There are a couple of ways by which people may fail to practice evidence based medicine. Some neonatologists, and physicians in general, fail to practice it by using therapies that have not been adequately tested for safety and efficacy. An example of this might be the liberal, unchecked use of high amounts of oxygen in the early days of neonatology that resulted in an epidemic of retinopathy of prematurity. (I'll post about another example in the next week or so.) At other times physicians don't practice evidence based medicine because they don't know the latest evidence. Neonatologists who continued to use steroids for BPD, even after studies showing their detrimental effect on neurodevelopmental outcomes, fall into this category. I know a couple of them.
Sometimes, though, we would like to practice evidence medicine but cannot, simply because evidence does not exist or is not conclusive regarding a given situation. Absence of data doesn't mean you don't have to treat a problem. I wish I knew the best way to treat hypotension (low blood pressure) in tiny premies - for that matter, I wish I knew when to treat hypotension in them - but the data isn't clear. That doesn't mean I can ignore the problem. In another example, I still have to feed babies, even though it's not well resolved how fast to increase a baby's feedings each day.
Although there is more to do regarding evidence based medicine in neonatology, there has been tremendous progress in obtaining good evidence through well done studies in neonatology. A very partial list of examples includes artificial surfactant, the use of IVIG to prevent infections, the use of oscillators versus conventional ventilators, Vitamin A for prevention of BPD, and phenobarbital for prevention of intraventricular hemorrage. All have been studied well, (some worked and some did not) resulting in better care for neonates. The establishment of networks like those of the NICHD neonatal research network and the Vermont Oxford Network have furthered the cause of evidence based neonatal practice and will continue to do so, as will the Cochrane collaboration.
We should all realize, too, that it is very difficult to do good studies in humans. Earlier in my career I did a couple of randomized controlled therapeutic trials. They were nothing earth shaking and very safe for the participants, but they were still tremendously difficult to carry out, requiring (appropriately) approval from the human investigation committee, informed consent from the parents, diligent supervision of the NICU staff to make sure they followed the protocol, and, last but not least, money. This was all in addition to my regular job taking care of babies. To say the least, it was a real pain to do.
Speaking of money, we could use all the help we can get in obtaining money for research. The NIH's budget is getting lower, not higher. Are you writing your congressmen to protest this? Are you voting for people who want to put a reasonable portion of your tax dollar into research? We're not perfect in neonatology, but unless you're part of the solution, you might be part of the problem.