Protocols are important

Protocols are important. In case of the lab world, they are everything. Therefore when Harold Jacobs focuses attention of how well protocols are followed, he is legitimately defending Floyd Landis. Here is how this defense is more than weaseling out on a technicality.

Lab tests like those in the doping scenario are not a case where you look in a microscope and see the synthetic testosterone. Finding evidence of drug use such as EPO and testosterone involve calculations and interpretations. Protocols are there to not only provide chain of custody, but to provide objectivity and remove bias.

Bias is in everything we do. It isn’t necessarily bad, it just is the way we are programmed. Let me give you three studies that show how our brains focus on something and begin to filter out new data (bias).

First. Focusing too much. There is a very famous study from Harvard that asks students to watch people in video bouncing a ball. The instructor tells the students that they must count the times the ball bounces and the times each person in the group touches the ball. Accuracy is stressed heavily. As the video begins and the ball bounces, a man in a gorilla suit walks across the screen, beats his chest and leaves the frame. Most students never saw the gorilla. They were focused on their task to the exclusion of new data.

Second. Increasing focus during stress. In similar experiments, students were asked to interview individuals and given questions on a clipboard. Accuracy of the questions and recording answers were stressed. Distractions were introduced, including large obstructions between the interviewer and interviewee. This caused the student to focus on the task more. During one of the obstructions that passed between the two, the interviewee was changed. Most students never noticed.

Third and finally. How accurate are your results? Do you have Quality Control? One experiment among university math students asked them to add a list of multidigit numbers. Speed vs. the other students as well as accuracy were stressed. At the bottom of the page was the answer. What the students didn’t know was the answer was wrong. If they made a common error in their calculations, they would obtain the result given as the answer. When the students finished, two things happened. 1) Those who added the list correct and got a different answer than what was on the bottom of the page checked and re-checked their math. 2)Those that finished with the answer at the bottom never checked their math. When you reach a conclusion you anticipated, you generally don’t put the rigor in checking a right answer as opposed to finding where you went wrong.

All this is known and therefore the genesis of protocols. Bias and subjectivity are removed so accuracy and repeatability are all that remains. Therefore a well run lab looks as smooth and disciplined as changing of the guards at Buckingham Palace.


My only question is why does the LNDD always look like the Keystone Cops instead?



2 Responses to Protocols are important

  1. […] I’m sorry to say that positive drug tests do not happen as they do on CSI. A lab technician does not put a few drops of clear liquid into a test tube which promptly turns blue indicating guilt. It is a very complex process prone to biases, interpretations and errors. […]

  2. […] This may be strong enough to win the legal test. I do think he needs to be stronger on the public opinion front where he emphasises his innocence as opposed to non guilt. While I have posted on bias and protocols, I think there is a risk of being cleared in the legal sense and not in the public opinion sense. People don’t like to think someone gamed the system. […]

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Google+ photo

You are commenting using your Google+ account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )


Connecting to %s

%d bloggers like this: