Last updated on November 27th, 2017 at 12:16 pm
Two recent studies published in respected journals seem to indicate that diet soda increases risk of stroke and dementia. Not to give a free pass to sugary drinks, one of the studies seemed to indicate that either artificially sweetened or sugar filled drinks might be linked to a higher risk of Alzheimer’s disease.
Typical of the popular press and your average website, they accept the findings of these studies without any serious critique of these studies. US News blares a headline that says, “Health buzz: drinking diet soda linked to stroke, dementia risk, study says.” Thankfully, many of the headlines use the qualifier “may be linked,” but I’m afraid most people will overlook that nuanced discussion of these studies.
But what does the actual science say? Do these studies provide us with robust evidence that cutting out diet soda will suddenly decrease our risk of stroke, dementia or Alzheimer’s disease? Probably not, but let’s see what this data actually tells us.
Diet soda increases risk of stroke and dementia – the paper
The study was published online in the journal Stroke on 20 April by MP Pase et al. Basically, the researchers seemed to show that consumption of even one can of diet soda (or more) each day was tied to an increased risk for stroke or dementia over a 10-year follow up period compared to those individuals who did not consume artificially sweetened beverages.
The authors studied 2888 participants, age greater than 45 years (mean age was 62), for occurrence of stroke. They also studied 1484 participants, age greater than 60 years (mean age of 69), for occurrence of dementia. Beverage intake was quantified using a questionnaire for each cohort.
The researchers adjusted the results for age, sex, education, caloric intake, diet quality, physician activity and smoking, so most confounding variables were eliminated from the study. For example, the data was adjusted so that smokers, who might be more at risk for stroke, would not make up more of one group or another.
The results were somewhat surprising. Those who had higher recent and cumulative intake of artificially sweetened soft drinks had a 2.96X higher risk for ischemic stroke than those individuals who had consumed none and a 2.89X higher risk for Alzheimer’s disease.
In this study, the researchers found that sugar-sweetened beverages were not associated with stroke or dementia. However, and this is a huge however, the same research group, in an article published earlier this year in Alzheimer’s & Dementia, concluded that “higher intake of sugary beverages was associated cross-sectionally with markers of preclinical AD.” So if you actually accept the findings of this paper, don’t count on drinking sugary drinks as being nutritionally superior to diet drinks.
Diet soda increases risk of stroke and dementia – the critique
Those of you who follow this blog may know that I rarely accept the findings of one article as “conclusive” evidence that correlation or causation exists. First of all, in the hierarchy of scientific publications, where a meta-review sits at the top and any article from Natural News sits at the bottom, a one-off primary research study which hasn’t been repeated by other, independent labs, has limited value as evidence.
But that’s just a meta level criticism of the study. Primary research can often lead to a new consensus in science, but more often than not, it is a dead end of science. But I have more specific appraisals of this study:
- There is a lack of hardy reasoning that would provide us with a plausible biochemical or physiological pathway whereby diet sodas would be causal to these two neurological issues. Of course, the inter webs are filled with claims that aspartame is a neurotoxin – no, they’re not. There is no evidence of aspartame being correlated with anything.
- The errors around the mean of most of the results were quite huge, and it included unity – in other words, there was a possibility that the results showed no increased risk to the consumption of these artificially sweetened drinks.
- The actual number participants in this study who had either dementia or stroke was tiny – there were less than 100 individuals who had ischemic stroke, while there were less than 60 individuals who had Alzheimer’s disease. Although one can show statistical significance with such low numbers, I’m always wary of unintentional bias showing up in such small numbers.
- The overall incidence for ischemic stroke for those who consumed artificially sweetened drinks was around 2.86%. If the risk is really 2.96X higher, then it can be argued that the difference in risk is not clinically significant, and can be biased by other risk factors which may have a small effect individually.
- The use of surveys in epidemiological studies has been questioned as to their utility and quality.
- The study cannot show cause and effect, because it is an observational study, meaning the results were observed after the consumption of either sugar filled or artificially sweetened soda. The problem is that it’s difficult to find which came first, the neurological issues or the consumption of the soda. Look at this way – maybe you have some cardiovascular risk factor, obesity, type 2 diabetes, or other issue, that leads you to a diet soda instead of a sugary one. So, which led to the stroke or dementia? Some physiological issue or the soda? This type of study can’t get us the answer.
- The study had 108 different statistical tests, all trying to give us a p-value, which states what the probability that the null hypothesis (for example, that diet soda is not related to Alzheimer’s disease) is true. This study has so many p-values that there’s a good possibility that there are false positives – we cannot tell which are false unfortunately. Moreover, many statisticians are finding problems with the overuse of p-values – for example, we can’t tell if the higher risk is a clinically higher risk. Will cutting off diet sodas lead to an actual significantly lower risk for either ischemic stroke or Alzheimer’s disease (or other dementia)?
- The study shows that Type 2 diabetes (T2DM) was higher in the diet soda group – obviously these patients are attempting to reduce their sugar consumption. T2DM is a known risk factor for both Alzheimer’s disease and stroke. If I were to fund a study of diet soda, I would hypothesize that “for a patient population that has been diagnosed with T2DM, the group that consumes diet soda would have a lower risk of ischemic stroke and Alzheimer’s disease.” I would have a much higher patient pool that can tease out small differences.
- This study ignores the possibility that the results have nothing to do with whether the drink is sugary or artificially sweetened – maybe another component of the drink is involved.
- I could find no other published studies that showed a link between Alzheimer’s disease and dementia. A fairly large study did show a statistically significant link between stroke and diet sodas – however, the increased risk was so small as to be almost clinically insignificant. Otherwise, there is little robust data out there that supports the hypothesis presented by Pase et al.
Diet soda increases risk of stroke and dementia – the summary
I am skeptical of this study’s ability to provide us with robust evidence that diet sodas are related to a clinically significant increased risk for ischemic stroke or dementia. I’m always driven mad when the popular press and your favorite website touts the danger of one food or another (for argument’s sake, an artificially sweetened soda is a food, though it has no nutritional value). There seems to be a buzz on the internet that diet soda has so many negatives, it’s probably wise to avoid those drinks. But that same buzz occurs for GMOs (completely safe) or high fructose corn syrup (just sugar), and “buzz” is not equivalent to actual scientific evidence.
If I were to make a recommendation, and I will because I’ll follow it myself, I’m going to stay neutral. I’m not going to change my diet soda intake because of this study. But if more evidence, and it has to be high quality evidence, confirms this study, I could change my mind. And those of you who buy into the precautionary principle, which I think is logically deficient and is similar to the argument from ignorance, you will probably cut out diet sodas. But if you have other risk factors for dementia or stroke, you’re probably not changing your overall risk.
On the other hand, I think this is a very interesting set of observations – in other words, stay tuned for better data in the future. But let me pop open my favorite libation, Diet Mountain Dew, and enjoy the afternoon.
- Bernstein AM, de Koning L, Flint AJ, Rexrode KM, Willett WC. Soda consumption and the risk of stroke in men and women. Am J Clin Nutr. 2012 May;95(5):1190-9. doi: 10.3945/ajcn.111.030205. Epub 2012 Apr 4. PubMed PMID: 22492378; PubMed Central PMCID: PMC3325840.
- Nieuwenhuijsen MJ. Design of exposure questionnaires for epidemiological studies. Occup Environ Med. 2005 Apr;62(4):272-80, 212-4. PubMed PMID: 15778263; PubMed Central PMCID: PMC1740990.
- Pase MP, Himali JJ, Beiser AS, Aparicio HJ, Satizabal CL, Vasan RS, Seshadri S, Jacques PF. Sugar- and Artificially Sweetened Beverages and the Risks of Incident Stroke and Dementia: A Prospective Cohort Study. Stroke. 2017 May;48(5):1139-1146. doi: 10.1161/STROKEAHA.116.016027. PubMed PMID: 28428346.
- Pase MP, Himali JJ, Jacques PF, DeCarli C, Satizabal CL, Aparicio H, Vasan RS, Beiser AS, Seshadri S. Sugary beverage intake and preclinical Alzheimer’s disease in the community. Alzheimers Dement. 2017 Mar 6. pii: S1552-5260(17)30050-X. doi: 10.1016/j.jalz.2017.01.024. [Epub ahead of print] PubMed PMID: 28274718.