Guest post by S. Stanley Young and Warren Kindzierski
We were asked by Heritage Foundation to write two papers on the alleged link between climate change, predicted changes in air quality, and public health effects. Both papers are available on their website. One paper is a discussion of a climate change−fine particulate matter (PM2.5)−public health link (or lack thereof). The second paper is a discussion of how ozone fits (or rather how it fails to fit) into a climate change−ozone−asthma link. A summary of both papers is provided, and an open question is posed to WUWT readers below.
First and foremost… for both PM2.5 and ozone, ALL of the studies we refer to in our two papers are not founded on proven biological plausibility of these factors causing diseases or deaths. They are founded on an assumption of what may be a cause of disease or death—for example, PM2.5 or ozone.
PM2.5
Our first paper examined the two key public health endpoints that the U.S. Environmental Protection Agency (EPA) claims result from PM2.5 exposure—nonfatal heart attacks and premature deaths. For both endpoints, we show that health effect studies—observational epidemiology studies—cited by the EPA to support their claims do not take proper accounting of hidden biases, nor do they apply rigorous tests for reproducibility of these studies.
The most glaring biases to us, but hidden to most readers, are use of questionable (poor) research practices, multiple hypotheses testing, and irreproducibility (falseness) of the research claims. Our paper also points to numerous null association studies in scholarly literature showing that PM2.5 does not cause nonfatal heart attacks and premature death.
It is up to the interpretation of the scientific method to answer the PM2.5−nonfatal heart attack/premature death claims: If the methods are flawed, so is the evidence. Studies engaging in poor research practices should be treated as untrustworthy until proven otherwise. Studies that perform many statistical comparisons tend to produce more errors of false-positive associations in the absence of statistical corrections. Finally, PM2.5−heart attack/premature death association studies failed our statistical reproducibility tests using p-value plots.
Herein lies the problem with PM2.5−health effect studies relied upon by the EPA: They are non-randomized designs with no effective researcher controls. They allow infinite researcher flexibilities in designing, analyzing, interpretating, and reporting of their results. Their data sets are not available to others for independent reproducibility testing. These types of designs and lack of controls cannot address biases and confounding. Spurious risk statistics that resemble genuine effects can (and do) easily occur in these studies when, in fact, they are nothing more than artifacts of hidden biases and confounding.
With no reproducible causal effects and given persistent, hidden biases and confounding in PM2.5 health research, any purported link between PM2.5 and public health is entirely unsupported and should not be taken seriously.
Ozone
Our second paper shows that the EPA’s own data for emissions of ozone precursors and ozone levels recorded at air monitoring stations in the U.S. over the past several decades show very different trends (declines) to what climate change models are predicting in the future. For example:
- mean ozone levels from 132 monitoring sites over the 42-year period 1980–2022 have steadily decreased by 7 parts per billion (ppb) per decade
- total number of days reaching “Unhealthy for Sensitive Subgroups” or above for ozone based on the EPA Air Quality Index among 35 major cities over the 22-year period 2000–2022 have steadily decreased by 600 days per decade
Thus, any potential impacts theorized from climate change models on air quality and asthma should be treated with skepticism.
The ozone−asthma link
The EPA considers respiratory effects, including asthma, as the key health effect for ozone exposure. They use results from controlled human exposure studies (i.e., chamber studies) and animal toxicology studies to claim that ozone can cause lung function effects. They use results from observational epidemiology studies to claim that ozone levels in outdoor air are associated with asthma attacks. They use results from observational studies and animal toxicological studies to claim that ozone causes onset of asthma. Sounds complicated, but not really.
First off – chamber studies. We present numerous compelling arguments demonstrating that no adverse effects or differences in effects can be shown to exist between non-asthmatics AND asthmatics exposed to ozone in controlled chamber studies. These include confounding from awareness bias and measurement variability, and lack of external validity.
Secondly – observational epidemiology studies. Our findings can be succinctly summarized as follows… same methods as PM2.5, same flaws, no proof.
