How I Learned to Stop Worrying and Love Diet Coke
Aspartame and Why Everything Causes Cancer
In a world where everything seems to cause cancer, the World Health Organization has added one more item to your "Things to Worry About" list: aspartame. Yes, that seemingly harmless sweetener in your diet soda has just been branded a "possible carcinogen." But before you pour your beloved Diet Coke down the drain, let's take a moment to consider whether this is a story of diligent public health protection or a cautionary tale of institutional failure. (Spoiler alert: it's the latter.)
IARC's Classifications
Last year, the World Health Organization's International Agency for Research on Cancer (IARC) made headlines when it declared aspartame, the artificial sweetener commonly found in Diet Coke and many other low-calorie products, a “possible carcinogen”. This classification was primarily based on the findings of three epidemiological studies: Stepien et al., Jones et al., and McCullough et al.
The studies present a mixed and inconclusive picture: Stepien et al. identified a small but statistically significant correlation between artificially sweetened beverages (ASBs)1 and liver cancer, while Jones et al. observed this association only among individuals with diabetes. McCullough et al. examined cancer associations across the entire body2 but found no link between ASBs and liver cancer in the general population. However, McCullough et al. did report an association with certain obesity-related cancers, though it became statistically insignificant after adjusting for BMI, and a significant link between ASB consumption and pancreatic cancer3.
Overall, I find the evidence unconvincing. Although the researchers in epidemiological studies attempt to correct for confounding factors, the studies are still limited in the ways that all observational studies are. It’s unlikely that a massive epidemiological study would find absolutely nothing4. In fact, the most statistically significant finding for ASB consumption was a reduction in the risk of lung cancer found by McCullough et al. But no one is promoting aspartame as a way to prevent lung cancer.
To get a sense of how these studies are limited, imagine the following two people:
Alice: She starts her day with a kale smoothie, drinks eight glasses of water, and unwinds with chamomile tea. Her idea of a wild night is having an extra slice of gluten-free, sugar-free, joy-free pie.
Bob: His blood type is Diet Coke. He starts his day with a can, washes down his lunch with two more, and falls asleep clutching a two-liter bottle like a teddy bear.
Now, if Bob develops health issues, is it fair to point the finger solely at his aspartame intake? This is why we should be careful about weighing observational studies too much.
The results of these observational studies, in particular, could be due to a simple confounding variable. Imagine someone who leads an unhealthy life, drinks sugary beverages all day, and gets diabetes. As a diabetic, they now have a higher risk of cancer, specifically liver cancer. They go to their doctor who tells them they need to be more careful with their sugar intake so they should switch to diet soda. Now they have a higher risk of cancer and higher consumption of aspartame, but the aspartame wasn’t the cause of the increased cancer risk.
Another reason to be cautious about labeling aspartame a carcinogen is that we understand how it is broken down in the body, and the resulting compounds are innocuous. After aspartame enters the body, it is rapidly hydrolyzed into three naturally occurring compounds: aspartic acid (a non-essential amino acid), phenylalanine (an essential amino acid), and a small amount of methanol (an alcohol also found in fruits and vegetables). No new or harmful chemicals enter the bloodstream as a result5.
Overall, it’s very unconvincing that aspartame causes cancer.
Fortunately, the Joint FAO/WHO Expert Committee on Food Additives (JECFA) also evaluated the data and “concluded that there was no convincing evidence from experimental animal or human data that aspartame has adverse effects.” In addition, the FDA has opposed this ruling, stating that it “disagrees with I.A.R.C.’s conclusion.” The FDA estimated that a person weighing 132 pounds would need to consume more than 75 packets of aspartame a day to exceed the Acceptable Daily Intake. Good on the JECFA and FDA, but this doesn’t undo the damage of yet another “X causes cancer” story making the rounds.
The Boy Who Cried Carcinogen
The problem goes far beyond a single errant classification of aspartame. It sounds scary to think that something might be a carcinogen, like it’s a dangerous product you should avoid. But when you look into it, you see that IARC has a habit of being a bit slap-happy with that label.
Let’s back up and look at what IARC does. IARC classifies things into four (or five) groups:
Group 1: "Carcinogenic to humans"
Group 2A: "Probably carcinogenic to humans"
Group 2B: "Possibly carcinogenic to humans"
Group 3: "Unclassifiable as to carcinogenicity in humans"
Group 4: "Probably not carcinogenic to humans"
If you look into those groups you’ll see what I mean. The full list of IARC classifications is available on their website. If you go to the link, the top item you’ll see on the list is aloe vera, which is in the same group—2B, possible carcinogens—as aspartame. That soothing gel you put on sunburns? Yes, possible carcinogen.
