Dr.RW in his increasing prolific and useful blog recently referenced an article by Dr. Norman Laton, a professor of neurology at Cornell.Laton points out that RCTs are not always needed nor always possible. There is a long list of things doctors do that do not have RCTs supporting their use. A few are: appendectomy for acute appendicitis,inserting a chest tube for tension pneuomothorax,setting a fractured bone,draining an subdural hematoma,taking time to listen to a patient,doing a lumbar puncture with suspected meningitis,getting an ekg on a patient with chest pain. Dr. Laton particularly criticizes EBGs (evidence based guidelines) that consider only RCTs as evidence.
The current EBM gurus as well as those who are credited with the foundational thinking of EBM do not maintain that RCTs and meta-analyses are the only evidence that we can use in determining the value of what we do.They realize that clinical judgment and patient values cannot and should not be cast aside.However,I hear physicians at meetings as well as medical student's comments that indicate they seem to believe that if there is not a RCT to a specific point then the tenets of EBM dictate that that procedure,treatment or whatever cannot be rationally used and to use it anyway will not only waste precious resources but may well be harmful. Here's the thing;There are many things that we do that are correct and which no RCT addresses. Further there are RCTs that are facile and basically empty (such one that compares a new antibiotic or ARB , or ACE or PPI with a barely adequate or inadequate dose of a comparator drug.There are many of these studies that drug companies do to show comparative efficacy with a drug which has proven efficacy).
A similar thought to some of what Dr. Laton said was expressed at a recent CME on hypertension by Dr. Tom Giles, President of the American Society of Hypertension when an audience member questioned some statement Giles made because there was " no evidence based medicine" proving the value of the proposed medical management issue.Paraphrasing Giles's reply.'RCTs can be very useful although some provide little useful new information.And some things may not need a RCT for validation. It would be hard to recruit subjects for a RCT to investigate the efficacy of parachutes to manage gravitational challenge.'