- #51
SW VandeCarr
- 2,193
- 77
Interesting, but isn't this is only a demonstration of the "garbage in, garbage out" principle? There have been many well conducted epidemiological studies that have changed both medical practice and the general behavior of the US population for the better. "For the better" is the evidence for better outcomes in treatment based on decreasing rates of cardiovascular mortality and increased longevity.
http://circ.ahajournals.org/content/92/11/3350.full
Do you still think tobacco smoking doesn't increase your risk for lung cancer (among other nasty things)? Note "increase in risk" is the way to phrase it. "Cause" only applies to individuals, not populations. Unfortunately, populations of humans are not the same as populations of undifferentiated particles. So what causes cancer in one person may not necessarily cause cancer in another. This is seems obvious enough, but apparently is not the common wisdom here.
Regarding this coffee study (which involved over 400.000 subjects in a large NIH study) I have my own reservations. I gave an example of confounding in an earlier post. The authors excluded people with a history of stroke, but not hypertension (high blood pressure). This could confound the study if people with hypertension were less likely to drink coffee and were classified as non coffee drinkers.
On the other hand, a clear dose-response relationship seems to exist. The more cups per day (up to 3), the greater the putative protective effect. This is not commonly seen in studies of this type. It's more commonly shown in n=1 studies. This is fairly good evidence there is an effect in the population. However, no one can claim drinking more coffee will extend your life or mine.
http://circ.ahajournals.org/content/92/11/3350.full
Do you still think tobacco smoking doesn't increase your risk for lung cancer (among other nasty things)? Note "increase in risk" is the way to phrase it. "Cause" only applies to individuals, not populations. Unfortunately, populations of humans are not the same as populations of undifferentiated particles. So what causes cancer in one person may not necessarily cause cancer in another. This is seems obvious enough, but apparently is not the common wisdom here.
Regarding this coffee study (which involved over 400.000 subjects in a large NIH study) I have my own reservations. I gave an example of confounding in an earlier post. The authors excluded people with a history of stroke, but not hypertension (high blood pressure). This could confound the study if people with hypertension were less likely to drink coffee and were classified as non coffee drinkers.
On the other hand, a clear dose-response relationship seems to exist. The more cups per day (up to 3), the greater the putative protective effect. This is not commonly seen in studies of this type. It's more commonly shown in n=1 studies. This is fairly good evidence there is an effect in the population. However, no one can claim drinking more coffee will extend your life or mine.
Last edited:
