We can add decaffeinated (decaf) products to the list of things you probably shouldn’t worry about but someone wants to make you worry anyway. You may have read recently that some decaf coffee and tea brands have “chemicals” in them that may be a health risk. The real story, as you might imagine, is more complicated.
The cause of the recent headlines is that several activist groups have petitioned the FDA to ban the most common chemical used in the decaffeination process, methylene chloride. Once again this is essentially a debate about hazard vs risk. But let’s start with some background on the decaffeination process.
There are several common methods for decaffeination, the most common being the so-called European method, or the indirect solvent method. This method first boils green coffee beans to soften them up and remove some flavor components from the beans. It then exposes the beans to methylene chloride or ethyl acetate to bind and remove the caffeine. The beans are then washed to remove the solvent and caffeine. The beans are then heated again to further evaporate off the volatile chemicals and then recombined with the water to return the flavor chemicals. The beans are then dried and roasted.
The advantage of this method is that it is the fastest and cheapest method of decaffeination. It is therefore the most popular. Using this method with ethyl acetate is often marketed as a “natural” alternative, or a “sugar” method because ethyl acetate can be sourced as a byproduct of sugar processing. However, it can also just be manufactured.
Alternatives include various water methods that allow the beans to soak for a longer period of time, leaching out the caffeine and many other chemicals. Caffeine is the largest chemical, however, so it can be removed with charcoal filters, and the other flavor chemicals then returned to the beans. There is also a carbon dioxide method where CO2 is bubbled through the beans, binding to caffeine, and then removed.
Is there any difference among these methods in terms of the health safety of the resulting tea or coffee? The answer here is pretty clearly no. The FDA has set the safety limit of methylene chloride in food products at 10 parts per million. This is based on safety data, usually with an order of magnitude buffer built in. Many news outlets, including WebMD, mention that a 2020 test by an activist group, the Clean Label Project, did a study of different decaf brands and found methylene chloride in many of them. But most outlets fail to mention that in every single case the levels were below the FDA limits, often far below. There is no evidence that exposure at these trace levels poses any health risk.
Which brings us back to the hazard vs risk debate. The FDA uses a risk-based model in determining whether or not to allow use of chemicals in products and sets safety limits. However, some advocacy groups and regulatory agencies rely on a hazard approach (whether or not they explicitly acknowledge this). If a chemical can potentially be harmful, it should be banned, even if it is not demonstrated to be harmful at the doses consumers will be exposed to.
There isn’t necessarily one correct approach – it depends on what tradeoffs we are willing to make. Do we want to err on the side of the precautionary principle (the hazard approach), or see everything as a risk vs benefit calculation? I do think the risk vs benefit approach is more rational. The hazard approach casts too wide a net, and leads to unintended consequences that may actually increase net risk for consumers. In order to avoid imagined risks people may change their behavior in a way that increases their risk (such as drinking more caffeine).
Economics should not be ignored either. Often consumers make economic choices, especially those at the lower SES levels. Mandating more expensive products or methods in order to avoid a potential hazard with no proven risk may have a net negative impact on overall food quality. Even just scaring people into more expensive options, like organic produce (which has no proven health advantage over traditionally-farmed produce) may result in lower consumption of fresh fruits and vegetables, which is likely to have a negative health impact.
Often groups like the Clean Label Project explicitly argue for consumers to make informed choices, and I agree. But this works both ways – you can’t advocate for informed choices and then fearmonger about hazards without putting them into context of actual risk. Trace chemicals from solvent-based decaf methods are below safety limits and pose no known health risk.
There are, however, other issues that complicate the choice of decaf method. You can find lots of taste tests comparing different methods. They do remove and then replace lots of chemicals that influence the flavor of the coffee, so there is a plausible mechanism here. I would just take them with a grain of salt, unless they are well-blinded tests (unblinded taste tests are notoriously subject to bias and expectation). Also, the only taste test that matters is you.
There is one very legitimate issue, and that is industrial and environmental exposure to chemicals like methylene chloride. Again, the chemical is a carcinogen, and can be a risk to workers who are exposed. This is less of an issue for the FDA and more one for OSHA and the EPA. Work standards should protect workers and keep exposures below safe limits, and we should definitely regulate which chemicals find their way into the environment and in what amounts. For these reasons the use of methylene chloride has been limited over the years, and that’s reasonable.
But there is no reason for consumers to worry about the decaf method from a personal health perspective. This is essentially another round of hazard-based fearmongering from activist groups, not spawned by any new science or evidence of risk.
