The Skeptics Society & Skeptic magazine

Diet Sodas:
Are the Dangers in the Chemicals or the Headlines?

In April 2017, there was a flurry of news reports with alarming headlines:

  • “Diet Sodas May Raise Risk of Dementia and Stroke, Study Finds”
  • “A Daily Diet Soda Habit May Be Linked to Dementia–Alzheimer’s”
  • “Is Diet Soda Harming Your Brain Health?”
  • “Diet Sodas Tied to Dementia and Stroke”
  • “Here’s Another Reason You Might Want to Quit Diet Soda”
  • “Drinking Too Much Soda May Be Linked to Alzheimer’s”
  • “Is Soda Bad For Your Brain? (And is Diet Soda Worse?)”
  • “Diet Soft Drinks Triple the Risk of Dementia”
  • “Two Things Diet Soda Does to Your Brain”

Some of these headlines were from respectable sources like The New York Times and The Washington Post. It’s not surprising that many people were alarmed by the news and assumed that diet sodas had been proven to cause dementia. Some people were frightened enough to stop drinking diet sodas.

Headlines are designed to get people’s attention so they will read the article. They are notoriously unreliable. The first thing is to check whether the information in the body of the articles matches the headlines; sometimes it doesn’t. Even when the information matches, the article may selectively report some but not all of the studies’ findings, and it may put an unwarranted spin on the meaning of the findings. These headlines all referred to a single study; I wanted to know what that study actually reported, so I read it.

What Did the Study Show?

The study, by Matthew Pase and his colleagues, was published in a reputable journal, Stroke. The subjects were 2,888 individuals in the community-based Framingham Heart Study Offspring cohort. During a 10-year period of observation, 97 subjects over the age of 45 had a stroke: 82 ischemic (restriction of blood flow) as opposed to hemorrhagic (ruptured vessel and bleeding), and 81 subjects over the age of 60 developed dementia (63 consistent with Alzheimer’s). They estimated cumulative consumption of artificially-sweetened soft drinks based on self-reports on a food-frequency questionnaire. They found that drinking one or more artificially-sweetened soft drinks a day was associated with a nearly 3-fold increase in the incidence of ischemic stroke, all-cause dementia, and Alzheimer’s disease compared to drinking none. Sugar-sweetened soft drinks were not associated with stroke or dementia.

The study authors themselves clearly identified the limitations of their study:

  • Absence of ethnic minorities (so the results might not apply to them).
  • The study only showed a correlation; causation cannot be inferred.
  • Self-reported dietary intake is unreliable, subject to recall bias.
  • There could be confounding factors.
  • There was no adjustment for multiple comparisons, so some findings could be due to chance.
  • They said, “future research is needed to replicate our findings and to investigate the mechanisms underlying the reported associations.”

The lead author, Pase, is not yet recommending against diet beverages based on this study. In an interview with Medscape he brought up the possibility of reverse causation: “It is not clear whether the diet sodas are causing stroke and dementia or whether unhealthy people gravitate more towards these drinks than healthier people.”

Most of the media stories failed to report those caveats. Reporters may not have read the study or may not be qualified to evaluate it; often, a reporter simply regurgitates the contents of a press release about the paper and the study.

Applying the SkepDoc’s Rule

The SkepDoc’s Rule applies here: before you accept a claim, try to find out who disagrees with it and why. A little googling easily identifies several people who disagree. Sy Mukherjee, writing in Fortune, said “Stop Freaking Out About That Study.” He went on to say that an accurate description would have read “Study determines minor observational link (but no direct cause-and-effect) between certain people who drink artificial sugar beverages, but it has a small sample size that doesn’t include minorities or account for a whole bunch of other critical factors.” But of course, that’s too long for a headline, and it isn’t sexy.

