Chevron’s Cancer-causing Fuel
Did the EPA approve fuel ingredients with a 130% cancer risk?
An article published by the news source ProPublica has caused a lot of controversy around the internet in the past few months after claiming that new jet, boat, and gasoline fuel ingredients developed by Chevron have been approved by the Environmental Protection Agency (EPA) despite having a lifetime cancer risk of 1.3 in 1 for those exposed to it.
When I first read this, I immediately had alarm bells going off in my head. A 1.3 in 1 absolute risk means that people exposed to this fuel ingredient would be 130% likely to develop cancer in their lifetime. But, the likelihood of an event happening must be between 0% and 100%, with 0% meaning that there is no chance of the event happening and 100% meaning that the event is guaranteed to happen. When a risk calculation like this returns results greater than 100%, it means that there are errors in the data or in the calculation itself. So, let’s look into ProPublica’s report and see if we can figure out what’s going on.
The first thing to note is that the 1.3 in 1 figure did not come from the EPA or the scientists who work for Chevron. Rather, it came from the journalists at ProPublica. ProPublica got access to the risk assessment document via a Freedom of Information Act request and used the data they found in the document to calculate the 1.3 in 1 figure. It’s important to note that the figures 1.3 in 1 and 130% are nowhere to be found in the actual risk assessment document. Despite the calculations being done by journalists and not the scientists who wrote the risk assessment, ProPublica claims that their calculations were confirmed by the EPA. However, they did not state whether the EPA simply confirmed their math or determined that the data they used to do such calculations were accurate. After all, you can do correct calculations but still end up with bogus results if your data points are inaccurate. When ProPublica asked a scientist from the EPA about the 1.3 in 1 figure, they stated “This is ridiculously high.” I’d agree considering that the figure doesn’t even make sense.
If we assume that the journalists at ProPublica did their math correctly, then that just leaves bogus data as the culprit; and in my opinion, that’s likely what has happened. According to the risk assessment, a large majority of the data points are estimations reported with high levels of uncertainty. For example, when reporting on the risks of cancer through inhalation, the assessment states:
“For cancer inhalation risks for the general population, NCD [New Chemicals Division] used an average annual air concentration when calculating the lifetime average daily dose estimate and assumed exposure to that concentration throughout the lifetime. The use of this assumption would overestimate risks to individuals who spend less than a lifetime at the affected location.”
The risk assessment is flat-out saying that the risk estimations assumed that a person would be exposed to the affected air throughout their entire lifetime and those that spent less than their entire lives exposed to the affected air (pretty much everyone) would have lower risks. This is the exact type of overestimation that could lead to bogus cancer risk calculations as most individuals would not be spending every day of their lives in areas polluted with these fuel ingredients.
ProPublica knew about these estimations and even reported that the EPA stated:
“Cancer risk estimates were extremely unlikely and reported with high uncertainty.”
They also reported that:
“The EPA said it had significantly overestimated the cancer risks posed by both the jet fuel and the component of marine fuel. The agency assumed, for instance, that every plane at an airport would be idling on a runway burning an entire tank of fuel, that the cancer-causing components would be present in the exhaust and that residents nearby would breathe that exhaust every day over their lifetime.”
These are bogus assumptions and allow us to see how such overestimations can lead to bogus calculations. The estimations assumed that a person would be breathing in the exhaust fumes produced by all the planes at an airport idling until their tanks were empty. They also assumed that the chemicals in question would be found in the exhaust fumes in the first place.
ProPublica did their due diligence by mentioning these problems, but the details were hidden further down in a lengthy article with the title “EPA Approved a Fuel Ingredient Even Though It Could Cause Cancer in Virtually Every Person Exposed Over A Lifetime.” Of course, the public is going to freak out over such a sensationalized headline. Only a small percentage of those exposed to the article are going to read any further than the headline, and of those that do, an even smaller percentage are going to read the 203-page risk assessment for themselves to verify any information.
Under the “Risk Conclusions” section of the risk assessment, estimates of cancer risk due to drinking water pollution were between 1.3E-10 and 1.7E-08. This means that, on the low end, only 1.3 people out of every ten billion people exposed to drinking water polluted with these fuel ingredients would be expected to get cancer. On the high end, 1.7 people out of every one hundred million exposed would be expected to get cancer. For reference, in 2020, 403 out of every one hundred thousand Americans got cancer according to the Centers for Disease Control and Prevention. According to the risk assessment, cancer risks from drinking water polluted with these fuel ingredients are more than 237,000 times lower than the rates of general cancers diagnosed in America.
In terms of inhalation, fugitive air risks of the ingredients ranged from 8.3E-8 and 1.2E-4 (8.3 in one hundred million to 1.2 in ten thousand). Fugitive air refers to accidental leaks and/or irregular releases of gases from things like pipes, storage containers, wells, and other equipment.
The highest cancer risk estimation reported in the risk conclusions from the assessment was 2.5E-01, meaning that 2.5 out of every ten people (one quarter) would develop cancer, but this risk was associated with stack air from one of the 18 ingredients. Stack air refers to air coming directly from smokestacks (the big pillar-looking chimneys at factories that allow smoke, water vapor, and other gases to escape the building). While this sounds alarming, it’s important to remember that this risk only refers to air directly leaving the smokestacks. In order to inhale this air you’d need to be right above said smokestack. After the air leaves the smokestack it immediately starts to dissipate into the larger volume of air around it, diluting any potential pollutants and lowering the cancer risk. Smokestacks are often built high into the air (some around 500 feet or higher) in order to limit the amount of pollution that makes it into the air closer to ground level. Other technologies can be employed to clean pollutants from the gas before it is sent out of the smokestacks, such as scrubbers and precipitators, to further reduce pollutants. Because of these precautions, the 2.5E-01 figure would not apply to your average person breathing the regular air unless you just so happen to routinely work on top of smokestacks featuring this particular ingredient in their exhaust without proper ventilation equipment. Even if the 2.5E-01 risk did apply to the average person, it’s still more than five times lower than the 1.3 in 1 provided by ProPublica.
These fuel ingredients are likely carcinogenic to some degree given adequate exposure and the risk assessment states such. But, the risk of cancer has likely been exaggerated to the general public so much as to say that every single person exposed to them is guaranteed to get cancer. It’s not clear why these ingredients would be given the okay by the EPA if this were truly the case. Not only are the calculations based on highly exaggerated and estimated data points, but they also assume that people will be exposed to high doses of these ingredients every single day of their lives. These are worst-case scenarios and were reported with high degrees of uncertainty. It’s more than likely that the EPA has taken these estimations and uncertainties into account when granting approval. On top of that, ProPublica did not reveal the calculations they used to come up with the 1.3 in 1 figure so it’s impossible to verify. Their figures do not show up anywhere in the actual risk assessment document.
