Why Aspartame Isn't Scary

kace_kay wrote: »

Diet sodas may be tied to stroke, dementia risk.

http://www.cnn.com/2017/04/20/health/diet-sodas-stroke-dementia-study/

Don't confuse CNN editorials based on statistical analysis of survey's for scientific studies.

Here is the source for that CNN editorial.

http://stroke.ahajournals.org/content/early/2017/04/20/STROKEAHA.116.016027

From that paper here is what they actually did.

"We studied 2888 participants aged >45 years for incident stroke (mean age 62 [SD, 9] years; 45% men) and 1484 participants aged >60 years for incident dementia (mean age 69 [SD, 6] years; 46% men). Beverage intake was quantified using a food-frequency questionnaire at cohort examinations 5 (1991–1995), 6 (1995–1998), and 7 (1998–2001). We quantified recent consumption at examination 7 and cumulative consumption by averaging across examinations. Surveillance for incident events commenced at examination 7 and continued for 10 years. We observed 97 cases of incident stroke (82 ischemic) and 81 cases of incident dementia (63 consistent with Alzheimer’s disease)."


What that basically means is they looked at survey data for people who answered that they drank diet beverages and found that those people had a higher incidence of stroke and dementia relative to people who said on the survey that they did not drink diet beverages.. None of the variables were controlled as it was an after-the-fact evaluation of survey data to search for correlations. From that they concluded "Artificially sweetened soft drink consumption was associated with a higher risk of stroke and dementia."

Note the word "associated" which is another term for it being a correlative.

Of course someone who writes for CNN picked that up and ran with it with an editorial.

People who are overweight often drink diet beverages. That doesn't mean diet beverages make you fat. Correlation is not causation, a phrase I am sure you have heard given this is one of the most abused things out there in the media's interpretation of scientific data.

To see what you can do with correlations this is a website that pulls aggregate data and uses algorithms to find correlations and then auto-produces graphs to show those correlations. If you think correlation is equivalent to cause then here might be some other things to avoid:

http://www.tylervigen.com/spurious-correlations


Here is the thing with correlations. They are very useful in scientific study....for disproving that something is a cause of something else, but not for demonstrating cause. In otherwords if you hypothesize that A causes B you can do a correlation study and show that A correlates with B not as proof of cause but as part of due-dillegence of testing the null hypothesis. If you did the correlation study and showed A did not correlate with B that would demonstrate that A is NOT the cause of B. The reason for that is correlation is a requirement of cause but not all correlations are causes. If A causes B then A will correlate with B but many MANY things that aren't causes will also correlate with B.


The issue isn't the studies or the people doing the studies, the issue is the media picking up correlations like they are proof of causation and then writing the editorials that are the actual public exposure since the public by and large is not science-literate and does not read the actual studies.