Wednesday, January 11, 2006
US researchers found the prostate specific antigen (PSA) test did not improve survival rates.
The researchers found men screened using the PSA test died from the disease at about the same rate as those who were not checked.
...The researchers found that combining the PSA test with a digital examination produced even worse results.
Now that's interesting, because we always tell people that they should have both a digital exam (done by feeling the prostate with a finger inserted into the rectum) and a PSA to screen for prostate cancer. The abstract doesn't make clear exactly how badly the combined approach faired or why. (The full text needs a subscription, which I don't have.)
The value of the PSA as a screening test has long been in question. However, it continues to be commonly used because men have come to expect it. Doctors are always told that they need to discuss the pros and cons of the test and all of its uncertainties with patients before performing it, but I've found that one of the most difficult subjects to convey to a patient - even when I've spent the whole visit discussing only that topic. I have yet to meet someone who has completely grasped it. Or maybe I just haven't discovered a good way to convey the message.
In any case, people who are prevention oriented always end up having it done. Those who are prevention phobic don't. And in the back of my mind, there's the always the question of how the patient or his family will react if we don't do the test, then he ends up with prostate cancer some years down the line. Too often, the response will be - "that doctor told me I didn't need the test. I should have listend to my friends (or the health news) instead." It's a very real concern, as this case illustrates. It's one of those tests that gets ordered more for legal reasons than for medical reasons.
posted by Sydney on 1/11/2006 08:50:00 AM 5 comments
Logically, PSA testing has not been
By 6:23 PM, at
I've found that one of the most difficult subjects to convey to a patient - even when I've spent the whole visit discussing only that topic. I have yet to meet someone who has completely grasped it.
By 9:03 PM, at
Case-control studies are OK for rare diseases, not for a common disease like prostate cancer. Where's the cohort study?
By 1:06 PM, at
The logical explanation, em, is the variable natural history of the disease. Early diagnosis does not equal higher rates of cure. You may be treating tumors which might never be clinically significant, as well as those which may already have microscopic metastases. No treatment benefit in either case.
By 2:08 PM, at
At autopsy 67% of men over age 80 have prostate cancer. But only a small fraction of that will die of it. The challenge of prostate cancer detection is preventing deaths, but we do not really understand who will die of prostate cancer and who will just be "carriers." Until we can separate out the aggressive from the non-aggressive cancers, or at least stratify risk, we will be shooting in the dark.