Physician Aaron Carroll, in The Incidental Economist, wrote “They did not prove that diet soda causes Alzheimer’s Disease. THEY DID NOT!” He pointed out several problems with the study:

  • No ethnic minorities
  • Their conclusion was based on results from one model (model 2) that adjusted for demographics, diet, physical activity, and smoking. They also applied a model 3 that adjusted for more possible confounders, but those results weren’t as dramatic, so they downplayed that model.
  • Different artificial sweeteners are different molecules with likely different effects.
  • Multiple comparisons, so some results might be due to chance.
  • Self-reported intake is unreliable due to recall bias.
  • It was an observational study that could only establish correlation, not causation.
  • They emphasized relative risk (3×) rather than absolute risk (actual numbers of patients with stroke or dementia were small).
  • Even if there is an association, there’s no evidence that changing your behavior (drinking fewer diet sodas) will reduce your risk of stroke or dementia.

A story in ScienceDaily further pointed out that the study did not differentiate between different types of artificial sweeteners and only reported intake of diet sodas; it did not account for other possible sources of artificial sweeteners (in coffee and other foods).

The new findings differed from the findings of previous studies. The Nurses’ Health Study reported only a slightly increased risk of stroke (relative risk 1.16) associated with drinking one or more diet sodas a day; the risk was the same for sugar-sweetened soda. The Northern Manhattan Study reported an increased risk of all vascular events, including stroke, with daily diet soda consumption, but the relative risk was only 1.43, and there was no association with regular soft drinks or light diet soda consumption. There were no previous studies on dementia.

Do Artificial Sweeteners Make People Gain Weight?

Artificial sweeteners are intended to replace sugar. There is a correlation between sugar consumption and various health problems. Gary Taubes goes overboard, arguing that sugar is the cause of obesity and most chronic diseases. It’s not that bad, but everyone agrees that limiting sugar in the diet is advisable. Sugar provides calories but has no other nutritive value; we refer to it as “empty calories.” So it seems only logical that replacing sugar with artificial sweeteners would reduce total calorie intake, help people lose weight, and have other health benefits because of the reduced sugar intake.

Paradoxically, recent studies have suggested that intake of non-nutritive sweeteners may be associated with increased weight. Again, this is correlation, not causation, and the media reports have been very misleading. There are three possibilities with a correlation between A and B: A causes B, B causes A, or some other factor causes both A and B. Being overweight may lead people to use artificial sweeteners, rather than artificial sweeteners causing people to be overweight. Weight loss is complicated, and individual success depends on a lot of factors. And there is a psychological backfire effect: when you have virtuously limited yourself to diet soda, you may rationalize that you can afford to reward yourself by eating dessert.

Steven Novella covered this subject on Science-Based Medicine. He says the research shows that:

  1. Replacing high-calorie sugary drinks with low calorie drinks offers a modest health benefit.
  2. There is no “backfire” effect where artificial sweeteners somehow trick the brain into eating more.
What About All Those Alarming Reports of Side Effects?

The Internet is full of dire warnings about several artificial sweeteners. Saccharin causes bladder cancer in rats! The FDA wanted to ban it; but subsequent research exonerated it. It causes cancer only in male rats, and it does so by a mechanism not found in humans. It is safe for human consumption.

Aspartame has been demonized in a concerted campaign by scaremongers, epitomized by the book Sweet Poison. It has been accused of causing headaches, seizures, Alzheimer’s, cancer, diabetes, multiple sclerosis, birth defects, tinnitus, memory loss, and all kinds of other problems. It doesn’t cause any of those things. Hundreds of studies have been published that examined and dismissed those claims. Aspartame has been evaluated far more extensively than any other food additive. It is safe for everyone except the few people with the genetic disorder phenylketonuria.

No, You Don’t Need to Stop Drinking Diet Sodas

The recent study only showed a modest correlation with dementia and stroke. Even the lead author of the study has said it does not constitute a reason to avoid diet sodas. Its findings have not been confirmed by other studies, and the findings only showed correlation, not causation. More importantly:

  1. There is no evidence that if you stop drinking diet sodas, your risk of stroke or dementia will decrease.
  2. There is good reason to believe that switching from diet sodas to sugar-sweetened drinks will result in worse health outcomes.

There are a number of artificial sweeteners on the market. There is no credible evidence that any of them are harmful. There is good evidence that high sugar intake is harmful.

