Regular readers of this blog know of my ongoing disappointment with the MDs at the Canadian Association of Physicians for the Environment (CAPE). No group has so consistently disappointed me with the variance between the reports they are capable of producing and their actual output. As I have detailed in previous blog posts, they have produced bad research on BC LNG,  bad epidemiology, bad takes on projects like the Site C Dam and the Trans Mountain Expansion Project, and their biggest and ongoing bad work on natural gas flaring, fugitive emissions and the climate effects of natural gas.

Naturally this led me to approach their latest campaign: Fossil Fuel Ads Make Us Sick with a jaundiced eye. What I have determined is this campaign builds on the misinformation and bad epidemiology described above and then adds new bad angles and newer bad data. The crux of their campaign is that:

air pollution from the burning of fossil fuels is one of the leading causes of premature mortality in Canada

In a recent article in the National Observer: “Doctors know banning fossil fuel ads is a matter of life or death” they have added another questionable claim that:

fossil fuel air pollution is responsible for one in seven premature deaths in Canada

As is normal with the work by CAPE, there is always a sliver of truth in their articles and campaigns but invariably their campaign seem to be built on an incomplete (or simple misreading) of the the underlying research. That being said, now I need to support my claims with actual research. So let’s begin:

Fossil Fuels are a leading cause of premature death

Let’s look at that claim that: “air pollution from the burning of fossil fuels is one of the leading causes of premature mortality in Canada”. Their claim links to a Health Canada report with the title: Health Impacts of Air Pollution in Canada: Estimates of morbidity and premature mortality outcomes – 2021 Report. Upon reading this report one thing becomes abundantly clear: the claim made by CAPE is not supported by (or even made in) the report.

Reading the Health Canada report I was most struck by the absence of any significant statements about fossil fuels. Specifically, the term “fossil fuels” appears only a single time in a discussion about the formation of nitrogen dioxide compounds. The report does indicate that fossil fuels are a leading cause of premature mortality in Canada.

The Health Canada report identifies “air pollution” as a major cause of premature mortality and then it goes on to discuss the causes of air pollution but in doing so it provides the data to point out that the burning of fossil fuels represents only a very minor cause of that air pollution.

Lets start with the PM_2.5 which according to the Health Canada report represents the source of approximately two-thirds of the premature deaths. For those unfamiliar with the term, PM_2.5 refers to a class of fine particulate matter (dust) that is sufficiently fine as to fit through a 2.5 micron filter. PM_2.5 is a particularly troubling component of air pollution because it is believed to have a disproportionate effect on human health and may even be small enough to affect fetuses in utero. Decreasing exposure to PM_2.5 is a solid goal that will improve community health. So what does the report say about the source of the PM_2.5? Let’s look at their table:

Yes, you read that right, the “Oil and Gas Industry” and “Transportation and Mobile Equipment” contributed 48,000 tonnes of PM_2.5 to the national emission total…out of a total of 1,600,000 tonnes! Doing the math fossil fuels contributed approximately 3% of the total anthropogenic PM_2.5. Notice that qualifier: “anthropogenic”. That is an incredibly important proviso because forest fires produce almost the same amount of PM_2.5 as humans activities but are not included in the accounting in this report (“fires” in the report represents cooking fires) but absolutely affect human health.

For those of you who like graphics, see below a figure depicting the various sources of PM_2.5. What you will notice if you blow up the figure is those fossil fuels don’t appear. They are combined in the “Other” category because don’t even warrant their own colour.

With respect to mortality the report indicates:

Chronic exposure to PM_2.5 air pollution contributed to 8.0% of all-cause nonaccidental mortality among Canadians over 25 years of age, equivalent to 10,000 deaths per year or 27 deaths per 100,000 population.

As a source of 3% of the PM_2.5 that would make fossil fuels responsible for about 300 deaths per year. Certainly a tragedy but not “one of the leading causes of premature mortality” considering that “dust” is responsible for 5,000 deaths a year and CAPE is not leading a campaign against “dust”.

Now admittedly, fossil fuels represent a much higher percentage of NOx emissions (about 75% see below) but NOx is only responsible for 1300 deaths a year. Ground-level ozone is reported to be responsible for 2800 deaths a year and fossil fuels are responsible for up to 52% of those premature deaths but when added together they don’t hold a candle to our biggest nemesis: “dust“.

Now I am not trying to belittle those deaths but the point is this is the citation provided in support of the CAPE claim that “air pollution from the burning of fossil fuels is one of the leading causes of premature mortality in Canada” and this citation does not support that claim.

Fossil fuel air pollution is responsible for one in seven premature deaths in Canada

The second claim to be examined is the suggestion that fossil fuel pollution is responsible for one in seven premature deaths in Canada. This claim is derived from a paper Global mortality from outdoor fine particle pollution generated by fossil fuel combustion: Results from GEOS-Chem. Now in this case the reference actually makes that claim, but, as I will demonstrate, that claim is so obviously wrong as to be incredibly puzzling and clearly represents a failure in the peer review process

This article identifies the connection between PM_2.5 and premature death and using a proprietary model argues that 34,000 deaths a year in Canada can be attributed to PM_2.5. This is presented in Table 1 of the article below.

This number (34,000 deaths a year) is so much higher than the Health Canada value (10,000 deaths a year) that it has to raise alarm bells. Are we really to believe that Health Canada, the organization that compiles all this data for Canada, so completely missed out on all these deaths? Now remember Health Canada not only compiles these results but they correlate all the other causes of death and so we are to believe that Health Canada managed to misattribute 24,000 deaths a year?

That being said, even if their numbers were reliable it would still be a mess because look at Table 1. The study incorrectly asserts that fossil fuels are the source of 85% of PM_2.5 emissions in Canada. But wait, didn’t we establish above that fossil fuels represent only 3% of anthropogenic PM_2.5 emissions and likely less than 2% of total PM_2.5 emissions (when you add forest fires). That means instead on 1 in 7 premature deaths being caused by “fossil fuels” the number is closer to 1 in 100 of premature deaths. But that number doesn’t make nearly as good a headline for a political campaign.

As for the claim that 1 in 5 deaths worldwide being caused by fossil fuels, that number is similarly ridiculous. In the US the study attributes 81% of PM_2.5 deaths to fossil fuels while in Europe it is 75.7% of PM_2.5 deaths. Given that fossil fuels represent about 3% of Canadian anthropogenic PM_2.5 do we really imagine that either of those figures are at all credible?

I am quite certain the reply I will get from my critics is that this paper was PEER-REVIEWED [yes, it will be in all caps because that is how they address me on social media]. But as I have pointed out previously, the peer review process is notoriously challenged by multi-disciplinary work. This journal can’t get dozens of reviewers so they had to concentrate their review on the stuff that journal is about (modelling and epidemiology). It is likely no one stopped to ask whether the inputs for the models were appropriate because none of the reviewers would recognize the data looked wrong. It takes someone from that field to look at the data and say: “wait that number is wrong”. In this case anyone familiar with PM_2.5 would ask about forest fires (not included in the inventory as they are not “anthropogenic”) or about construction dust or about simple road dust. Sadly, none of the reviewers asked those questions and so we have a paper with a huge attribution problem based on bad inputs.

So once again we have well-intentioned health practitioners demonstrating my father’s (a physician himself) adage: “never trust an MD on any topic that is not related to medicine”. CAPE’s “Stop Fossil Advertising” campaign, while clearly well-intentioned, is simply not supported by the research it cites. Rather, the campaign appears to be based on a combination of bad science and a bad reading of the science all mixed together with a lot of good intentions..