3 ms·

I'm skeptical that it is sound to just throw out smokers and make a claim about coffee health benefits. By taking people that use a lot of one common addictive

by yojo 11y ago

I'm skeptical that it is sound to just throw out smokers and make a claim about coffee health benefits. By taking people that use a lot of one common addictive substance (caffeine) and none of another (nicotine) they might just be selecting people with low potential for addiction/high will power. I believe those kinds of people might have lower mortality rates for reasons other than coffee consumption.

yaniksilver 11y ago
Just want to tell my experience here. Last Tuesday, I ordered a Cappuccino (strong). And I lit up a cigarette with it. Just after few minutes coming back to my work desk, I felt extreme headache and I fell from my seat unconscious. I was taken to a hospital near by and the doctor said your blood pressure is way higher than normal. (160/100) precisely. I believe that happened because of excess nicotine because I had no history of such blood pressure problem. I might be wrong but that was the case.
keithpeter 11y ago
Yes, I think that coffee-consumption could simply be a label that brings in other lifestyle factors. However, even among the non-smokers, the meta-study did find a difference between no-coffee and moderate-coffee groups. That may be pointing towards some of those 1000 compounds for further study...
polyfractal 11y ago
It's very common to remove smokers from studies like this (anything looking at mortality). Smoking increases your chance of so many health issues: heart disease, lung cancer, COPD, high blood pressure, stroke, aortic aneurysm, general infection due to chronic inflammation, etc etc. It's well documented and pretty unambiguous. If you smoke, your risk of all the above increases dramatically. Which means as a researcher you can't leave these people in a study looking at mortality rates since they would confound the results too much. Basically, the risk of confounding your results with smoking-related mortality is a bigger problem than biasing your population with fewer people who consume addictive substances. Source: I'm an ex-molecular biologist, my significant other is a PA.
- paulddraper 11y ago
  That's not what confounding means, from a stats perspective. https://en.m.wikipedia.org/wiki/Confounding https://en.m.wikipedia.org/wiki/Confounding Excluding smokers can cause a confounding variable, not the other way around. A better choice of word would be "obscure", as including them will make it harder to reach significance (though the end result should be the same, if the variable was not confounding).
  - polyfractal 11y ago
    Oops, sorry, you're right. It was very early this morning and pre-coffee when I wrote that, was paraphrasing colloquially instead of technically (and not thinking) :)

by yojo 11y ago

yaniksilver 11y ago

Just want to tell my experience here. Last Tuesday, I ordered a Cappuccino (strong). And I lit up a cigarette with it. Just after few minutes coming back to my work desk, I felt extreme headache and I fell from my seat unconscious. I was taken to a hospital near by and the doctor said your blood pressure is way higher than normal. (160/100) precisely. I believe that happened because of excess nicotine because I had no history of such blood pressure problem. I might be wrong but that was the case.

keithpeter 11y ago

Yes, I think that coffee-consumption could simply be a label that brings in other lifestyle factors. However, even among the non-smokers, the meta-study did find a difference between no-coffee and moderate-coffee groups. That may be pointing towards some of those 1000 compounds for further study...

polyfractal 11y ago

It's very common to remove smokers from studies like this (anything looking at mortality). Smoking increases your chance of so many health issues: heart disease, lung cancer, COPD, high blood pressure, stroke, aortic aneurysm, general infection due to chronic inflammation, etc etc. It's well documented and pretty unambiguous. If you smoke, your risk of all the above increases dramatically. Which means as a researcher you can't leave these people in a study looking at mortality rates since they would confound the results too much. Basically, the risk of confounding your results with smoking-related mortality is a bigger problem than biasing your population with fewer people who consume addictive substances. Source: I'm an ex-molecular biologist, my significant other is a PA.

paulddraper 11y ago
That's not what confounding means, from a stats perspective. https://en.m.wikipedia.org/wiki/Confounding https://en.m.wikipedia.org/wiki/Confounding Excluding smokers can cause a confounding variable, not the other way around. A better choice of word would be "obscure", as including them will make it harder to reach significance (though the end result should be the same, if the variable was not confounding).
- polyfractal 11y ago
  Oops, sorry, you're right. It was very early this morning and pre-coffee when I wrote that, was paraphrasing colloquially instead of technically (and not thinking) :)