I find the Wikipedia page for IARC group 2B much easier to navigate. There, in addition to aspartame and aloe vera, you’ll find:
Pickled vegetables (but only the “traditional Asian” kind, because apparently cancer can tell the difference)
Nickel (bad news for coin enthusiasts)
Kava extract (sorry, relaxation seekers)
Radiofrequency electromagnetic fields (yes, that’s cell phones, wifi, etc.)
Carpentry and joinery (Jesus was a carpenter. Coincidence? I think not.)
In IARC group 2A, the probable carcinogens, we have red meat, art glass, being a hairdresser or barber, and night shift work. There’s also IARC group 1, substances that have been classified as carcinogenic. It includes HIV and plutonium, but also alcoholic beverages and processed meats.
Why are bacon, beer, and HIV on the same list of carcinogens? These aren’t lists that are designed to help people. They’re lists created by a bureaucracy, where the path of least resistance is paved with caution tape.
The WHO's communication strategy is not just confusing; it fundamentally undermines its credibility. After labeling aspartame a “possible carcinogen,” their head of nutrition said, “Our results do not indicate that occasional consumption should pose a risk to most.” After red and processed meats were added to the list, the spokesman for the WHO said that “IARC’s ruling did not mean people should stop eating meat”.
I don’t think you can do that. What on Earth does it mean for something to be on a list of carcinogens, along with HIV and plutonium, if it doesn’t mean “don’t intentionally put it in your body”?
One Flew Over the Bureaucrat’s List
IARC's classification of aspartame as a "possible carcinogen" is not an isolated incident, but rather a symptom of a deeper, systemic problem plaguing many public health institutions. This failure stems from the combination of misaligned incentives, risk aversion, and bureaucratic inertia.
The result is a classification system so cautious it borders on absurdity. Consider this: out of all the substances IARC has ever evaluated, only one has been classified into group 4. That is, there is only one thing in the world that IARC will tell you is “probably not carcinogenic to humans.” It’s caprolactam, a compound used in synthetic fibers like nylon.6
Another source of problems is that adding items to the “potentially dangerous” list is easy, but they are loath to declare something less dangerous. In 1991, coffee was added as a possible carcinogen. In 2016, 25 years later, they moved it to group 3—unclassifiable. Maybe in 20 more years, they’ll figure out what the rest of us already know—coffee doesn’t cause cancer.
Cell phones probably never should have been on that list. But even if there was some uncertainty about cell phones and brain cancer, it’s been resolved by now. Unless you’ve spent a career working for the CIA or NSA in tightly sealed Faraday cages, you’re constantly inundated with radiofrequency electromagnetic fields. If there was a link to cancer, we would have seen it by now. As the Harvard School of Public Health says, “a number of large studies have found no real evidence of such a link.”
The result is a bloated catalog of maybes and might-bes, where your morning bacon is deemed as risky as handling radioactive waste. It's not science; it's CYA masquerading as public health policy.
We Need Better Public Health Institutions
The story of aspartame's classification as a "possible carcinogen" is more than just an example of bureaucratic overreach—it's a microcosm of a larger institutional failure in public health communication. These institutions that we rely on to tell us what’s dangerous have cast such a wide net in their classifications that they've essentially rendered their warnings meaningless. When everything from aloe vera to cell phones is labeled as potentially cancer-causing, the public is left not with useful information, but with a sense of confusion and cynicism toward the very warnings meant to protect them.
Instead of endless worry and safetyism, I propose another way to think about aspartame: We are incredibly fortunate that it exists. We’re facing an obesity epidemic and sugar plays a large role in it. We need to get off sugar but we’re still consuming more than 300% of the recommended daily amount. And it just so happens that there exists a chemical that tastes 200 times as sweet as sugar, so a tiny bit of it is all we need, and it hydrolyzes into organic compounds.
Aspartame is a workable answer to our obesity epidemic. But instead of vocally promoting it as a cause we should be celebrating, the WHO managed to turn it into another scary “possible carcinogen”.
We need public health institutions. But it’s worth these institutions going back to the drawing board and asking, “How can we help people make the decisions they need to make?” Most of us don't have the time or expertise to evaluate the safety of every substance we encounter. These lists are filled with countless chemicals I know nothing about. I have no idea what N-Ethyl-N-nitrosourea is, but it’s on IARC’s Group 2A list. What does that mean? Is it safe to touch it? Make a shoe out of it? Spread it on a piece of toast? I have no idea.
