Key takeaways:

• PFAS, known as “forever chemicals,” are manufactured chemicals found in many food and plastic products.
• Areas with higher PFAS levels also had higher rates of Alzheimer’s mortality.

Parts of southeastern Pennsylvania with higher rates of Alzheimer’s disease mortality had levels of per- and polyfluoroalkyl substances that exceed federal minimums, according to data from Temple University researchers.

Findings from the ecological study were presented at the Alzheimer’s Association International Conference 2026. Erin Kulick, PhD, MPH, assistant professor of epidemiology and biostatistics at Temple, said the Alzheimer’s disease mortality rate in the study area is approximately 13% higher than the overall U.S. rate.

Data derived from Kulick E, et al. Spatial distribution of PFAS-contaminated public drinking water supplies and Alzheimer’s disease mortality. Presented at: Alzheimer’s Association International Conference; July 12-15, 2026; London.

“Regardless of sex, we saw overall that the highest exposure areas had consistently higher mortality rates,” Kulick told attendees during a presentation.

Per- and polyfluoroalkyl substances (PFAS), also known as “forever chemicals,” are manufactured chemicals that are used in the creation of a variety of materials, including many food and plastic products such as waterproofing materials, nonstick cookware and stain-resistant fabrics, according to Kulick. For people, the primary exposure to PFAS is through contaminated drinking water.

When consumed by a person, PFAS can accumulate in brain tissue and disrupt normal neurological function. Based on early experimental evidence in animals, PFAS can promote neuroinflammation, oxidative stress, mitochondrial dysfunction and the accumulation of amyloid beta and tau proteins, all of which Kulick said can contribute to Alzheimer’s disease and related dementias (ADRD).

“Human studies have started to show associations between PFAS exposure and ADRD, but the evidence is still pretty inconsistent, mostly because there’s not a lot of it,” Kulick said.

Kulick and colleagues sought to map the spatial distribution and clustering of PFAS-contaminated drinking water with the distribution of Alzheimer’s mortality.

To do this, they examined links between PFAS exposures and Alzheimer’s mortality using vital statistics from 408 municipalities across 12 counties in southeastern Pennsylvania. They also analyzed death certificate data in the area from 2000 through 2020.

Exposures included the six PFAS chemicals that are most prevalent in the water supply in the study area, which included 134 public water systems. The six chemicals were PFBS, PFHpA, PFHxS, PFNA, PFOS and PFOA.

Water samples were categorized into three groups: those with no PFAS among any contaminants; those with some level of PFAS but that were under the minimum reporting level (MRL) for any contaminant found; and those with PFAS levels above the MRL for any contaminant.

In some parts of Pennsylvania, PFAS levels reached as high as 1,290 parts per trillion (ppt), well over the federal MRL of 4 ppt, Kulick said.

In total, 45% of the municipalities analyzed had PFAS levels above the MRL, Kulick said.

The annual Alzheimer’s disease mortality rate for the study area was 26 per 100,000 people. One part of the study area had a mortality rate of 372 per 100,000 people, Kulick said.

“When I first looked at this rate of 372 people per 100,000, I thought it was fake, but it really is real, and it’s clustering right in the highest PFAS area, essentially where we expected,” Kulick said.

Half of the municipalities that fell in the upper tertile in terms of mortality rate had PFAS levels above the MRL, Kulick said. Among municipalities in the second and third tertiles, 47% and 39% of them had levels above the MRL, respectively.

Using Global Moran’s I to gauge the degree and significance of spatial autocorrelation, Kulick and colleagues found significant high-high positive clusters, with a positive spatial correlation of 7.43 (P = .001).

High-high positive clusters were found near the cities of Philadelphia, Reading, Lancaster and Allentown, in areas Kulick said typically had military bases or airports.

Moving forward, Kulick said it will be important to conduct individual-level analyses of PFAS exposure and Alzheimer’s disease mortality in the high-risk areas identified in the study.

“We are almost done modeling individual-level exposure for multiple PFAS types over time, going back into the ’70s, obtaining people’s residential histories and matching them with their PFAS exposure, and then doing more in-depth spatial-temporal analyses incorporating these exposures,” Kulick told attendees.

For more information:

Erin Kulick, PhD, MPH can be reached at neurology@healio.com.