Science, observation, hypothesis, experiment, analysis

With the water quality in Rio for swimmers in the Summer Olympics such a hot topic this year, a number of people pointed out that in the 2007 Pan American Games some swimmers were later diagnosed with bowel diseases.

Link to observation about 2007 Pan American games:

However, other world class swimmers who were not at the 2007 Pan American games in Rio have also been diagnosed with bowel conditions (Ulceritive Colitis and Crohns) and despite the UC and Crohns conditions, went on to medal in Rio: Link for that here:

So that is two different observations. Some people would simply stop there and go “OMG! Swimming causes bowel problems!!” and rush to tell the world about their amazing new “discovery” which is nothing more than an untested hypothesis at this point.

The next step is to form a hypothesis, and a null hypothesis.

The hypothesis is, “swimming as a sport is associated with higher rates of bowel disease.”

The null hypothesis is, “swimming as a sport is not associated with higher rates of bowel disease.”

Then you have to design an experiment. This is where things get expensive, especially in human studies. In this case, I would try to do a “controlled pair” study. A well designed experiment will have an experimental group, and a control group. You must design the study to accurately measure a large enough population to not artificially inflate the numbers if someone in a small sample size contracts a disease, and you must have a good control group that is as similar to your experimental group as possible.

If you can’t do a study that large requiring large amounts of funding, you can always mine through the existing literature on the subject you want to study.

One microbe looked at:
Water treatment and risk of bowel diseases post water sport recreation:
Another microbe looked at:

So the current literature does have a solid link between swimming and bowel conditions. The literature is doesn’t have a solid link between the short term gastrointestinal distresses and full blown UC or Crohns, so there is room in the literature for more research on the subject. Which means a well designed experiment could be useful to other researchers later on.

To design this hypothetical “controlled pair study” I would try to recruit college swimmers and college track athletes for my controlled pair trial. I would try to recruit them from multiple colleges across as broad an area as I could get, and I would try to match them male for male, and female for female, as close as I could for ethnicity. I would try to get data points from them at least twice a year using self reporting questionnaires backed up by medical examinations and pool maintenance records. Ideally we are looking at an initial population of about a thousand each at the start, so that even with a 50% drop out rate we can have 500 or so in each group four years later.  See what I mean about human studies being expensive?

And once the data points were all gathered, we could crunch the numbers to do the analysis part and see if anything popped out with an interesting result. If the hypothesis was supported, great, go ahead and publish. If the null hypothesis is more likely, great, go ahead and publish (although researchers tend to not publish non-sexy results). If there is a cluster of diseases that pops up in one school, the pool maintenance records would be helpful, but maybe not illuminating.

And that is science, observing something and thinking, “huh, that’s interesting” and then trying to figure out if there is really anything there or just a random sequence of events that our pattern matching brains put together. Even more important, a well designed study of college athletes should not be taken as gospel in other populations. It should be understood that these sorts of studies don’t take into account the endless minor details of real life outside of the study.

So I hope this refresher in the scientific method has been fun, and why “science” isn’t something that tells the truth, but seeks the truth.

This entry was posted in Uncategorized. Bookmark the permalink.

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