Thirdly – laboratory animal studies. The EPA has hypothesized several mechanisms for how ozone exposure may cause an asthma attack or onset of asthma. Of course, laboratory animals are not humans and suffer from numerous problems. The most important being that they do not spontaneously develop asthma, nor can they be tested in traditional ways that people are diagnosed for asthma or asthma symptoms.
Only by imaginative use of smoke and mirrors, perhaps with the aid of AI, could one conclude that evidence cited by the EPA from chamber, epidemiology, and animal studies support casual effects of asthma or asthma symptoms from ozone exposure.
Our question for WUWT readers
We pose an open question for WUWT readers which is part and parcel of the problems we encountered in our assessment of an alleged climate change−air quality−public health link.
The Reference Manual on Scientific Evidence is a document created for U.S. federal judges to help them understand and interpret difficult issues involving scientific testimony. The third edition of the Reference Manual, a 1000+ page document, was published in 2011 as a collaboration with the Federal Judicial Centre and the National Research Council (NRC) of the U.S. National Academies of Sciences, Engineering, and Medicine (NASEM).
NASEM is primarily comprised of established academics. Authors contributing to the 2011 Reference Manual included many such academics. Of 16 chapters in the 2011 Reference Manual, 14 had established academics as primary author or as co-authors. Of 30 authors or co-authors contributing overall, 22 were academics.
Junk science—spurious (false) or fraudulent scientific data, research, or analysis—has proliferated in peer-reviewed literature over the past several decades simultaneously as these ‘established’ academics built their careers. False or fraudulent evidence from junk science has been used to advance special interests and hidden government regulatory agendas on many topics important to society – climate change, environmental pollution, health impact, etc. As government funds much academic research, they should not be viewed as independent.
The U.S. Supreme Court recently overruled itself on a longstanding, controversial doctrine that gave regulatory agencies, e.g., the EPA, an unfair advantage in court – the so-called “Chevron doctrine.” This decision now places the obligation on U.S. Congress to legislate more explicitly and on the courts to interpret the law without defaulting to agency judgment, which in the case of the EPA is suspect—as we have established above.
Given this outcome, it is essential in our view that the Reference Manual be properly updated for judges (and lawyers), so they better understand junk science and how it has proliferated in academic research, government policy making, and the setting of regulations.
[Our question] How can updating the Reference Manual to address junk science best be accomplished given such a massive input from entrenched academics in the past?
Little has changed regarding the credibility of peer-reviewed literature since John Ioannidis published his 2005 paper Why Most Published Research Findings Are False, and NASEM published their two reports on irreproducibility of scientific results (one in 2016 and one in 2019), and the National Association of Scholars published their science irreproducibility report in 2018.
The last thing that should be allowed in updating the Reference Manual is for the makeup of authors to be overweight with established academics to figure out how to deal with the 800-pound junk science gorilla they helped create and/or allowed to flourish on their watch.
S. Stanley Young is CEO of CGStat in Raleigh, North Carolina and is Director of the National Association of Scholars’ Shifting Sands Project. Warren Kindzierski is a retired college professor (public health) in St Albert, Alberta and a contributor to the Shifting Sands Project.
Doing away with accepting L-NT (linear-no threshold) assumptions as to scaling risk is another bit of quasi religious doctrine. It is ultimately an Ad Ignorantium argument, from that no one really knows the risk, so of course we must assume a worst case, to the presumption that L-NT is true. Because we say so, and you are an apologist for polluters if you doubt it.
In several cases, where L-NT has been investigated, the model was not confirmed. But the State of California, for one, has legislated that model into Proposition 65 warnings on a large number of products.
Our World in Data has a ‘prevalence of asthma map (2021)’ that displays data for asthma in countries 1990 — 2021 (Data source: IHME, Global Burden of Disease).
There appears to be weak correlation between industrialisation and asthma prevalence, for instance in 1990 the most affected countries included Greenland Norway Sweden Australia and New Zealand.
Further the prevalence in almost every country pointed to has decreased, in some remarkably so, over the period 1990 — 2021.
