Nearly half of the tap water in the US is contaminated with ‘forever chemicals,’ government study finds

news article

peer-reviewed paper

A peer-reviewed paper entitled “Per-and polyfluoroalkyl substances (PFAS) in United States tap water: Comparison of underserved private-well and public-supply exposures and associated health implications,” by Smalling et al describes a study in which over 700 locations across the  United States, including Puerto Rico and the Virgin Island territories, were sampled for a variety of Per- and Polyfluorinated Substances (PFAS) (Inserted Figure 1) (1). More importantly, this study addresses a gap in data and aims to understand point-of-source exposures by sampling from public supply tap-water and privately owned wells.

PFAS or “forever chemicals” have become a point of interest within the scientific community, particularly for the role they play as drinking water contaminants, due to their long lifetimes, toxicities, and ubiquitous use throughout the industry in consumer products manufacturing (2).  PFAS refers to a vast array of compounds that all share the common feature of containing a carbon-fluorine bond and have been linked to numerous health disorders and several types of cancer (1). However, there are no federal PFAS regulations in drinking water (1). In fact, the only benchmark that this paper mentions is a 2023 EPA maximum contaminant level (MCL) proposal of 4 ng/L for two common PFAS: perfluorooctanoic acid (PFOA) and perfluorooctane sulfonate (PFOS) (1). Clearly, a baseline level for a variety of PFAS across different locations and timeframes will provide an incredibly useful metric when government regulations are eventually put in place. 

The Smalling et al study amassed a suite of data collected from 2016 to 2023, consisting of 716 samples in total. All of the samples were collected by volunteers in plastic tubes and were shipped on ice to one of three research facilities to be tested, depending on the year in which they were collected. The time of sampling varied by year, month, and time of day. A total of 26 samples were measured by the EPA National Exposure Research Laboratory (NERL) in 2016, 82 samples were measured by the Colorado School of Mines (CSM) from 2017-2018, and the remaining 608 samples were measured by the USGS National Water Quality Laboratory (NWQL) from 2019-2022. All three laboratories used slightly different quantification methods which targeted specific PFAS and yielded varying detection limits. The EPA NERL method targeted 10 PFAS, the CSM methods targeted 28 and 44 PFAS, and the USGS method targeted 34 total PFAS. The authors state that any variation between the methods was accounted for during their statistical analysis. It is important to note that the EPA proposed MCL for PFOA and PFOS of 4 ng/L is below the reporting limit for over half of these methods. The CSM, NWQL and EPA NERL reporting limits for PFOA and PFOS were 1.3 ng/L for both, 5 ng/L for both, and 2.0 ng/L and 7.4 ng/L respectively. This suggests that any conclusions related to proposed governmental controls are likely conservative, that any PFOA or PFOS detection is significant, and that PFAS concentrations may be underreported due to detection limitations.

The main takeaways of the study involved analysis of the PFAS concentrations across all samples, a comparison of detected PFAS concentrations to EPA proposed MCLs, estimated exposure toxicities, and correlations between PFAS concentrations and their potential sources. The authors found that of the 44 PFAS that were tested for, 17 were observed at least once and of those 17, all but one were measured in over 600 samples. Of all the tap water samples, 30% contained at least one PFAS and 20% of all private-well samples contained at least one PFAS. Within a sample for which a PFAS was detected, the total number of PFAS detected ranged from 1-9 and total PFAS concentrations ranged from 0.348 to 319 ng/L. In short, there is nearly a 50% chance that any water sample will contain at least one PFAS regardless if it is a tap water or private well source. 

As previously mentioned, it is important to take into account the fact that over half of the reporting limits for the methods used to detect PFOA and PFOS were below the EPA proposed MCL, and the remaining reporting limits were within the same order of magnitude. Any PFAS detection is therefore meaningful. When measured PFOA and PFOS concentrations were compared to the 4 ng/L EPA proposed MCL, they found that PFOA and PFOS levels were greater than 4 ng/L in 6.7% and 4.2% of all of the tap water samples respectively, and greater than 4 ng/L  in 48% and 70% of samples in which PFAS was detected. It should be noted that the language in the paper at this point makes it difficult to understand the same trend for private well samples. However, the authors generally concluded that any sample containing PFOA or PFOS would likely exceed the EPA proposed MCL for both tap water and private well sources.

