tag:blogger.com,1999:blog-2843486188833087449.post1389927453410567258..comments2024-01-12T05:48:42.480-05:00Comments on FasterCures Blog: Magical ThinkingFasterCureshttp://www.blogger.com/profile/00119318415684617652noreply@blogger.comBlogger2125tag:blogger.com,1999:blog-2843486188833087449.post-68856482622474045792008-10-01T23:31:00.000-04:002008-10-01T23:31:00.000-04:00how does "personalized" medicine meet the requirem...how does "personalized" medicine meet the requirements of "evidence-based" medicine that are a precondition for coverage by insurers -- both private and public?asnethttps://www.blogger.com/profile/09398741708012871395noreply@blogger.comtag:blogger.com,1999:blog-2843486188833087449.post-11342963765339393252008-09-23T14:52:00.000-04:002008-09-23T14:52:00.000-04:00Of course we can do it, but not just by talking ab...Of course we can do it, but not just by talking about it, which is what we have so far gotten from Critical Path, for example. Change in applied science comes at the nuts and bolts level. <BR/><BR/>How are we going to run clinical trials of "one?" Personalized medicine requires it, and we could already be doing it, using the patient's baseline molecular and macro characteristics as the control, and then observing very closely (through sequential sampling and characterization of the patient's affected and unaffected tissues and detailed monitoring of clinical outcomes) what happens as we treat the patient, then adjusting to changes in the patient as they occur. This would also be a lot closer to the ethical practice of medicine than what we are doing now, and would be more attractive to patients.<BR/><BR/>In fact, it would be more scientific, more ethical and more effective than what we do now (e.g., forcing dying patients into blinded, no cross-over, placebo-only control arms so they will die on the schedule of an untreated patient to provide a column of "survival" data to FDA statisticians). <BR/><BR/>We would of course have to conduct many clinical trials of "one" to learn what we need to know about variability. Pooling the results into populations for statistical analysis after characterizing the cause and effect relationships of drug/disease interaction would then be of very little use in making progress, because it would be applicable only to some calculated "average" patient - which we would know did not exist from the indivdualized data we collected - or we would know the population did exist - based on the cause and effect observations - also rendering the statistics meaningless. <BR/><BR/>We can move to the pursuit of first-principle science because we now have the technology to do it. But first we have to get past the seeming unshakeable belief by the clinical research and regulatory communities that only outcomes in "populations" matter. The science is telling us, in fact screaming at us, that population-based medicine is wrong for many of the diseases that still kill us. The right direction, however, treads perilously close to the statistically-disdained world of "anecdotal" observations. Nonetheless, that is where we have to go, because the populations we have been relying on for 50 years simply don't exist at the molecular level. <BR/><BR/>We actually don't have a science problem. We can do the science. We have a cultural problem. Our clinical researchers and regulators don't believe in first-principal science - they believe only in statistics derived from population-based outcomes. <BR/><BR/>Stalemate.<BR/><BR/>The real challenge is breaking the cultural resistance to change that permeates medical/clinical research and government regulation of that research. <BR/><BR/>A good place to start would be excising the term "well-controlled" from the FD&C Act, which mandates that all clinical research must be designed to produce a statistical result. What other field of science has managed to move forward using that comnvention? The answer is - none.<BR/><BR/>Imagine Congress mandating in any other field of science (as they have in clinical research) that there is only one way to design an experiment, and that way is to call a statistician? <BR/><BR/>Let's let the scientists do some science for a change. Science moves forward by itself if we give it enough room to move. The current crop of status quo defenders sense the truth in that, and their efforts to make sure that our system doesn't change in any fundamental way is a reflection, in part, of that sense. No one wants to be made obsolete, but that is the inevitable effect of progress. Keep up, or become irrelevant. But when the powerful institutions responsible for fostering change become the opponents of change (the current situation), progress can't happen. <BR/><BR/>The answer to this question posed in this article is quite simple. We need to get out of our own way.Anonymousnoreply@blogger.com