Is PM2.5 killing people, or not?
It seems like a basic scientific question.
Yet, as a recently published study shows, the science is anything but settled.
Despite significant questions in this new paper, which echo a growing number of other studies published in recent years, there is almost zero discussion in the scientific and regulatory communities of whether PM2.5 is truly a killer.
The silence is oddly deafening.
Yes, it involves complex scientific information. But you need to pay attention to the details because it affects you in very direct ways.
Regulations mandating reduction of PM2.5 (tiny particulate matter in soot, dust and exhaust) are what keep people from lighting fires on cold winter nights here in the San Joaquin Valley, which is nowhere near compliance with national standards and has some of the most restrictive rules on PM2.5.
More important than cold, empty fireplaces, PM2.5 regs cost us a lot of money, every single day.
That’s because all long-haul trucks, tractors, bulldozers, etc., in California made before 2014 are required to be retrofitted with expensive filters and/or replaced, driving up costs for consumers on just about everything we use, eat or wear.
Those regulations are justified on a public health basis by the claim that PM2.5 causes thousands of “premature deaths” every year in California and nationwide.
That idea was born in 1995 after a major study authored by C. Arden Pope found an association between exposure to PM2.5 and premature deaths.
INTO THE WEEDS
Here’s what “association” means in this context.
This was a very large epidemiological study that looked at data on more than 500,000 Americans enrolled by the American Cancer Society in 1982 and followed for death through 1989.
Pope combined each subject’s lifestyle data with their estimated PM2.5 exposure level based on their address in 1982.
He then did a statistical analysis to estimate the contribution of PM2.5 to their deaths through 1989.
Pope compared that data, which included the subjects’ lifestyles, ages, address and deaths, to exposure levels of PM2.5 and created a statistical analysis of whether PM2.5 may have played a role in their deaths.
The role PM2.5 may have played is labeled as the “relative risk” in this and other studies.
In that seminal Pope study, the relative risk was found to be 1.07, meaning for every 10-unit increase in PM2.5 exposure, you have a 7 percent greater risk of dying early.
The 1995 American Cancer Society study by C. Arden Pope created a huge outcry at the time with many scientists questioning its conclusions.
That didn’t stop the Environmental Protection Agency from creating a national ambient air quality standard for PM2.5 in 1997, based primarily on Pope’s work.
The annual standard back then was set at 15 micrograms per cubic meter. The current annual standard is 12 micrograms per cubic meter.
The San Joaquin Valley has yet to meet the 1997 standard and has virtually no hope of hitting the new standard by the 2025 deadline, by the way.
Between then and now, a number of studies have come out showing no association between PM2.5 and premature deaths.
These other studies are not based on the “gold standard” American Cancer Society data, however.
And they are routinely discounted by the EPA and the California Air Resources Board, or CARB, when the latest scientific information is evaluated for proposed new regulations.
The latest study, published in the journal Dose-Response last month by James Enstrom, did use the American Cancer Society data.
For anyone who’s followed air pollution science for any length of time, the name “Enstrom” will likely elicit a response.
Enstrom has been arguing the merits of Pope’s and other researchers’ PM2.5 findings for the better part of 15 years.
It was Enstrom who discovered a CARB researcher who helped propagate those truck rules I mentioned earlier had lied about his Ph.D. credentials and CARB staffers knew about the deception.
It wasn’t long after that when UCLA attempted to fire Enstrom from his longtime research position with the Environmental Health Sciences Department.
In a termination letter, he was informed “your research is not aligned with the academic mission of the Department.”
As part of the 2012 settlement, UCLA rescinded Enstrom’s termination, paid him $140,000 and allowed him access to school resources, which he still has.
What I’ve learned over the years of covering Enstrom and this topic is that while Enstrom’s views may be unpopular, no one has been able to throw shade on his actual work.
I find that telling.
In his latest paper, Enstrom uses the American Cancer Society human data — which he won’t say how he obtained (which is bound to be a big deal, see sidebox at right) — to recreate Pope’s 1995 study.
He shows that you can get a relative risk similar to Pope’s or you can get no relative risk.
It all depends on which PM2.5 exposure data you use and the best available PM2.5 data yields no relative risk.
Of particular interest, Enstrom found when he separated the Ohio Valley area from the rest of the country, there’s no significant effect in either area no matter which PM2.5 data he used.
That kind of sensitivity brings the whole idea of PM2.5 as a killer into question.
SILENCE ISN’T AN ANSWER
Since Enstrom’s paper directly calls out Pope and his findings, I called him up.
Pope would only say that he knows Enstrom to be a very bright scientist, that he takes the questions in Enstrom’s paper seriously and would be reading it closely and writing up a response.
“Hey, this is what science is about,” he said. “We need people to be skeptical and pick apart our work. It forces us to be more rigorous.”
That’s a refreshing attitude and one I hope is catching.
Unfortunately, up to now researchers who’ve been vocal in questioning PM2.5’s effects have been essentially blackballed, as I’ve been told.
Regulators ignore their studies. They can’t get funding from EPA, CARB or the Health Effects Institute, the main funding sources for air pollution research (itself a conflict of interest). And essentially all have been turned down by the American Cancer Society for access to that human data used by Pope.
Even so, studies have been coming out showing virtually no effect from PM2.5, particularly in the western United States.
Even one study in 2016 by George Thurston, which purported to show an effect, actually has a relative risk of dying early from PM2.5 exposure of 1.02 for California. When you add the margin of error to that figure, it’s basically a nonrisk.
DON’T FORGET THE LIVING
“Air pollution does kill people,” another longtime researcher, Fred Lipfert, told me, citing the terrible London smog that killed hundreds of people in 1952. “But we don’t know who it will kill or why.”
And these large epidemiological studies don’t get at those questions because there’s no way to say what an individual’s actual PM2.5 exposure is over time.
Aside from that, he said, PM2.5 isn’t a pollutant.
“It’s a regulatory construct,” he said. “It includes everything of a certain size, which includes all kinds of stuff. Some toxic, some not.”
Science should focus on what people are actually exposed to instead of these mass studies where exposure levels are so variable.
“I think Enstrom’s paper cracked the door on that issue,” he said. “We need more science on this issue and perhaps a recognition that regulation is not the end all, be all.”
Because, he said, while air pollution studies focus on deaths, “It’s the living population that pays the bill for all these regulations.”
As one of those people, I’d like to at least see an open, honest conversation about this among scientists and regulators.