The US is an outstanding exception and compared to say China there must be factors applying to US other than industrialisation air pollution and urbanisation.
The human body health is a confluence of many factors, air being just one.
Unless factors such as diet and exercise, and even, perhaps, mental health are included, a holistic assessment cannot be achieved.
No study should be used in a court, or for regulatory purposes, unless it has been truly independently replicated. The independence of the replication must be verified.
The law of unintended consequences occurs to me. The recent attribution of severe weather events, particularly rainfall to climate change may have another cause, if true. Have there been any studies to establish whether the reduction in air pollution has the effect of delaying the nucleation of water vapour. This would result in heavier showers and flash floods.
Perhaps any perceived temperature increase is due to increasing levels of water vapour – a greenhouse gas – as a result of the desire to ‘improve’ air quality? Millennia of burning fuels with particulate emissions may have kept us cool!
increased water vapour as a result of climate change or climate change as a result of increased water vapour? Chicken or egg?
The whole idea that PM2.5 causes significant negative health effects, independently of what the PM consists of and of any other pollutants mixed with it, is highly disputable. This idea is basically a doctrine being pushed by the UN’s WHO. In the UK, the government’s COMEAP (Committee on the Medical Effects of Air Pollution) in 2009 issued a report which simply parroted the WHO’s line, and even its recommendations. And yet, real world evidence from the Great Smog of London (1952), and subsequent smogs and hazes, tells us that PM is only dangerous when breathed in together with sulphur oxides. The entire “science” of air pollution toxicology needs a thorough, independent, evidence-based audit.
At the same time, one can purchase ozone generators for your basement, bedroom, kitchen, etc., for “air purification and odor elimination.” https://www.rabbitair.com/pages/not-acceptable-ozone-generators
The flawed PM 2.5 study was used because it gave EPA the result it wanted, free from scientific validity. In my years of reading ever tightening air and water regulations and their justifications, I found EPA selected the outcome first and then found “data” that supported that outcome.
If you look deeply you will observe they found models that supported their predetermined conclusions. Multiple times.
[Our question] How can updating the Reference Manual to address junk science best be accomplished given such a massive input from entrenched academics in the past?
In the present political climate, all academic authorities should be excluded, except perhaps as a reference source. What is needed is working, hands-on pathology, chemistry, biology, medical experts plus a significant representation of engineers (all fields), and applied physicists. A few expert statisticians would also prove their worth. It also might be worthwhile to have a practicing legal authority (judge, attorney, whatever) be involved only in the review such that clarity rather than interpretations is established.
Why engineers? Engineers are problem solvers. Analysis of alternatives is a key function. Engineers have to live with and deal with anything they get wrong and are motivated to not have that kind of result.
This cannot be a hobby or part time involvement. Nor can it be an environment with people working in isolation. They have to come together to work as a team.
Above all else, politics needs to be exorcised from the process.
My concern is about the upcoming explosion of large language model AI programs that will scour the data bases with all the unverified information listed as factual.
We are about to be inundated with junk science propaganda with climate data just being one of the corrupted data sets that will be foisted upon the public.
If anyone comes up with a plausible and practical means of countering this, I have investment money to commit.
What’s really scary is the leftist’s desire to automate all research results including medical protocols. I read somewhere that over 80% of research has some level of data manipulation.
I think it all depends on what the programmers determine to be valid data sets. The propensity for leftist influence is almost guarantied.
Leftist does not get exclusive access. Anyone with an agenda can corrupt AI results.
There is no study that shows that PM2.5 particles cause premature deaths. Nor is there any way to measure it, though they claim with absolute certainty that it causes, for example, 5 million premature deaths annually, or some other arbitrary number. This is how they come up with that spurious claim:
People with emphysema, asthma, and other chronic respiratory illnesses tend to have shorter lifespans than people without those illnesses.Dust, ozone, and air pollution tend to exacerbate the symptoms of respiratory illnesses.Therefore, more PM2.5 must cause premature deaths.Ignoring the fact that the premature cause of death is a chronic illness, not particulates or ozone, they use that as the basis to manufacture statistical “evidence” that PM2.5 “causes” premature deaths.
