Evidence Based Medicine

[Bieber]

So, after reading another thread I got to thinking that it would be nice to have a place where we could all share our knowledge and form a compendium of evidence based medicine in relation to the treatments we in EMS and emergency medicine in general provide, have provided, or may provide in the future. I am aware that many if not most of the treatments we give are not supported by scientific evidence, but I realized after reading another thread on here that there are more treatments unsupported by studies than I thought.

The rules of the thread are thus: every point is disputable, however all claims must be backed and supported by references to peer-reviewed academic studies. And while the studies may tell us one thing, I also want to know what your personal and individual experiences with these treatments are; because we all know that what we read in a book and what happens in the streets are two very different things.

The goal is to find as much credible evidence for or against as many prehospital treatments as possible, and also for in-hospital treatments. Oftentimes, we look at these studies in relation to how we in EMS should be doing things, but what about the hospitals? Are they advising us against the same unfounded treatments they themselves continue to provide and are any of these treatments beneficial for ANYONE in any setting?

[NYCEMS9115]

Unfortunately, medicine is a practice. You can never get everyone to agree on the same thing. I believe something; others may dispute. Science based studys are only thing that changes the practice of medicine. Ideas are things that starts a study. Every practice must use education, continuing education, protocols, guidelines, experience, & discussions to make medical disicions. This is what we all must do; sometimes it can be the wrong dicision. Mostly, due to the misunderstanding of the science. You will meet resistance and bias because you have ideas. Be prepared. See many posts; how members get rediculed because they have different ideas. Go for it. Good luck.

[chbare]

One thing we need to realise early on is that the best we can hope for is an indeterministic assessment of the physical world. In that I mean, our most complicated tools are really only based on the toss of a coin or die. Therefore, it can be exceedingly difficult making sense of our world in meaningful and consistent ways. So, we have to understand the limitations of EBM and the techniques we use to make sense of evidence.

With that said, it is not difficult to appreciate the variance in opinions and the ways in which people interpret information. I will give you a very basic example:

Let's say I have a data set of 5 people. Let's say we are trying a new EMS treatment on trauma patients? The data set will be patient's who responded with an increase in their mean arterial pressure above some threshold. We can have say three thresholds. One is minimal response, two is moderate and three is good response. Patients who did not make threshold 1 will be identified as threshold 0.

patient one: threshold 3, patient 2: threshold 3, patient 3: threshold 1, patient 4: threshold 1, patient 5: threshold 0

This gives me, 3,3,1,1,0

Let's say I want to get an average of these findings. Well, I can choose three different techniques:

1) Mean: Add em up and divide: 8/5 = 1.6

2) Median: Order and choose the middle: 0,1,1,3,3 ---> = 1

3) Mode: Most common number, I can go with either one or three

So, in our simple problem above how do we interpret the data? The mean and median suggest it may not be all that effective (of course, I never defined effective), but the mode could go either way. What do we do? Look at more patients, get a confidence interval and Z score, look at the standard deviation, do a new study and define our responses differently and so on. What if we keep drawing blanks? Clearly, something is going on with our data, but not everybody will agree on what that something is or how it should be interpreted or implemented.

This is where the importance of consensus and multiple studies that are peer reviewed comes into play I think. Many guidelines such as the surviving sepsis guidelines are based on consensus where people basically look at data and then play a game of give and take compromise on recommendations that they can at least live with.

So, I absolutely agree with EBM and thinks it's the best we currently have available, but we must recognise that it can be difficult to make any definitive conclusions in many cases.

Take care,

chbare.

[Bieber]

Excellent point, Chbare! And one I forgot to mention earlier. You're one hundred percent right, data interpretation, including relevance, relationship, and variable factors, is a major drawback to studies done in the field of emergency medicine. It's often cited, but perhaps not nearly enough, just how hard it is to get reliable, understandable, straight forward information regarding the results of studies done on emergency medical patients. To use another example, though I'm afraid it's far more simplistic than yours but still one that I like to present, you could say that in a study of mortality rates that patients who are intubated by EMS personnel have a higher mortality rate than those who are not intubated by EMS. And without the relevant details, it sounds like intubation is a bad idea (and bear in mind this is just an example, not to start a debate on prehospital intubation), however all that study would actually say is that sicker people die more often. For studies on EMS practice to be truly sound and not just number crunching, we have to look at all of the relevant variables.

Great post, Chbare, and definitely a take home point with all of the studies presented in this thread: proper interpretation of the data.

[Bieber]

This is a great site where I've found a lot of interesting research on various practices in EMS. I haven't gone through and examined all of the studies just yet, but if someone wants to highlight one in particular feel free.

http://emergency.medicine.dal.ca/ehsprotocols/protocols/toc.cfm

[D.P.Gumby]

This is a great site where I've found a lot of interesting research on various practices in EMS. I haven't gone through and examined all of the studies just yet, but if someone wants to highlight one in particular feel free.

http://emergency.med...otocols/toc.cfm

If you haven't heard of it, the Cochrane library maintains a similar database for many medical topics, although it is not EMS focused. I am sort of jealous of that Canadian document, though. Maybe I should have moved up there, I hear the skiing is good too....

Creating a compendium like that is a huge undertaking, I suspect you'll have more success with smaller steps, going through individual topics. Why don't you start off the discussion - pick a topic that interests you and tell us what you think of the evidence. I'd be happy to play along with a more manageable activity like that, and maybe some of smart experienced people could give some more specific input with a start like that.

[Bieber]

If you haven't heard of it, the Cochrane library maintains a similar database for many medical topics, although it is not EMS focused. I am sort of jealous of that Canadian document, though. Maybe I should have moved up there, I hear the skiing is good too....

Creating a compendium like that is a huge undertaking, I suspect you'll have more success with smaller steps, going through individual topics. Why don't you start off the discussion - pick a topic that interests you and tell us what you think of the evidence. I'd be happy to play along with a more manageable activity like that, and maybe some of smart experienced people could give some more specific input with a start like that.

Hmm, that's not a bad idea. Now to find one that hasn't been overdone, such as ETI. Standby.