The point of earning a doctorate degree is that you've mastered the subject material and statistics and have been deemed competent to evaluate and contribute new research. Doctors are competent not just to function in their field, but to evaluate ongoing research and incorporate the best of it in their daily functioning. There is a concerted effort to convince the public that physicians are brainless pawns of big pharma and we need guidelines or we wouldn't know what to do. The New York Times frequently displays this bias in their "news" articles. If health care reform is successful at transforming physicians from executive decision makers to technicians the public will suffer. There will be no one standing between a government wanting to save money and a patient who needs care. Guidelines are not always in a patient's best interest.
Guidelines are values based documents. They tell people what they 'should' do. The pretense that they are pure revelation is utterly foolish. Evidence is selected and interpreted according to one's biases. The evidence itself can be manipulated. Close to 90% of all studies confirm the pre-existing bias of the researcher or are favorable to the funder of the study. If someone wants to prove money saving diuretics are just as good as more modern drugs they keep the study short enough so that cardiovascular problems wont show up and they don't bother testing for kidney problems because that would be embarrassing. Pharmaceutical companies keep databases of research sites that give them good results. This ends up skewing the sample so it no longer is representative of the general population. This violates the assumptions of every statistical methodology.
Lets look at an example. If you're doing a study on an antibiotic and have some sites that consistently give you good results, what will we find if we examine the people involved? The people who will give you results showing your antibiotic is effective will be people who metabolize it slowly (so it has higher concentrations and lasts longer in their bodies) and who don't report adverse effects. That may be because they don't have adverse effects or they're not prone to complaining. Think of this and look at Zithromax. Supposedly highly effective and lasts in the body five days after you stop taking it. What happened was that bacteria became resistant to it in record time, just like our professors told us would happen if we used an antibiotic less than ten days. One might guess that lots of people metabolize it faster than the study population. I did an informal study that showed 1/3 of the people on Zithromax for 5 days had relapses between day 7 - 10 after starting it. That's by no means a definitive number because of the informal nature of the study. But it does make you wonder if there are lots of people who should be taking it longer than five days, no matter what the guidelines say. As a physician, I reserve the right to evaluate guidelines and adjust my practice to reflect reality. As a patient, I would think you'd want me to be able to offer you informed choice rather than offer you only treatment that benefits the one making the guidelines.