Boiled chicken, huh? Well, since we're in a critical mode, let me point out two additional shortcomings of the health sites we've been talking about (neither of which, by the way, will dissuade me from recommending patients to them, in particular the National Library of Medicine site--I still believe it performs a valuable service for most patients).
The first problem concerns the paucity of information about rare diseases, the so-called zebras of medicine. A person diagnosed with PNH (paroxysmal nocturnal hemoglobinuria), for example, the disease I had, will find either nothing (OnHealth.com) or a brief, encyclopedic discussion of the condition (NLM, drkoop.com). The latter, while mentioning a few signs and symptoms, won't likely be satisfying since it doesn't convey what PNH really means for them or what might happen to them. Not that an expert in the field, however, would be able to give accurate predictions either; there are simply not enough patients to generate good data. Now I wonder, should the cloud of uncertainty that hangs over zebra-conditions like PNH--and the truth is, similar clouds can be said to surround lots of more common diseases as well--be publicized on consumer-oriented Web sites? As a doctor, I've become inured to the fact that we don't know as much as the layperson thinks we know about many things. But being "reintroduced" to that fact as a patient was terrifying. At least in the office a doctor can get a sense of what type of patient he or she is dealing with before deciding to confess his ignorance.
My other area of concern involves conditions for which there are no clear-cut guidelines for care. For example a friend of mine was recently diagnosed with chronic myelogenous leukemia (CML). Unlike patients with, say, breast cancer, where management is pretty much standardized, the treatment for CML varies from institution to institution; some centers will recommend a bone-marrow transplant, others chemotherapy; and those advising the latter often offer different types. Medical information sites will list and explain each of the treatment options but won't really help a patient choose between them. Then again, as you pointed out before, that probably shouldn't be the role of the Internet anyway--to enable patients to become doctors. And that's what I told my friend: You're at Johns Hopkins, a great institution, I said; you have a great doctor who has given you his recommendation; you've done all your homework (including countless Web searches); there's nothing left to do but make your decision. Just as I had done five years ago in a similar situation.
But, thankfully, my friend wasn't like me and didn't play the part of model--i.e. passive--patient. Unconvinced there was nothing left to do, he didn't stop at Johns Hopkins but sought out second and third opinions. Nor did he stop at drkoop.com and NLM but rather continued searching on the Web until he finally found what he was looking for: on a leukemia support group chat room. And this brings us to something you mentioned earlier, the art of being a patient. My friend could be said to be Picasso in that department.
He learned--not from doctors but from other patients --about an experimental drug that was having remarkable preliminary effects in more advanced stages of his disease. He also learned who the principal investigator was and traveled to Oregon to meet with him. Although his original doctor told him it could be a year, or years, before the drug would be available, my friend is scheduled to begin a study in 2 months, largely through his own efforts as well those of other patients. According to a recent Wall Street Journal article (6/6/00), patient activists like my friend have, with the help of the Internet, been able to put enough pressure on the pharmaceutical giant Novartis to increase production of their experimental compound for CML and speed up clinical trials (even though it may not necessarily be in their financial interest). Imagine that! I suspect this is what you meant, Atul, when you talked about the potential of the Internet to transform medical care.
So perhaps you're right that patient-supported Web sites as opposed to physician-supported ones may ultimately have the most impact on the practice of medicine--in both positive and negative ways. While my friend's story turned out well (though it is far from over), I imagine other patients might just as easily find bad advice and get involved in things they probably shouldn't.
Times are a' changing pretty fast, hey? I can't wait to see what cyberspace has in store for us down the road.