October 2006 • Vol. 2, No. 4
	Advancing Orthotic and Prosthetic Care Through Knowledge

Much of What I Know About Research,
I Learned in the Second Grade

There was this kid, Chris (his name really was Chris), who lived in the house behind mine, and he was in the second grade too. I used to go to his house every day after school to play. One day he turned around and hit me in the arm. My not so cavalier response was, "I'm gonna tell," and Chris's response..."You'll have to prove it!"

So there it started.

In my quest to prove that Chris hit me in the arm, I realized that I needed EVIDENCE. Well, certainly I knew he had hit me, because he had hit me before, and I knew that when he hit me it felt the same as it did that day. But that was merely ANECDOTAL, just because it was true to me didn't make it true to whomever I wanted to convince. I needed to verify that he had hit me. What I needed was an OUTCOME. Well my arm hurt, that certainly was an outcome…but maybe my arm hurt from falling down while we were playing basketball earlier that day or from when my little sister threw what used to be that nice glass piece of art that was on the living room table at me. Just having the outcome of pain wasn't enough evidence. So I immediately looked for further evidence, a bruise perhaps. Well nothing yet, a bit of redness but again, nothing to tie his actions to the outcome.

What if I had Chris hit 50 of my friends and they all hurt in a similar way? Well that certainly would give my outcome a bit more POWER but there still seemed to be too many VARIABLES—what if he was tired and didn't hit as hard or he stood further away or the recipient of such a blow had a higher pain tolerance and claimed it didn't hurt at all. So while I may have a bunch of similar outcomes, the test really lacked RELIABILITY AND REPRODUCIBILITY.

What I needed was some CONTROL. I needed a bunch of people to play basketball all day, and then randomly choose one group that Chris would hit in the arm in the same place at a set distance, at a certain force, and the other would not get hit and then we could compare arm pain, redness, etc.—A RANDOMIZED CONTROLLED TRIAL...whew, not only would this take time, but finding enough people to volunteer to get hit in the arm would be tough, not to mention Chris probably wouldn't do it because the outcome for him would be punishment. Not to mention what an Institutional Review Board (IRB) would have to say. This research stuff isn't so easy.

The other thing that comes to mind is that the LEVEL of EVIDENCE that I might need to prove my point really depends on whom I have to convince. If I happened to be at my house when Chris hit me and I happened to tell my mom and show her the slight redness on my arm…that may be enough evidence to get him in trouble; the desired result. But what if I have to convince HIS mom, or the doctor, or the police. The level of evidence necessary to achieve a desired result depends on whom I am trying to convince. So before I really laid out a full blinded controlled study, I should have some idea what type of study is possible and what level of evidence is necessary to achieve the desired result and get me paid for the prosthesis... (oops, I digress, back to Chris).

So then, if I could find a way to provide enough evidence to prove that not only did Chris hit me, but that there was a reproducible negative outcome from his actions (i.e., pain), that I could then convince someone else that maybe they shouldn't go to Chris' house (changing their practice) based on the evidence.

So there you have it: Evidence Based Practice. The goal is to be able to alter our practices through diligent study, recognizing the variables that effect the outcomes and convincing others that what we do is valid, meaningful, and provides the desired results…or by definition: conscientious, explicit, and judicious use of current best evidence in making decisions about the care of individual patients. Simple, right?

Recently, I bought Chris a ring with his initials on it, that way when he hits me, there would be plenty of evidence, even if it is just a single CASE STUDY and it may not be applicable to other times he or someone else hits me.

Thirty-eight years later, I'll get my proof.

What do you believe? Write and let me know at . Well-written beliefs may be published in the Academy Today.