That’s how the game is played by alarmists.
Thank you Authors
I have been sent a quick reaction by an economist with experience in air quality regulation and litigation and they would be happy to discuss more to prepare a more robust report to Heritage. I am posting this by request. Contact me if a deeper consultation would be beneficial.
++++++++
The fundamental questions are biological in the first place
i) “Premature deaths” are definitional and the assigned cause at death does not mention co-morbidities.
ii) There is no exposure-response function for PM2.5 of different composition and sources and particular diseases, most of which do not lead to death (depending on age/sex, confounding factors).
iii) EPA assumes equitoxicity for all PM2.5 and assumes there is “no threshold”. Since the basic association of PM2.5 and diseases is itself doubtful – from the Harvard Six Cities study to its reworking, if I remember correctly, or maybe a 30-city study – these assumptions are moot.
The biggest observational cohorts were active and passive tobacco smoking, but smokers do not die of the same disease or have a set “pre-maturity”: five years or 20 years, against the arbitrary standard of 86 years in the IHME’s Global Burden of Disease (GBD) or other such, though the US does not subscribe to the GBD modeling in the first place.
iv) The climate change –> air pollution argument is again, based on specific events – more wildfires, say – that are quite contextual. Washington DC metro area has a baseline of urban fires of x events and y bursts of pollution over the last 50 years. I am pretty sure fire management and fire prevention techniques have improved a lot in these 50 years and the events are far less frequent now than 50 years ago. Climate models rely on certain predictions of regional temperatures, precipitation, and population to generate the “rise in premature deaths” fear, but basically, these models are spurious, ergo fundamentally wrong. (See above the comments on exposure-response, equitoxicity, comorbidity and confounding).
Practically NONE of the global modeling is based on particular cohorts in each region – say, cities in sub-Saharan Africa, whose population would have grown from something like 100 million in 1980 to over 1 billion in 2040, associated with massive changes in the built habitat and sources of pollution or other risk factors.
++++++++
From me:
If you have never heard of the Bill Gates-funded IHME and its GBD guesswork, assumptions and modelling, read a little so you know where this comes from, lest you assume it is mainstream science in good standing. If someone dies before the age of 86, the shortfall is assigned to about 175 contributing causes, including murder, suicide and smoking. A little bit each. They add up the little bits until it comes to 86 years and then say “one in x-million people died from it”.
You will note in literature that claims a “contribution to premature deaths” is converted at the whisk of a pen into a “cause of death” and “causes deaths”. Those authors (and there are many) lack an understanding of what is being claimed. A simple paper attributing exposure to (not inhalation of – that is an assumption) smoke contributes to the premature deaths of 4.2 m people per year is easily turned into “causes 4.2 million deaths per year.” In 2012 the WHO, misquoting its own 2011 paper, made just such as assertion. Everyone else quotes the WHO, 2012 because it sounds official. The 2011 paper traces to the IHME. Surprised?
The EPA assumption of equitoxicity of all air polluting particles is based on the bizarre (and old) assumption made “because we do not have any evidence that they are not” (equally toxic). I think people with mesothelioma would disagree.
The parts about ozone are almost humorous. No one knows what causes asthma. Claiming that ozone causes asthma is very different from claiming that if someone has asthma it will be aggravated by ozone. That is hardly news. The recent attempt to “ban gas stoves” was based on a claim that it “causes childhood asthma”. Yeah? If you see that claim, ask them for the reference to the article saying so and see if it contains any facts.
With respect to the question “How can updating the Reference Manual to address junk science best be accomplished given such a massive input from entrenched academics in the past?”
it is suggested that a Congressional Inquiry be constituted with a challenge to the basic assumptions and methodologies that is first circulated to all the “entrenched” agents.
Most of them would find it difficult to defend the assumption of equitoxicity, or of a “single cause of death”, or any exposure-response curve they have used. The conclusions then fall away.