The Dangers of Reading Headlines

The moral of the story: diet sodas are not dangerous, but reading headlines can be dangerous. Media reports of scientific studies, not just headlines, often give the wrong impression. They put a slant on results to make them more newsworthy. They tend to make preliminary research sound like definitive proof. Early studies are often superseded by later studies with the opposite findings. We should never trust a single study; we must look at the total weight of all published findings. Most published research findings turn out to be false. That might sound discouraging, but it shouldn’t be. Science is a self-correcting endeavor. Don’t believe the headlines, but stay tuned for further developments. END

About the Author

Dr. Harriet Hall, MD, the SkepDoc, is a retired family physician and Air Force Colonel living in Puyallup, WA. She writes about alternative medicine, pseudoscience, quackery, and critical thinking. She is a contributing editor to both Skeptic and Skeptical Inquirer, an advisor to the Quackwatch website, and an editor of, where she writes an article every Tuesday. She is author of Women Aren’t Supposed to Fly: The Memoirs of a Female Flight Surgeon. Her website is

This article was published on April 18, 2018.


25 responses to “Diet Sodas:
Are the Dangers in the Chemicals or the Headlines?

  1. LC says:

    I have to say I love the fact people are even skeptical of this site. I’m a smoker & drink diet soda like it’s going out of style. Soda is probability bad for you diet or otherwise. Just like “light” cigarettes are no better for you than regular. One thing I find missing from a lot of these studies is genetic makeup.

  2. Mike T says:

    A point worth noting – many people have commented with things like “I’m 75 and I drink gallons of soda and I’m healthy so obviously soda doesn’t cause any harm….”
    This is a serious fallacy – like saying “my grandfather smoked 60 cigarettes a day and lived to be 105 so cigarettes are harmless.”
    Imagine for a minute that soda was really bad and harmed 50% of drinkers in some way – that would be a really big issue – but it would still mean that 50% of people drinking it were absolutely fine. So the person in my first quote could be one of the 50% who are not harmed. Just because some people are unaffected is not evidence that something is harmless. There are so many other factors involved.
    Most things are not “all or nothing.” e.g smoking increases your RISK of say getting cancer – but because so many other factors are involved, not everybody who smokes will get cancer – it just multiplies your risk by a lot. But the fact that some smokers will avoid cancer completely is not “evidence” that smoking doesn’t cause cancer.
    If we use examples like in my quotes, we are just using anecdotes – a trial where n=1. That is nowhere near being a sufficient number to make any kind of prediction about anything.
    Just look at the number of confounding factors mentioned in Harriet’s quotes – in order to rule out the possibility of these influencing the result you would need a significantly large number of people in the study (and of course you would have to use a different design in order to settle questions like association vs causation.

  3. Dr. Sidethink says:

    What all this seems to say is that aspertame tastes pretty much like sugar and is not as likely be unhealthy as previously thought.

    ut sugar tastes exactly like sugar and the weight gain if any is not as likely be unhealthy as previously thought.

    sounds like jive to me,

    what happened to moderate eating and exercise

  4. Cynthia says:

    If you see a heavy person buying diet soda, don’t assume that the diet soda made them heavy. That is just ridiculous. Consider what else is in their shopping baskets. Consider that they may have just switched to diet. Consider that they may have already lost a large amount of weight. Consider that they may be diabetic and unable to consume sugar soda. Consider that they may even be purchasing it for someone else.

    If you don’t know anyone who lost weight drinking diet soda, consider how much they might have weighed if they drank sugar soda and what the rest of their diet and lifestyle look like.

    There are an awful lot of comments here with false causality and/or correlation!

  5. skeptonomist says:

    Studies based on any kind of survey can be unreliable for several reasons. For example you don’t know why people are doing what they do – people seem more likely to drink diet soda if they are already overweight.

    The best results are experimental – people are split into groups and assigned certain medicines, foods, or whatever and the result ideally evaluated by people who don’t know what the subjects are taking. This is obviously more difficult and expensive than surveys. If you see results from a survey which conflict with those from experiment, prefer the experimental results. Many reporters don’t seem to be aware of the difference or don’t care.

