EPA’s Health Advisories
In August, I wrote Pulling Back the Curtain on EPA’s PFAS Health Advisories, discussing the EPA’s updated health advisories of 0.004 parts per trillion (ppt) for PFOA and 0.02 ppt for PFOS. Health advisories are guidance, not regulations, on acceptable drinking water levels. These levels are thousands of times lower than EPA’s 2016 Health Advisories of 70 ppt for both chemicals - so low that they are not even measurable in drinking water and are based on a flawed scientific methodology.
Several other countries and agencies have recently proposed safe levels for PFAS in drinking water much higher than the EPA’s.
- Last week Canada proposed guidelines for drinking water quality of 200 ppt for PFOA and 600 ppt for PFOS, 50,000 and 300,000 times greater than the EPA’s health advisories! These guidelines are measurable, practical, and based on animal, not the human, studies that EPA used as the basis for their health advisories. Canada’s goal is to eventually replace these guidelines with one guideline of 30 ppt for the sum of total PFAS detected in drinking water.
- The World Health Organization (WHO) proposed drinking water values of 100 ppt for both PFOA and PFOS and 500 ppt for total PFAS in September 2022. These values are 25,000 and 5,000 times greater than EPA’s health advisories. The WHO reviewed the health and exposure data in detail and concluded that there were inconsistencies and uncertainties between the health values and proposed “practical” drinking water values.
Some states adopt EPA’s guidance as rules, such as Alaska, Colorado, Connecticut, Delaware, New Mexico, and Ohio which have adopted EPA’s 2016 Health Advisories of 70 ppt for their limits for PFOA and PFOS.
Safe Drinking Water
The EPA currently does not have drinking water regulations for PFOA or PFOS. EPA had scheduled the release of its proposed regulations for December 2022, but this has been rescheduled for March 2023.
Under the Safe Drinking Water Act, when EPA sets a regulation for a contaminant in drinking water, it first must establish a health goal based on risk to sensitive populations - the Maximum Contaminant Level Goal (MCLG). It then sets a legal limit, the Maximum Contaminant Level (MCL), as close to the MCLG as feasible, taking costs, analytical methods, and treatment technologies into account. EPA must also do a Regulatory Impact Analysis (RIA), showing that the benefits exceed the costs.
For most chemicals, EPA sets the MCLG at the same level as the health advisory (if one is available). For PFOA and PFAS, the EPA could set the MCLG at the health advisory level and the MCL at the detection limit (4 ppt). However, treatment costs would be enormous at this level, and just as importantly, the benefits would be practically nonexistent.
In estimating the RIA, the EPA uses models of how many deaths or illnesses will be avoided with the proposed regulation. The type of illness avoided is based on the critical endpoint used in setting their health goal (the MCLG). For PFOA and PFAS, that critical endpoint is a decrease in antibody levels after immunizations for diphtheria and tetanus. The study (carried out in the Faroe Islands, a group of islands belonging to Denmark) did not show that PFAS caused the decrease in antibody levels or any increase in diphtheria or tetanus on the islands. The study suggested an association between PFAS and decreased antibody levels. Denmark rarely has a case of either disease; the US averaged less than 1 case of diphtheria per year over the last 40 years, and tetanus is also very uncommon.
If the EPA sets the MCLG based on its revised Health Advisories, it will have to conclude that there are zero benefits, an unlikely conclusion! The EPA’s other option is to set the MCLG at a higher, more realistic value, but how will they explain how it is a value based solely on health?
PFAS as Hazardous Substances
In April, I wrote PFAS: Fear and Misinformation Runs Wild, discussing the EPA’s intent to classify PFAS as hazardous substances under the “Superfund Law .”[1] In early 2021, the EPA issued an Advanced Notice of Proposed Rulemaking (ANPRM) stating they have the authority under that law to respond to releases into the environment of any hazardous substance “which may present an imminent and substantial danger to the public health and welfare.” [emphasis added]
The EPA has never before attempted to list substances as hazardous, not listed under other environmental laws. The EPA proposed doing this for PFAS under its interpretation of the phrase “may present,” indicating that Congress does not require certainty that the substance presents a substantial danger or proof of actual harm. Once a substance is designated as hazardous, it has widespread repercussions across the US economy
In September 2022, the EPA proposed designating two PFAs, perfluorooctanoic acid (PFOA) and perfluoroctanesulfonic acid (PFOS), as hazardous substances. It would impact and have significant economic effects on 21 sectors, including firefighting foam, medical devices, wastewater treatment, and even car washes. [2]
When the Office of Management and Budget (OMB) initially considered the EPA’s proposed rule, it classified it as “other significant,” meaning that it was predicted to have costs or benefits of less than $100 million annually. After several comments, including one from the Chamber of Commerce that provided its estimate that the rule would have a cost impact of over $700 million annually, OMB changed its initial decision and said that EPA would have to carry out an RIA. The RIA would need to justify and explain that the designation is the least burdensome and most cost-effective way to achieve EPA’s goals. An RIA has yet to be released.
Both Canada and the WHO have proposed drinking water guidelines for PFAS at significantly higher levels, yet in their view, still safe than our EPA. They follow the lead of other countries, including Australia and the UK, that have set levels at 100 ppt and above. Will the EPA see the flashing red light and understand the significant economic impacts without a corresponding health benefit in its designation of PFAS as hazardous substances?
[1] Formally, the Comprehensive Environmental, Compensation, and Liability Act (CERCLA)
[2] A complete list would include, Aviation operators, Carpet manufacturers, Car washes, Chemical manufacturing, Chrome electroplating, anodizing, and etching services, Coatings, paints, and varnish manufacturers, Firefighting foam manufacturers, Landfills, Medical devices, Municipal fire departments and firefighting training centers, Paper mills, Pesticides and Insecticides, Petroleum and coal product manufacturing, Petroleum refineries, and terminals, Photographic film manufacturing, Polish, wax, and cleaning product manufacturers, Polymer manufacturers, Printing facilities where inks are used in photolithography, Textile mills (textiles and upholstery, Waste management and remediation services, Wastewater treatment plants.
NOTE TO READERS: I would like to issue the following clarification based on information I received from Health Canada after publication. On February 11, Health Canada proposed an objective of 30 ng/L for the sum of total per- and polyfluoroalkyl substances (PFAS) detected in drinking water (about 18 compounds). This value is 1,500 (PFOS) and 7,500 (PFOA) times higher than the EPA Health Advisories. Both animal and human studies were considered to understand the health hazard, but the number is based on treatment and analytical achievability, not health, and therefore cannot be directly compared to EPA’s Health Advisories. The guidelines of 200 ppt for PFOA and 600 ppt for PFOS that I quoted in my article are not new values, but were finalized in 2018. Canada’s goal is to eventually update these health-based guidelines and it believes that “due to the many uncertainties about health effects, levels of PFAS should be kept as low as reasonably achievable.” I apologize for my unintentional mischaracterization of Health Canada’s latest guidelines on PFAS. This was based on my interpretation of the information on Health Canada’s website which turned out not to be accurate.