Support journalism that lights the way through the climate crisis

Goal: $100k

High levels of carcinogenic gas near a medical sterilization facility in Scarborough, Ont., highlight a dangerous gap in Canada’s air quality monitoring, says the Environment and Climate Change Canada researcher who discovered the chemical in the area.

Ethylene oxide is a colourless gas used to sterilize medical equipment that can't be heated, like syringes or gauze; it’s also used in some industrial processes and to sterilize spices. The chemical has recently come under scrutiny in the U.S., where a review of government air quality guidelines showed several communities had been exposed to dangerous levels of the gas for years.

The ECCC research team found "detectable" plumes of ethylene oxide within a 900-metre radius of a now-closed Sterigenics medical sterilization plant in Scarborough during air quality testing over a week in December 2021.

The snapshot measurements were consistent with levels that exceeded Ontario's air quality guidelines, but because the guidelines are measured over a year, the researchers couldn't assess if they were being exceeded. A spokesperson for the province said: "The data cannot be compared to Ontario's air standards."

Since the tests were done, Sterigenics, an international company that operates in North America, Europe, Asia and South America, has replaced the Scarborough plant with a new one in Mississauga, which it says is equipped with a "state-of-the-art emissions control system” that can capture "99.9 per cent of total ethylene oxide emissions."

Most air monitoring in Canada focuses on the gases and particles behind smog and two toxic pollutants typically linked to fossil fuel production and use: benzene and polycyclic aromatic hydrocarbons (PAHs). Other so-called "air toxics" haven't "really been part of Canada's regular air monitoring," said research leader Elisabeth Galarneau.

Take ethylene oxide: Despite a 2001 federal risk assessment warning the gas has "a probability of harm at any level of exposure," the federal government has almost no information about where Canadians might be exposed to it.

The only way to estimate exposure to the gas comes from Canada's National Pollutant Release Inventory (NPRI), a database where mid- to large-sized companies must report their emissions of some chemicals, including ethylene oxide. In 2021, only five facilities reported ethylene oxide emissions to the database: the Scarborough Sterigenics plant, an Ontario environmental cleaning company and three petrochemical facilities in Alberta and Saskatchewan.

However, Galarneau said they are only required to report their emissions if they manufacture, process or use over 10 tonnes of ethylene oxide at concentrations over one per cent per unit of weight. That means smaller emitters like hospitals or spice distributors might be overlooked. For instance, she cited a 2009 study that found elevated levels of the gas in B.C., even though according to the NPRI, there are no ethylene oxide emitters in the province.

High levels of carcinogenic gas near a medical sterilization facility in Scarborough, Ont., highlight a dangerous gap in Canada’s air quality monitoring, says the ECCC researcher who discovered the chemical in the area.

"In terms of having a national picture, that's the information we have so far," she said, noting Canada currently does not even have the technical capacity to regularly monitor for the gas. Her research relied on a high-tech instruments American researchers loaned to her for the project.

That's a problem, said Canadian Association of Physicians for the Environment toxics program director Dr. Jane McArthur. While the releases might be too small to be captured in the NPRI, they could still reach levels "that are of concern in particular facilities," posing a risk to employees and nearby residents.

Health-care workers, especially women, are particularly concerned, she said, noting Canadian occupational health and safety data shows that about 2,400 Canadians are exposed to the chemical and 62 per cent are women. People working in health care and food processing are those most exposed, according to data from CAREX, a cancer research organization.

Recent studies have found ethylene oxide exposure is linked to breast cancer in American women living within a 10-kilometre radius of facilities emitting the gas. The U.S. Environmental Protection Agency (EPA) also noted last month that the chemical is linked to higher cancer rates for people who are regularly exposed. Canada does not make epidemiological data for such small geographical regions or job categories publicly available due to privacy concerns.

Despite these findings and the EPA's decision to regulate the gas more strictly, the federal government "is not currently considering the development of additional controls for ethylene oxide used as a sterilant," ECCC said in a statement. Canada currently regulates the chemical with guidelines developed in 2004 that are based on a risk assessment published a year earlier.

"It's a system that needs improvement," McArthur said.

She would like to see the government do a better job of monitoring exposure and assessing the risk that air toxics like ethylene oxide pose to vulnerable groups.

Provisions that would require the government to conduct this type of assessment are included in the proposed update to the Canadian Environmental Protection Act (CEPA) and in another bill targeting environmental racism. The CEPA update is still undergoing parliamentary review, she noted, adding questions remain about whether Canada has the technical capacity to properly monitor for the gas alongside other toxics.

These changes could have a major impact on Canadians' well-being, McArthur said. While some toxins might escape government monitoring efforts, she regularly talks to physicians and nurses who are seeing the impact of toxic chemicals in real-time.

"They are seeing it in the health of their patients and their communities," she said. "And those are very compelling stories."

Updates and corrections

| Corrections policy
May 11, 2023, 10:30 am

Editor's note: This story was corrected to more accurately reflect the criteria companies using ethylene oxide must meet to contribute to the NPRI. It was also corrected to clarify the study's findings in relation to Ontario's air quality guidelines and the type of equipment loaned to ECCC by American researchers.

Keep reading

Air quality monitoring across the board in Canada seems so woefully inadequate. (would like to know where that is not he case) One stop gap solution at least for PM2.5 monitoring, caused by for example highly localized wood smoke pollution, is for communities (and frankly citizens if their communities won't adequately invest) is low cost "personal air monitors" eg. Purple Air (other brands exist, but see their website for a real time example of how it is working and the variety of readings within even one city neighbourhood). Simply put, we need a total reset from the old model of one air monitoring station for a huge population. Additionally, shouldn't we monitoring for a much larger array of pollutants? Shouldn't we be setting lower thresholds for "acceptable" levels? Shouldn't we be educating people much more vigorously about all these things, and more pertaining to air quality. Does the average person even know what PM2.5 is, or how many people die a year from air pollution in Canada? Or the diseases that can be triggered and or exacerbated? Time to step up with a major overhaul of all things pertaining to air quality from top to bottom, from A to Z.

Our measurement of toxics in water is terrible, as well.
The main problems with setting levels, that I can see, are that there's no defensible way ot setting levels for substances that are toxic at any level, and what is for some people not particularly toxic can for others be ruinous.
And levels for toxic substances seem often to be based on what's used, rather than the other way around.
In addition, for licensing of new chemicals, Health Canada relies on studies produced by industry, with all the issues of bias that includes.