  6. Jimmy D Watson says:

    Every study read today will be debunked tomorrow,it’s the deal of the day…. still interested in learning the truth . thanks

  7. Roger Linse says:

    There is a front-fire effect as far as I am concerned. I drink
    none sugar drinks to off-set my huge intake of ice cream.
    My health right now at 65 years old is excellent. My advice is to do a little scholarship on any subject so you have multiple sources. And don’t vote for buffoons when
    voting for President. I understand why some of you did it, just don’t make a habit out of it.

  8. Chuck Buchanan says:

    I’ve never met anyone who lost weight by drinking diet soft drinks.

  9. Barbara Harwood says:

    Cause and effect can be a little hard to establish. You could say that divorce is caused by marriage. Quite clearly, 100% of all people who get divorced are married to begin with.

  10. George says:

    That’s because you voted for the Buffoon, didn’t you.

  11. tpaine says:

    Harriet says in the preamble…”it may put an unwarranted spin”
    Like this spin…”Some of these headlines were from respectable sources like The New York Times and The Washington Post.”

    In my household, these are not respectable sources.
    Just saying.

  12. Bruce says:

    ACW, you win the comments award for best comment according to me.
    Yeah, that’s not really a thing.

  13. ACW says:

    If Diet Coke could kill you, or cause dementia, I’d either have been dead or drooling in a nursing home decades ago. I drink it in stupefying quantities. It isn’t ‘natural’, and it has no nutritional value. It just happens that I like it. Moreover, I have to confess I grew up on regular Coke and Pepsi, so I’m in the habit. I was also an obese teenager, and in addition to adopting a rational, three-reasonable-meals-a-day diet and a regimen of moderate exercise, replacing the sugary soda with diet enabled me to take off, and keep off for more than 40 years, that 75+ lbs of avoirdupois.
    The major cause of dementia is, gasp, getting old. It’s been known since Nestor bored the Greek armies with his diatribes and Pentheus mocked Tiresias and Cadmus for decking their aged bones in deerskin to go dance with the maenads. Shakespeare described dementia as the seventh age of man in Jaques’ famous speech in ‘As You Like It’. If it’s ‘epidemic’ now, it’s because we Boomers, the demographc ‘pig in the python’, are reaching that stage.
    Not every molecule one consumes need be ‘virtuous’ — as long as you eat a reasonable, balanced diet consisting mostly of wholesome, nutritious, real food — mine is ethical vegan, therefore, fruits, vegetables (fresh is nice, but frozen is ok if it’s just the plain vegetable, not adding ‘sauce’ etc), legumes (if you don’t want to cook from scratch, some canned brands, such as Goya, don’t add sugar), grains (which can include pasta, rice, etc. and don’t have to be whole grain if, like me, you don’t like them), with the occasional frozen-dessert treat in small helpings, cooking at home and avoiding restaurant, fast-food, or processed.
    My experience, from observation, is that people who remain obese despite drinking diet soda or adopting a diet underestimate either the size of their portions or how often they eat, or think a Diet Coke at Outback suffices to offset a surf-‘n-turf entrée with extra melted butter on the lobster, a double side order of fries, an entire Bloomin’ Onion, plus a helping of death-by-chocolate cake.
    Gotta go; I’m thirsty. :D

  14. Richard Raymond says:

    I’ve never drunk “pee”, so I guess I wouldn’t know. It seems other people have, though. They’d know better than I if diet sodas taste like pee. But, just in case, I’ll continue to not drink diet or regular sodas even though I love the carbonation. I am getting to like the taste of water. I heard that most of the bottled water has plastic particles in it. Heck, can’t win!

  15. Bad Boy Scientist says:

    Thank you again, Dr Hall. Your articles are well researched and well written.

    My opinion of science journalism was shaped by an experience in grad school. A newspaper clipping was posted outside of a professor’s office with a highlighted section that had his name and quote marks around a statement that was exactly wrong. As if the writer omitted the word ‘not’ and made the professor look ignorant of something that all grad students and most undergrads knew. The rest of the article indicated the writer thought the erroneous statement was what the professor had actually said.

    Under the clipping was a sign that read “This is why you never speak with reporters.”

  16. js says:

    Diet soda tastes like pee.