The authors used two screening-level assessments to estimate exposure effects of the PFAS. 1) Sample specific cumulative exposure activity ratios (ΣEAR) which is essentially a ratio of the contaminant concentration and the low effect or no effect level of a dose, the lower the number, the better. 2) Sum of toxicity quotients (ΣTQ), or the ratio of the contaminant concentration and a health based benchmark of the compound, for 15 individual PFAS. The authors note that there are inherent limitations to this approach due to the availability of benchmarks for all PFAS and the difficulty in comparing the two approaches. The study reports that none of the samples had ΣEAR > 1, eliminating any probable risks due to long term exposure. However, there were samples for which the ΣEAR exceeded the precautionary screening level of 0.001 (yellow dashed line on inserted Figure 4). Conversely, 124 samples exceeded the benchmark value (ΣTQ > 1) indicated by the solid red line on inserted Figure 4, suggesting that risks due to long term exposure are relevant. With two separate approaches and contradictory results, it seems the best takeaway here is that we can not rule out the potential of long term health effects due to PFAS in drinking water from either tap water or private wells.

Finally, the authors sought to understand the relationship between PFAS concentrations and potential PFAS sources. They found that cumulative PFAS concentrations increased in locations surrounded by developed land and decreased as sample locations became further from the identified possible sources while individual PFAS concentrations would sometimes decrease with increased development suggesting that the type of PFAS source is more important than total potential PFAS source. The authors did find one relationship for a specific PFAS, PFBS, that showed a positive correlation with “open development” e.g. “large-lot single-family housing units, parks, golf courses, and vegetation planted in developed settings for recreation, erosion control, or aesthetic purposes.” In the conclusion the authors also mention that “drinking-water exposures may be more common in the Great Plains, Great Lakes, Eastern Seaboard, and Central/Southern California regions.”

The main takeaways that I would include when communicating the findings of this paper to a general audience would be that 1) PFAS is an umbrella term for thousands of compounds that are widely used and found within common household items, 2) There is approximately a 45% chance that at least one type of PFAS will be found in your drinking water, in both tap and well water, and there is a chance that this number could be higher in reality because of limitations to detection methods 3) The concentration of PFAS in your drinking water has the potential to cause health effects with long term exposure, 4) You can generally expect to see more PFAS in your drinking water if you are living on developed land and are close to any “open development” and PFAS water contaminants are more common in the Great Plains, Great Lakes, Eastern Seaboard, and Central/Southern California regions.

Overall, the CNN news article entitled “Nearly half of the tap water in the US is contaminated with ‘forever chemicals,’ government study finds” by Jen Christensen, did a fine job communicating the science of the Smalling et al study. The author describes PFAS as “a family of ubiquitous synthetic chemicals that linger in the environment and the human body,” and includes that exposure to PFAS has been “linked to problems like cancer, obesity, thyroid disease, high cholesterol, decreased fertility, liver damage and hormone suppression,” (3). The article correctly states that “at least one PFAS chemical would be detected in 45% of US drinking water samples,” and also notes that both tap water and private wells were tested and found to contain similar PFAS concentrations (3). They also mention the finding from the Smalling et al conclusion regarding the areas of the United States in which PFAS drinking water contamination is more likely. 

The article does not mention any other statistics, which I agree with because the toxicological study was relatively inconclusive. They also fail to mention the limitations in the PFAS detection methods and the conservative nature of the 45% statistic. A professor of pharmacology and toxicology, Dr. Jamie DeWitt, is quoted to say that “[people should not] necessarily need to be scared,” however, the author then concludes with a dissimilar quote from a professor of chemistry and biochemistry and PFAS researcher saying, “[PFAS is] really insidious this poison… we are going to have to get inventive on how to filter it out for all of our days,” (3). All this combined with the albeit helpful suggestions from Dr. DeWitt about household solutions for filtering out PFAS from drinking water, hints at an attempt to soften the PFAS blow and make the news of toxic contaminants in everyone’s drinking water, including well water, a little easier to swallow. For these reasons, I would give the CNN article an 8/10. The science is correctly communicated but the implications of the results are lacking in severity.

Sources:

(1) Smalling, K. L.; Romanok, K. M.; Bradley, P. M.; Morriss, M. C.; Gray, J. L.; Kanagy, L. K.; Gordon, S. E.; Williams, B. M.; Breitmeyer, S. E.; Jones, D. K.; DeCicco, L. A.; Eagles-Smith, C. A.; Wagner, T. Per- and Polyfluoroalkyl Substances (PFAS) in United States Tapwater: Comparison of Underserved Private-Well and Public-Supply Exposures and Associated Health Implications. Environment International 2023, 178, 108033. https://doi.org/10.1016/j.envint.2023.108033.

(2) Per- and Polyfluoroalkyl Substances (PFAS). U.S. Food and Drug Administration. https://www.fda.gov/food/environmental-contaminants-food/and-polyfluoroalkyl-substances-pfas#:~:text=What%20Are%20PFAS%3F,%2C%20and%20fire%2Dfighting%20foams.

(3) Christensen, J. Nearly Half of the Tap Water in the US Is Contaminated with ‘Forever Chemicals,’ Government Study Finds. CNN Health. July 6, 2023. https://www.cnn.com/2023/07/05/health/pfas-nearly-half-us-tap-water-wellness/index.html.