A smoky haze obstructs the view of the San Francisco skyline on Aug. 24. (Justin Sullivan/Getty Images)

Low air quality, even at pollution levels well below current Environmental Protection Agency thresholds, is associated with increased risk of dementia and Alzheimer's disease in later life, according to a new working paper published by the National Bureau of Economic Research.

Researchers Kelly C. Bishop, Nicolai V. Kuminoff and Jonathan D. Ketcham of Arizona State University cross-referenced more than a decade of Medicare records for 6.9 million older adults with EPA air-quality data to track how exposure to air pollution correlated with rates of dementia.

They found that levels of fine particulate pollution known as PM2.5 were closely associated with higher rates of dementia. “A 1 microgram-per-cubic-meter [μg/m³] increase in average decadal exposure [of PM2.5] increases the probability of receiving a dementia diagnosis by 1.3 percentage points.”

To put those numbers in context, that increase in dementia odds is “equivalent to the elevated risk associated with a female aging from 74 to 77,” Bishop and her colleagues write. There is a nearly 17 μg/m³ difference between typical PM2.5 levels in the least-polluted county in the United States and those in the most polluted.

The authors find that the harmful effects of the pollutants begin occurring well before exposure levels deemed actionable by the EPA, and they recommend tightening federal restrictions to save lives and dollars. In recent months, however, the Trump administration has made rule changes that would put more of the pollutants in the atmosphere.

Researchers and policymakers have only recently begun grappling with the effects of PM2.5 — the Environmental Protection Agency didn't have a separate regulatory standard for it until 1997. PM2.5 particles are tiny — about 1/30th the width of a human hair. Its small size “allows it to remain airborne for long periods, to penetrate buildings, to be inhaled easily, and to reach and accumulate within brain tissue,” Bishop and her colleagues write.

PM2.5 has been linked to impaired physical and cognitive ability in both blue-collar and white-collar workers. It's been linked to respiratory system damage, emergency room visits and death. It's been shown to be particularly harmful to the elderly. Post-mortem examinations have found that it accumulates in the human brain. “People living in more polluted areas, such as near roadways, for long periods tend to have elevated concentrations of PM2.5 in their brains, smaller brain volume, and higher rates of brain infarcts or areas of necrosis,” Bishop and her colleagues write.

In short: it's nasty stuff. In 1997 the EPA set a standard for average annual PM2.5 exposure of 15 μg/m³, and reduced that threshold to 12 μg/m³ in 2012. The good news: That regulation appears to be working. EPA data show that nationwide, average annual PM2.5 exposure fell from 13.5 μg/m³ in 2000 to 8.0 μg/m³ in 2017.

Average annual PM2.5 exposure in the United States.

But as the map below shows, in a number of places in the United States, particularly parts of the Rust Belt and Southern California, annual PM2.5 levels are well above that number. In 2014, the latest year for which the Centers for Disease Control and Prevention has countrywide data, the highest PM2.5 level was observed in Kern County, Calif., with an annual average of 19.7 μg/m³, or 64 percent higher than the current EPA standard. The county is home to 882,000 people, and its economy is linked to agriculture and petroleum extraction.

Map of annual PM2.5 exposure.

Researchers have suspected for some time that air pollution, particularly PM2.5, is a driver of dementia and Alzheimer's disease in the United States. But previous research has been limited to small sample sizes, or to people who live near major roadways that give off high levels of pollution. The primary advantage of Bishop and her colleagues' study is that it's huge and comprehensive, involving 6.9 million Medicare enrollees living all over the United States. They also tracked the enrollees over time, thus eliminating the confounding effects of, say, moving from a high-pollution area to a low one, or vice-versa.

"These exposure histories are the most comprehensive data ever developed to study how air pollution affects cognition,” they write.

To ensure that they were isolating, to the best extent possible, the effects of PM2.5, they spent a lot of time controlling for other factors known to affect health and dementia risk: age, race, gender, other medical conditions, overall medical expenditures and neighborhood economics and demographics.

In the end, they were able to distill their headline finding: that “a 1 microgram-per-cubic-meter [μg/m³] increase in average decadal exposure [of PM2.5] increases the probability of receiving a dementia diagnosis by 1.3 percentage points.” That's a stunning figure, particularly given that ambient PM2.5 levels vary by much greater than that on a county-by-county basis.

But they also took their analysis a few steps further by attempting to calculate a return-on-investment for existing PM2.5 regulation. In the mid-2000s, for instance, the EPA flagged 224 counties for failing to meet the 1997 PM2.5 standards. These counties were targeted for increased regulation, and the data collected by Bishop and her colleagues shows that over the following years those counties subsequently saw greater reductions in PM2.5 than counties that were already meeting the rule. The heightened regulatory scrutiny worked, in other words.

Bishop and her colleagues calculated the number of dementia cases avoided in these counties because of those regulatory efforts, and assigned dollar values (based on standard actuarial calculations) to the quality-adjusted life years gained in those counties as a result of the avoided dementia cases. The overall benefit, using middle-of-the-road estimates, worked out to $163 billion in those 224 counties alone.

They also found that stricter regulation, via a lower EPA standard for annual PM2.5 levels, would have saved even more lives and more dollars. In other words, to maximize public health benefits policymakers should set the threshold even lower.

The current administration, however, is taking the opposite approach. Earlier this year President Trump announced he was directing the EPA to give localities “more flexibility” in their compliance with national air quality standards.

More recently, in August the EPA announced the rollback of pollution restrictions on coal-burning power plants, effectively allowing them to put more PM2.5 in the atmosphere. The EPA's own analysis estimated that the rule change would lead to as many as 1,400 additional premature deaths from air pollution annually.