those fields to make an informed judgement about it. But as far as I can tell the new recommendations do represent physicians who are trying to do what they believe - rightly or wrongly - is best for patients.
That tells the story pretty well, but I'll add some additional notes on another subject on which I am not expert: Prostate cancer screening via the PSA test: This is another closely related situation, with an added twist or two. An elevated PSA level can signal prostate cancer and for some percentage of men, finding prostate cancer early can be lifesaving. But this test like any other, will have a certain rate of false postives. Meaning some of the men who test high won't actually have prostate cancer. But with this disease theres an extra complication, since quite a number of prostate cancers are very slow growing and not life-threatening. It's a common saying that a great many men will die in old age with prostate cancer, but not of prostate cancer. Many prostate cancers are diagnosed at autopsy after the man has died of something else.
So, with widespread screening for PSA, some men will have their lives saved by early detection, some will undergo a somewhat unpleasant biopsy procedure to find that they don't actually have cancer, and some will find a cancer and undergo surgery to treat the cancer, even though the cancer would never have troubled them. All of the treatments for prostate cancer have a certain rate of pretty unhappy complications - mostly impotence and incontinence. It's not a simple subject and not one that lends itself to soundbite discussions in a heated political atmosphere. Politicizing it from any point on the spectrum is highly irresponsible. I hope this is of some help in understanding the debate a bit better.