Science and the Scleroderma Girl: Controlled Experiments

I literally learned about controlled experiments at my mother’s knee; there was a new baby in the house, and she was slowly introducing solid foods into his diet and I was helping to feed him. She carefully added only one new food a week: applesauce, carrots, squash… it was boring and I wanted to give him lots of interesting foods all at once. Nope. She insisted that it was important to keep everything exactly the same EXCEPT for the new food that was being introduced; if the baby got sick then we would know that the new food was the cause. Bam! This is an explanation of controlled experiments that even my five year old self could understand.

Upset cat
The Mother of Cats was raised by a very smart woman…

As a person with a chronic illness I hang out in internet forums and support groups that are awash with lots of confusing and conflicting information. It is a lot to sort through, and I find myself searching for validating information generated by that method I first learned from my mom: results from controlled scientific studies.

This is what I’m looking for when I deep dive into the internet looking for results on medications, alternative medicine treatments, supplements, exercise, possible causes of my illness, whatever.

  • Who did this research? Is it being presented by someone with a financial stake in the results? For example, research showing that cigarettes had no impact on cancer rates funded by the tobacco industry was later shown to not be valid. There was research bias in the tobacco-funded research. I look for bias like that as I collect information.
  • Did this research study have a control group? This means that if a group of patients received a drug, was there another matched group that did not get the drug; the two groups should be compared to see if the drug impacted their illness.
  • How well matched were the two groups? Ideally, they should be as identical as possible. The groups should have patients of the same ages, genders, diagnoses, etc.
  • What was the data that was being collected? I feel that hard data like lung volume, CAT scans, arterial pressure or blood tests are more meaningful than a survey filled out by patients that reports how “bad” they feel.
  • How large were the research groups? A study done with only a dozen patients is not necessarily invalid, but one done with a couple of hundred patients is one whose results I’m more likely to regard as meaningful.
  • The gold standard in research studies is one that assigns one group the drug, and the other group a placebo (an inactive dose) as the control. The patients and the doctors do no know which patients are receiving the drug, and which are getting placebo. This type of study, a double blind, is considered the most reliable as bias has been removed: differences between the two groups is due only to the drug.

This is a lot to keep in mind when looking at research reports, but after a while it gets easier. I used to hunt for newspaper reports of medical research to use with my classes. One of the favorites was this study about a virus that causes obesity…

Mice
Mice used in medical research studies are genetically identical. In the fat virus study animals were kept in identical conditions and one group was infected with the virus, and the other was not. The fat boy got the virus. (okay, this picture was swiped off the internet… I have no idea what actually happened to these two particular mice, but you get the idea…)

I don’t have the article from the Denver Post anymore, but here is the same study described on WebMD. The research, done by Richard Atkinson, M.D., showed that animals infected with the Ad36 virus have a huge weight gain even though they were eating the same amount as the control group. It happened in any animal group exposed to the Ad36 virus: mice, rats, chickens, monkeys. When he checked the blood of humans for antibodies to Ad36, he discovered that 30% of humans had antibodies to the virus, and those people were significantly heavier than people without the antibodies. Here’s the breakdown of the experiment:

  • Virus exposed mice: experimental group
  • Mice not exposed to the virus: control group
  • Conditions kept constant: the students could like several like age, type of mouse, genders, food, cage size, etc.
  • Data collected: weight gain

My students could also come up with the hypothesis and make predictions for further experiments. It was always interesting to me that many students would insist that obesity was just the result of poor diet and lack of exercise and that this data couldn’t be true. That’s bias, I would tell them. You have to follow the data; these results don’t mean that diet and exercise aren’t important, but you can’t just dismiss them out of hand, especially since it was validated using several animal models…

This is the type of thinking that I try to cling to as I hunt for information that might help me handle my chronic condition. Will essential oils help? Which ones? What is this whole AP protocol? Is sugar inflammatory? What about dairy? Is CellCept better than methotrexate in the long run? What is the best treatment for PAH? Should I go to yoga class? What studies have been done, and what does the data say? Learn, learn, learn.

My best survival skill is now the one that I learned at my mother’s knee.

Advertisements