Drinking water contaminated with Pfas chemicals probably increases the risk of infant mortality and other harm to newborns, a new peer-reviewed study of 11,000 births in New Hampshire finds.

The first-of-its-kind University of Arizona research found drinking well water down gradient from a Pfas-contaminated site was tied to an increase in infant mortality of 191%, pre-term birth of 20%, and low-weight birth of 43%.

It was also tied to an increase in extremely premature birth and extremely low-weight birth by 168% and 180%, respectively.

The findings caught authors by surprise, said Derek Lemoine, a study co-author and economics professor at the University of Arizona who focuses on environmental policymaking and pricing climate risks.

“I don’t know if we expected to find effects this big and this detectable, especially given that there isn’t that much infant mortality, and there aren’t that many extremely low weight or pre-term births,” Lemoine said. “But it was there in the data.”

The study also weighed the cost of societal harms in drinking contaminated water against up-front cleanup costs, and found it to be much cheaper to address Pfas water pollution.

Extrapolating the findings to the entire US population, the authors estimate a nearly $8bn negative annual economic impact just in increased healthcare costs and lost productivity. The cost of complying with current regulations for removing Pfas in drinking water is estimated at about $3.8bn.

“We are trying to put numbers on this and that’s important because when you want to clean up and regulate Pfas, there’s a real cost to it,” Lemoine said.

Pfas are a class of at least 16,000 compounds often used to help products resist water, stains and heat. They are called “forever chemicals” because they do not naturally break down and accumulate in the environment, and they are linked to serious health problems such as cancer, kidney disease, liver problems, immune disorders and birth defects.

Pfas are widely used across the economy, and industrial sites that utilize them in high volume often pollute groundwater. Military bases and airports are among major sources of Pfas pollution because the chemicals are used in firefighting foam. The federal government estimated that about 95 million people across the country drink contaminated water from public or private wells.

Previous research has raised concern about the impact of Pfas exposure on fetuses and newborns.

Among those are toxicological studies in which researchers examine the chemicals’ impact on lab animals, but that leaves some question about whether humans experience the same harms, Lemoine said.

Other studies are correlative and look at the levels of Pfas in umbilical cord blood or in newborns in relation to levels of disease. Lemoine said those findings are not always conclusive, in part because many variables can contribute to reproductive harm.

The new natural study is unique because it gets close to “isolating the effect of the Pfas itself, and not anything around it”, Lemoine said.

Researchers achieved this by identifying 41 New Hampshire sites contaminated with Pfoa and Pfos, two common Pfas compounds, then using topography data to determine groundwater flow direction. The authors then examined reproductive outcomes among residents down gradient from the sites.

Researchers chose New Hampshire because it is the only state where Pfas and reproductive data is available, Lemoine said. Well locations are confidential, so mothers were unaware of whether their water source was down gradient from a Pfas-contaminated site. That created a randomization that allows for causal inference, the authors noted.

The study’s methodology is rigorous and unique, and underscores “that Pfas is no joke, and is toxic at very low concentrations”, said Sydney Evans, a senior science analyst with the Environmental Working Group non-profit. The group studies Pfas exposures and advocates for tighter regulations.

The study is in part effective because mothers did not know whether they were exposed, which created the randomization, Evans said, but she noted that the state has the information. The findings raise questions about whether the state should be doing a similar analysis and alerting mothers who are at risk, Evans said.

Lemoine said the study had some limitations, including that authors don’t know the mothers’ exact exposure levels to Pfas, nor does the research account for other contaminants that may be in the water. But he added that the findings still give a strong picture of the chemicals’ effects.

Granular activated carbon or reverse osmosis systems can be used by water treatment plants and consumers at home to remove many kinds of Pfas, and those systems also remove other contaminants.

The Biden administration last year put in place limits in drinking water for six types of Pfas, and gave water utilities several years to install systems.

The Trump administration is moving to undo the limits for some compounds. That would probably cost the public more in the long run. Utility customers pay the cost of removing Pfas, but the public “also pays the cost of drinking contaminated water, which is bigger”, Lemoine said.