It is true that guidelines change. By the same token, though, all knowledge changes. There are different ways to model the function of knowledge change. Are we asymptotically approaching truth? Does each generation of scientists invest in a new explanatory model, which is then discarded some time down the line in favor of another - a new paradigm, not necessarily closer to the truth? Or are scientists continually confronting new problems, with different narratives, so we're not so much finding new answers as dealing with new questions?
All these possibilities apply to medicine. Guidelines for, say, the optimal control of blood pressure don't change simply because we have a better idea of what the perfect blood pressure is. We have a different array of blood pressure treatments than we did just a few years ago; we have a different understanding of the relationship between systolic and diastolic blood pressure; we think a lot more about patient preferences than we did 10 or 20 years ago.
Where does this leave us? I hope not with widespread disillusionment that medicine, after all, does not inexorably march towards truth and health. Like any other empirical caravan, we trundle along for a while, get lost, find a new byway, and discover that we weren't lost at all, and now we are in an even better place than we thought possible. Or we discover that the folks with us are not merely passengers along for the ride, but they know how to drive as well as we can.
When you hear about changing health care, you might be worried about a loss of stability. I would say that understanding the world requires constant change, in a world of flux.