  17. PAUL VAN GULICK says:

    For me, the biggest takeaway from this is the warning against skimming headlines. They really are notoriously misleading. My favorite is this one from my local paper some years ago: “Nuclear Device Goes Missing”. You can bet I read that one. It turns out that a nuclear densometer, a common device in construction, was stolen from the back of a construction truck.

  18. Mike Colyar says:

    I live in a third world area where the water is bad. I buy my diet soda in cans which can be and are recycled. Almost every drop of moisture I drink comes from diet sodas. No sign whatsoever of memory problems and I am 75.

    Go away.

  19. Mike Colyar says:

    I live in a third world area where the water is bad. I buy my diet soda in cans which can be and are recycled. Almost every drop of moisture I drink comes from diet sodas. So sign whatsoever of memory problems and I am 75.

    Go away.

  20. Cliff says:

    I didn’t read the study, but people with memory problems (among others) self reported how much and how often they drank diet soda?

  21. Jerry Rampelt says:

    I agree with the holes that are poked into the study, but let’s look at it from the opposite side. The study did not indicate a decrease in the risk of dementia. That headline would read, “Drinking diet soda reduces the risk of dementia.” That would be a great advertising campaign for the diet soda industry.

    My view is why take the added risk of dementia by drinking diet soda. That is especially true for me because my father and his two sisters suffered from dementia as well as a first cousin.

  22. steve says:

    I like how theres a Amazon advertisement for diet soda at the bottom of this article… just saying…

  23. Michel Belley says:

    You report here that “The lead author, Pase, is not yet recommending against diet beverages based on this study. In an interview with Medscape he brought up the possibility of reverse causation: “It is not clear whether the diet sodas are causing stroke and dementia or whether unhealthy people gravitate more towards these drinks than healthier people.””
    What were the conclusions of his article in the journal Stroke? Exactly what he reported to Medscape. Further studies are needed because what they found was an association between two factors, and possibly not a cause/effect relationship.
    The problem arose because people read only the abstract and translated the word “association” by causation, instead of correlation.

    Extracted from his article in Stroke:
    Previous studies linking artificially sweetened beverage consumption to negative health consequences have been questioned based on concerns regarding residual confounding and reverse causality, whereby sicker individuals consume diet beverages as a means of negating a further deterioration in health.
    Clinical trials are needed to establish whether the
    consumption of artificially sweetened beverages is causally related to dementia or surrogate end points, such as cognitive decline or brain atrophy.

  24. Dr. Patrick Buick says:

    Thank you, I’ve been on a diet for 2 months and lost a few pounds but have been miserable. I miss my Fresca with 1 or 2 oz of OJ.

    And a piece of toast with my one egg breakfast made my day.

    I am old and healthy, so #^&! it.

    I am also an old (77+) retired science professor who has taught skepticism to all of my science students for years, and subscribed to The Skeptic.

  25. Pete says:

    Even if soda were not a health hazard (disputable), there are other good reasons to not consume it, like saving money and the environment. Further, those who consume soda show worrying signs of addiction to it. I choose for my life to be addiction-free.
    Water is the best thing for the human body to drink. It is also virtually free if you don’t buy it in plastic bottles, which is completely foolish and damaging to the environment. If you want a little flavor, go to homemade herbal tea. I sweeten mine with a little honey from my own bees.
    While it is purely non-scientific, observe those at the store who are loading up on cases of soda and tell me if they look like they would benefit more than the average person from saving money and reducing caloric intake. Almost all of them do.

This site uses Akismet to reduce spam. Learn how Akismet processes your comment data. Comments are closed 45 days after an article is published.

Skeptic Magazine App on iPhone


Whether at home or on the go, the SKEPTIC App is the easiest way to read your favorite articles. Within the app, users can purchase the current issue and back issues. Download the app today and get a 30-day free trial subscription.

Download the Skeptic Magazine App for iOS, available on the App Store
Download the Skeptic Magazine App for Android, available on Google Play
Download the Skeptic Magazine App for iOS, available on the App Store
Download the Skeptic Magazine App for Android, available on Google Play
SKEPTIC • 3938 State St., Suite 101, Santa Barbara, CA, 93105-3114 • 1-805-576-9396 • Copyright © 1992–2024. All rights reserved • Privacy Policy