5 ms·

I'm not too keen on the discouragement against self experimentation at the end of the article in this case. Self experimentation has greatly improved insulin pu

I'm not too keen on the discouragement against self experimentation at the end of the article in this case. Self experimentation has greatly improved insulin pump technology in the past few years, and this situation has much much less risk.

chapium 8y ago
How good are most experiments with a sample size of 1? Can the data gathered be independently verified?
- hanniabu 8y ago
  It can be verified by others self experimenting with a sample size of 1, creating an aggregated larger data set. A type of decentralized trial, if you will.
  - dpatrick86 8y ago
    Some research is adopting this pattern too. A great example is Dr. Satchin Panda's clinical trial on time-restricted eating. You self-select an eating window and then document it through the app. Pretty cool stuff. mycircadianclock.org
- TheAceOfHearts 8y ago
  When you are suffering from some chronic disease you really only care about finding a solution that works for you. If there was a place where one could submit anecdotal experiences then you could analyze that data and use it as a basis for professionally run medical experiments. For example, if 1000 people said X food triggered symptoms of severe depression for a few hours after consumption then a professional could look at that and say "this is something worth looking into". You would be correct in remaining skeptical of individual claims which haven't been carefully verified, but you would be a fool to just ignore and discard those claims. Mikhaila Peterson is one example of an individual that was suffering from serious chronic problems who found a solution that appears to work for her. She's now on a carnivore diet, which means she only consumes meat and water. Now, that doesn't mean that everyone should go out and start imitating her, but it sure as hell sounds like something that medical professionals should look into.
- hyperpape 8y ago
  I'm pretty down on a lot of self-experimentation, but it's worth noting that probiotics are frequently recommended by doctors for many other issues and are unlikely to have harmful effects. I'm not familiar with the strain discussed here, but normal strains are ~$1/day. So the danger is that people might get a bit of false-hope or imagine a connection that's not there. I won't say that's harmless, but it can be a lot worse. I'm more worried that people will latch on to the gut connection and experiment with other "cures". There's an entire world of people poisoning their kids to try and cure "gut problems".
- thatthatis 8y ago
  Let me phrase it backwards: how good is aggregate data that throws away outliers in a complex system that isn’t yet fully understood?
rootusrootus 8y ago
I work with a type 1 diabetic who programs his own insulin pump now. Still goes to the doctor, but says they just don't have the same commitment to finding a good solution that he does, and he feels much better now. Sample size of 1, obviously, but he says it's pretty common to find other people in his situation who have similar results.

chapium 8y ago

How good are most experiments with a sample size of 1? Can the data gathered be independently verified?

hanniabu 8y ago
It can be verified by others self experimenting with a sample size of 1, creating an aggregated larger data set. A type of decentralized trial, if you will.
- dpatrick86 8y ago
  Some research is adopting this pattern too. A great example is Dr. Satchin Panda's clinical trial on time-restricted eating. You self-select an eating window and then document it through the app. Pretty cool stuff. mycircadianclock.org
TheAceOfHearts 8y ago
When you are suffering from some chronic disease you really only care about finding a solution that works for you. If there was a place where one could submit anecdotal experiences then you could analyze that data and use it as a basis for professionally run medical experiments. For example, if 1000 people said X food triggered symptoms of severe depression for a few hours after consumption then a professional could look at that and say "this is something worth looking into". You would be correct in remaining skeptical of individual claims which haven't been carefully verified, but you would be a fool to just ignore and discard those claims. Mikhaila Peterson is one example of an individual that was suffering from serious chronic problems who found a solution that appears to work for her. She's now on a carnivore diet, which means she only consumes meat and water. Now, that doesn't mean that everyone should go out and start imitating her, but it sure as hell sounds like something that medical professionals should look into.
hyperpape 8y ago
I'm pretty down on a lot of self-experimentation, but it's worth noting that probiotics are frequently recommended by doctors for many other issues and are unlikely to have harmful effects. I'm not familiar with the strain discussed here, but normal strains are ~$1/day. So the danger is that people might get a bit of false-hope or imagine a connection that's not there. I won't say that's harmless, but it can be a lot worse. I'm more worried that people will latch on to the gut connection and experiment with other "cures". There's an entire world of people poisoning their kids to try and cure "gut problems".
thatthatis 8y ago
Let me phrase it backwards: how good is aggregate data that throws away outliers in a complex system that isn’t yet fully understood?

rootusrootus 8y ago

I work with a type 1 diabetic who programs his own insulin pump now. Still goes to the doctor, but says they just don't have the same commitment to finding a good solution that he does, and he feels much better now. Sample size of 1, obviously, but he says it's pretty common to find other people in his situation who have similar results.