President's Message - Is Your Practice Evidence-Based?
by Mike Fotis, ICHP President
June 4, 2014
I had the opportunity to attend an awards ceremony a few years back and found myself seated with a hospital administrator and several leaders from other healthcare fields during dinner. Usually when one is seated with five people you have never met, of course the other five end up being colleagues of some sort, known each other for years, and you get to be the odd man out during this happy reunion. As it turned out, the hospital administrator was a contemporary of mine, had graduated from pharmacy school at about the same time, and we ended up being a part of an interesting discussion. What a relief!
The administrator had not practiced pharmacy for decades and pointed out during our discussion that most likely he had forgotten almost everything taught in pharmacy school. He was a bit embarrassed by this. I remarked that perhaps having forgotten college material might not be as big a problem for him as he thought as most of the things we learned have been updated, and a lot of material we had learned is now flat out incorrect!
Of course I was being a wise guy during dinner conversation, but I think anyone who has been out of school for a dozen years or more has noticed that so much of the latest information tends to contradict earlier information we spent so much energy learning…which brings me to the question: Is your practice evidence-based?
How are you going to answer this question? My sense is that like me, most readers will state, “Well, of course MY practice is evidence-based. But I am not so sure about most of my colleagues!”
I think our goal as clinicians is to base our recommendations on findings that are reported in adequately powered randomized, double blinded studies using an active comparator that reported meaningful improvements in clinical outcomes in intact human beings. (Such studies are usually classified as Level 1 evidence.) We all agree with this statement of course, but sometimes the perfect study doesn’t exist (yet), and we need to make decisions about treatments that are based on weaker information. There is nothing wrong with using placebo comparisons or observational studies, or basing recommendations on expert opinion, anecdotes, or even on laboratory results…as long as a) it is the best information available; and b) the increased level of uncertainty is described and is transparent.
My point is that information from Level 1 evidence has a lesser degree of uncertainty and tends to hold true over time, and while we may need to modify our views a bit in the future, the basic structure of our understanding often does not change. On the other hand, information based on animal models, human cell cultures, laboratory analyses, observational studies and even expert opinions may often be corrected and substantially revised when later examined in a Level 1 type study.
Most speakers and authors go beyond merely listing a reference and point out when information was not measured in adequately powered randomized, double blinded studies using an active comparator that reported meaningful improvements in clinical outcomes in intact human beings. This step is an easy one: For example, “recent data from an animal model of kidney function has shed new light on the mechanism of action of drug X.” Or perhaps point out that an “antimicrobial dosing strategy is based on microbiological findings plus population pharmacokinetic data, but of course still needs to be adequately tested in humans.”
Most of us provide a necessary caveat to our audience most of the time. I would like to see us as pharmacists become more consistent at reporting the limitations of our recommendations. Whether the audience is our students, residents, or colleagues and whether the venue is KeePosted, the ICHP Annual Meeting, or the Spring or Regional Meetings, we should be careful to specifically point out whenever information is based on less than Level 1 evidence. By pointing out the limitations and inherent uncertainty of your analysis, you will let your audience know there is a good chance that someday in the near future your conclusion will need to be revised. And of course when instead recommendations are based on Level 1 evidence, it’s a good idea to plan on remembering this data into the future!
Please let me know your opinion on this topic. I plan to submit this recommendation to the Division of Educational Affairs for consideration.
The News Article appears in the following categories: