Someone switches from regular soda to diet soda, feels reasonably good about it, and then reads a headline saying diet soda causes strokes. That has been the pattern for about fifteen years now, and the evidence underneath it has moved enough to be worth going through properly.
Where the scare started
The original signal came from the Northern Manhattan Study. Gardener and colleagues followed 2,564 people for a mean of ten years and counted 591 vascular events, meaning stroke, myocardial infarction or vascular death. People drinking diet soft drinks daily had a hazard ratio of 1.43 compared with people drinking none, with a confidence interval of 1.06 to 1.94 (1).
What made it interesting is that the association held after adjusting for metabolic syndrome, diabetes, hypertension, cardiac disease and hypercholesterolemia, and that regular soft drinks showed no such association in the same cohort (1).
A confidence interval running from 1.06 to 1.94 is a study saying “something is probably here, and we cannot tell you how big.”
What came after
The Women’s Health Initiative analysis in 2019 is the one that made me take this more seriously. 81,714 postmenopausal women. Those consuming two or more artificially sweetened beverages a day, compared with less than one a week, had a hazard ratio of 1.23 for all stroke, 1.31 for ischemic stroke, 1.29 for coronary heart disease and 1.16 for all-cause mortality (2).
The subgroup finding is the one that sticks. In women with no prior cardiovascular disease and no diabetes, high consumption was associated with more than double the risk of small artery occlusion ischemic stroke, hazard ratio 2.44 (2). That subgroup matters precisely because it is the group where reverse causation is least likely to explain the result.
Then erythritol. Witkowski and colleagues published in Nature Medicine in 2023, working across a discovery cohort of 1,157 patients and US and European validation cohorts of 2,149 and 833. Circulating erythritol in the highest quartile was associated with increased three-year risk of major adverse cardiovascular events, and they followed it with mechanistic work on platelet reactivity (3).
Erythritol is not aspartame or sucralose. It is a sugar alcohol used heavily in stevia and monk fruit blends and in a great deal of what gets sold as keto-friendly. Somebody avoiding artificial sweeteners on principle may be consuming more erythritol than anyone in that study.
The problem with all of it
Reverse causation. It is the whole difficulty and it does not go away.
People do not drink diet soda at random. They drink it because they are worried about their weight, or because someone told them to cut sugar, or because they already have diabetes. So a cohort of daily diet soda drinkers is enriched for people who were already heading toward a vascular event before they ever opened a can. Statistical adjustment helps and does not fix it, because you can only adjust for what you measured.
There is no randomized trial showing that drinking diet soda causes strokes. There is not going to be one. The trials that do exist are short and use intermediate endpoints, and they generally show that substituting a sweetened beverage for a sugar-sweetened one produces modest weight loss.
The World Health Organization landed on this in 2023 and issued a conditional recommendation against using non-sugar sweeteners for weight control, on low-certainty evidence, noting possible long-term associations with type 2 diabetes and cardiovascular disease (4). Conditional and low-certainty are doing real work in that sentence. It is not a ban and it was not presented as one.
What I think this adds up to
Diet soda is not established as a cause of stroke. The associations are consistent across several large cohorts, which is more than nothing, and they are exactly what confounding by indication would also produce, which is why nobody has closed the case.
Here is what I tell patients, and I separate it by what they came in for.
If you have diabetes, diet soda does not spike your blood sugar the way regular soda does. So if you are not going to give up soda completely, diet is the better of the two, and I would rather you drink it than pretend you are going to stop.
That is the whole of the defense. Diet soda is not healthy, and we know that. Some of what is in it has been linked to health problems, which is most of this post. And there is a second issue that matters more in a weight-management visit than the vascular question does: some research suggests diet soda makes you hungrier.
Yunker and colleagues ran a randomized crossover trial in 74 adults, published in JAMA Network Open in 2021, comparing sucralose against sucrose. Sucralose produced less suppression of acyl-ghrelin, the hunger hormone, and the neural responses in appetite and reward regions differed by sex and by BMI, with women and people with obesity looking most sensitive to the difference (5).
Most people do not want to be hungry. They want to lose weight, not gain it. If a drink is quietly making the day harder, that is a real cost even if nobody’s carotids are worse for it.
I will give you the trial that cuts against me, because it exists and it is better designed than most of what is in this post. Harrold and colleagues randomized 493 people in a 52-week behavioral weight-management program to either water or non-nutritively sweetened beverages. The sweetened-beverage group maintained more weight loss, 7.5 kg against 6.1 kg, a difference of 1.4 kg (6). That is a real randomized result and it says that for someone actively working a program, swapping to diet drinks did not sabotage them and may have helped.
So my position is not that diet soda is dangerous. It is that it is not good for you, the appetite signal is worth taking seriously in someone trying to lose weight, and if you are diabetic and choosing between two sodas, choose the one that does not move your glucose.
For patients
If you switched from regular soda to diet, do not switch back on the strength of these headlines. That trade still looks favorable.
If you are drinking several a day and trying to lose weight, cutting down is reasonable, and the best reason is appetite rather than any proven vascular danger. If you specifically avoid artificial sweeteners and use stevia or monk fruit blends, check the label for erythritol, because that is the one with the most concerning recent data.
For colleagues
The small artery occlusion subgroup in the WHI analysis is the finding worth knowing, because it is the one least explained by reverse causation.
Also worth knowing that patients hear “conditional recommendation on low-certainty evidence” as “WHO says diet soda is dangerous,” and correcting that is a two-sentence conversation that is usually worth having.
The Bottom Line
Daily diet soda has been linked to vascular events in several large cohorts, with hazard ratios in the 1.2 to 1.4 range, and the association survives adjustment. It has never been shown to cause them, and the people who choose diet soda are different from the people who do not in ways no adjustment fully captures. Diet soda beats regular soda. Water beats both. If you have diabetes and you are not giving up soda, take the one that does not move your glucose. If you are trying to lose weight, the argument against it is appetite rather than your arteries. And the most interesting recent finding is not about aspartame at all, it is about erythritol, which is in a great many products marketed to people who are trying to avoid artificial sweeteners.
Related Reading
Stroke Warning Signs: What to Know About CVA and TIA
Headaches: Migraine, Cluster, Tension and Chronic Daily Pain
How Do You Actually Lose Weight? A Doctor Explains
Newly Diagnosed With Type 2 Diabetes: What You Should Know
Doctor Supervised Weight Loss: What Works Long Term
Scott Rennie, D.O.
Board Certified in Obesity Medicine and Family Medicine
This blog is for educational purposes only and does not constitute individual medical advice. Always consult your own physician before making changes to your health, medications, or treatment plan.
Sources
1. Gardener H, Rundek T, Markert M, et al. Diet soft drink consumption is associated with an increased risk of vascular events in the Northern Manhattan Study. J Gen Intern Med. 2012;27(9):1120-1126. PMID 22282311. https://pubmed.ncbi.nlm.nih.gov/22282311/
2. Mossavar-Rahmani Y, Kamensky V, Manson JE, et al. Artificially Sweetened Beverages and Stroke, Coronary Heart Disease, and All-Cause Mortality in the Women’s Health Initiative. Stroke. 2019;50(3):555-562. PMID 30802187. https://pubmed.ncbi.nlm.nih.gov/30802187/
3. Witkowski M, Nemet I, Alamri H, et al. The artificial sweetener erythritol and cardiovascular event risk. Nat Med. 2023;29(3):710-718. PMID 36849732. https://pubmed.ncbi.nlm.nih.gov/36849732/
4. World Health Organization. Use of non-sugar sweeteners: WHO guideline. 2023. https://www.who.int/publications/i/item/9789240073616
5. Yunker AG, Alves JM, Luo S, et al. Obesity and Sex-Related Associations With Differential Effects of Sucralose vs Sucrose on Appetite and Reward Processing: A Randomized Crossover Trial. JAMA Netw Open. 2021;4(9):e2126313. PMID 34581796. https://pubmed.ncbi.nlm.nih.gov/34581796/
6. Harrold JA, Hill S, Radu C, et al. Non-nutritive sweetened beverages versus water after a 52-week weight management programme: a randomised controlled trial. Int J Obes (Lond). 2024;48(1):83-93. PMID 37794246. https://pubmed.ncbi.nlm.nih.gov/37794246/