Feb 10, 2011

Diet Soda and Heart Attack/Stroke Risk

So, one of my initial plans when starting this blog was to talk about nutrition topics in the news and various research that has come out. Instead, it has been mostly random rambling b/c that is who I am. In an effort to be more educational and helpful, today we are going to talk research. Specifically this new study that has come out linking diet soda and stroke risk.

Let me preface this with I LOVE DIET COKE. A lot. I don't even pretend. I know it isn't good for me. I don't care. I need it to live.

I've seen several articles and posts mentioning this study. I feel the media tends to take research and make sweeping statements as to why we're all going to die immediately. They like to jump on "study results" and go nuts. There is much, much less information about the study methods, results, and actual statistics. When I went to grad school I thought I would be learning *more* about nutrition, since that is my field. Nope! It is all research. How to analyze, perform, and interpret whatever research is happening in your area. Y'all, I hate hate doing research. I don't mind doing a nice lit review or meta-analysis, but I would rather be punched in the throat than have to design and actually do my own study. The moral of this story is: grad school wants to teach you how to be a good consumer and interpreter of data. It is your job as the expert in whatever field to look at the actual study/stats/methods and turn it into easily understandable, meaningful information for your target audience.

The number one rule of interpreting results is Correlation does not imply causation. It is very, very difficult to prove that one variable causes anything. 

So let's talk about this study, shall we? To be fair, I can't access the full actual paper yet since it is from a meeting and hasn't been published yet. Interestingly, there is exactly 1 result when "diet soda stroke" is searched for in PubMed (and it has to do with epilepsy.) So there obviously aren't tons of studies out there on that subject. Researchers at the University of Miami Miller School of Medicine found that in a group of 2500 participants, daily diet soda drinkers had a 61% greater risk for stroke/heart disease than drinkers of regular soda.

First the good stuff: 
-It is longitudinal. It took place over 9 years which yields much more information.
-It has a large-ish sample size-2500 participants.
-It identifies soda vs diet soda drinkers, not soda vs non-soda drinkers. (Also, it is very hard for my extremely southern self to call it soda. Everyone knows Coke=any soda.)
-Sample was diverse. 
-Researchers did control for age, history of stroke/heart disease. 

Now, here are some criticisms

-Yes, the study had 2500 participants. Of those, 900 participants claimed to drink no diet soda. Only 160 (8%) said they had diet soda everyday. Results found in 160 participants aren't necessarily applicable to the general population.

 -They gave a food intake questionnaire (and I've spent literally months in epi discussing the efficacy of each possible intake tool. not sure which one they used or it's validity.) at the beginning of the study and then never followed up. They just assumed intake remained the same over a decade.

-The study used self-report to identify intake. Self report is just a bad idea in nutrition research. People vastly overestimate the good and underestimate the bad stuff. Or they lie, b/c they think it is what you want to hear or they are embarrassed. 

--It is hard to tease out and quantify, but there is a segment of the population that eats terribly and drinks diet coke. Diet coke is associated (meaning it doesn't cause) obesity, in some part due to the fact that those who are overweight can be trying to cut calories by drinking diet. There is discussion on how insulin responds to artificial sweeteners and whether or not that can be linked to insulin resistance/ diabetes.


-They didn't look at sodium intake which is a known risk factor for vascular problems.  

-They did not look at weight/obesity over time. Which again, is a known risk factor for heart disease/CVA.

--But here is the key: THEY DID NOT CONTROL FOR OR ANALYZE THE DIETS OF THE PARTICIPANTS. So the participants might be drinking 1 Diet Coke with the 3 Big Macs, supersized fries, and apple pie daily. We don't know. -



Studies like this make headlines because the results are sensationalized. The media makes it sound like "If you are totally healthy, run 5 miles per day, are vegan, and a healthy weight...but drink 1 diet coke/day you are 60% more likely to have vascular complications." Which is just not possible to say based on these results. Again, this is the first study to link diet soda and vascular disease. If there is a link, it will be be able to be duplicated in studies that come after.

The biggest problem with nutrition research is that it is so difficult to isolate effects. Disease risk is multi-faceted, and sometimes we just don't understand it. Scientists are able to agree that it is better to get your nutrients from foods vs supplements, but they have been unable to determine exactly why. There are so many individual components of food: vitamins, minerals, phytochemicals, macronutrients, etc and they all work together in a unique way that isn't found artificially. Good health is a combination of so many genetic, environmental, and behavioral components that it is very hard to isolate the effect of each little piece.


I'm obviously not making a case that Diet Coke is good for you. It isn't. Go drink some water or green tea instead. What I'm trying to do is help people be informed consumers of research. Look at the methods, use common sense, and above all trust an RD ; ) Anyone who looks at the methods can see they didn't analyze or control for diet and realize that makes the results pretty weak. Just not everyone goes that next step. I would encourage everyone to really try to examine the study, not just the 'results' the media grabs on to. You might look at the methods and find them convincing. That is great! Just make sure it is an informed decision. Or just ask me and I will tell you what to do.

What do you think? Is this study significant enough to cause a change in your Diet Coke consumption? Am I completely wrong and blinded by my love for coke? Have any questions? 

7 comments:

Rachel said...

eek that's scary. particularly since i'm sitting next to a big and empty soda cup from subway right now hahah.

Carrie said...

Girl, you didn't read ; ) I think you will likely survive.

Megan said...

The primary thing that graduate school did was make me cynical about all research. (Well, ok, maybe not the PRIMARY thing, but it was a big thing). I especially never believe the media representation of research. This one seems to be especially problematic. Seriously, no controls for changes in diet or weight. That is laughable.

Rose said...

This was awesome to read because man, I love diet coke, too, and the study didn't seem that...sciency.

Part of my PT massage involved the guy pulling my arm back so my hand was resting on my butt. Awkward.

After the BJ conversation with my trainer he thanked me for being the kind of person he could have those types of conversations with.

Rose said...

I try to reply on the other person's blog so they get notified. Blogger's comment system is pretty lame.

I used to knock down 2-5 cans of diet coke a day. Now I save it for weekends or days when I'm not exercising after work. I probably have 3-5 cans a WEEK, which is a huuuuuge thing for me.

I used to pound that many cans of coke, but yeah, the calories. Now I love the flavor of diet coke, and can't drink regular coke.

I goatsed my trainer within a week of knowing him, so that's the kind of relationship we have.

Angela (the diet book junkie) said...

excellent post, thanks for sharing! personally, i don't drink soda, except for the odd vodka diet-coke here and there. interesting that they didn't analyze the diets of the participants. that's a very good point!

Carrie @ Confessions of a Dietitian said...

Thanks, Angela! I love diet coke even though I know I shouldn't. I can't help it.

Yeah, I kind of can't figure out why they would announce these findings or follow participants for 9-10 years and not analyze their intake. It really makes no sense to me.