They classify many substances, such as aspartame, as "possible carcinogens". By now, you might be assuming I'm a shill for Big Aspartame, but even I won't claim it's IMpossible for aspartame to cause cancer—that's simply not true for any substance. I do believe there's an amount of aspartame you could force-feed a rat that would result in harmful health effects. So, while their declarations are "defensible" in the sense that almost anything can be labeled a potential carcinogen, the labeling classifications are almost guaranteed to cause concern and confusion for any substance they investigate.
The current approach fails on three levels:
It ignores the fundamental principle of toxicology: As the 16th-century Swiss physician Paracelsus wisely noted, "the dose makes the poison7." By failing to distinguish between high-dose animal studies and realistic human exposure, these institutions create false equivalences that muddy the waters of public understanding.
It creates a boy-who-cried-wolf scenario: When institutions label both plutonium and pickled vegetables as carcinogens, they undermine their own credibility. The public becomes desensitized to these warnings, potentially ignoring legitimate health concerns in the process.
It can lead to unintended consequences: In the case of aspartame, scaring people away from artificial sweeteners might inadvertently push them towards sugary drinks, trading a dubious and hypothetical risk for a certain one.
A Better Way Forward
So, how do we move forward? We shouldn’t simply abolish these institutions—public health is an important role that we need institutions to play in society. People lack the time and expertise to evaluate the safety of every substance they encounter. Instead, we need a recalibration of how these organizations approach risk communication:
Communicate in the way people think: IARC evaluates hazard rather than risk. This means they identify whether a substance could potentially cause cancer, not the likelihood of it doing so under typical exposure levels. By not differentiating between hazard and actual risk, the IARC's classifications create public confusion. They should communicate the risk.
Prioritize transparency: Clearly communicate the strength of evidence behind each classification. Is it based on robust human studies or preliminary animal research?
Continue to reduce uncertainty: Substances shouldn’t be placed on “possibly carcinogenic” lists indefinitely. It shouldn’t be a final resting place. We should prioritize understanding in what cases they might be carcinogenic, and not be afraid to declare them generally recognized as safe if they are.
Consider real-world implications: Before issuing warnings, carefully weigh the potential for unintended consequences. Will this information lead to better health outcomes or just more confusion or apathy from overload?
Empower the public: Instead of simply issuing edicts, provide tools and frameworks for individuals to make informed decisions based on their personal risk tolerance and lifestyle.
The magnitude of this institutional failure is not just astonishing—it's a public health crisis in itself. In their current form, I don’t see how IARC adds any value to society. I don’t know what we’re supposed to do as non-experts but ignore the lists they create. It's time we demand better from our public health institutions.
The studies didn’t measure aspartame usage directly but instead used ASBs as a proxy for it.
McCullough et al. looked for associations with the following cancers: larynx, esophageal, stomach, liver, pancreatic, gallbladder, lung, small intestine, colorectal, breast, uterine, ovarian, prostate, kidney, bladder, melanoma, brain, non-Hodgkin lymphoma, multiple myeloma, and leukemia.
It sounds scary that McCullough et al. found an association between ASB consumption and pancreatic cancer, but it’s important to see this in context. The study checked for cancer all over the body; not because we have some plausible reason to believe that any of these would occur, but just to look for correlation. It’s a dragnet search. These studies are good to do, but we need to understand their limitations.
Assuming they don’t use a Bonferroni correction, which many studies do not.
Note that this is not true of other sweeteners, which is why it would have been better for the studies not to have combined them. Sucralose, for example, passes through the body unchanged. 10-15% will be absorbed by the small intestine and enter the bloodstream, so sucralose will circulate in the blood until it is filtered out by the kidneys.
In case you’re a caprolactam manufacturer and are feeling left out of the carcinogen list, you might be relieved to hear that, according to the EPA, “Acute (short-term) exposure to caprolactam may result in irritation and burning of the eyes, nose, throat, and skin in humans. Headaches, malaise, confusion, and nervous irritation have been observed in workers exposed to caprolactam by inhalation.”
This is the common paraphrase of "All things are poison, and nothing is without poison; the dosage alone makes it so a thing is not a poison."
One example of an organization that does public health communication well is examine.com
Sad fact: Diet coke seemingly everywhere in the world other than the US contains ace K as well as aspartame. (I reckon that ace K is probably fine, but no other sweetener has the same evidence for safety as aspartame.)