As a semi-retired epidemiologist, in a higher risk age group and with attendant co-morbidities, I have followed the Covid-19 pandemic with scientific curiosity mixed with a tinge of personal anxiety.  Much of the data being reported is of abysmal quality, and it’s a major professional disappointment to me that, after more than four months, the situation hasn’t improved much.

Family members and friends, some of whom I haven’t heard from in years, continue to pepper me with their well-intentioned questions, fueled by the latest reports in the mainstream and social media.  Like many others, they have a difficult time sorting through the multitude of competing claims – many of which are downright crazy -- to make sense of things.

Growing weary of the conflicting and confusing reports, the locally mandated “stay at home” orders and the dire consequences to their personal finances, many Americans are clearly frustrated. A surprising number is no longer heeding the CDC’s advice to avoid crowded in-door settings, practice social distancing, wear masks in public and frequently and vigorously engage in hand washing.  The public’s trust in science, already shaky even before the pandemic, has become further eroded—and too many are now dismissing public health advice.  


Did "endocrine disrupting chemicals" at trace levels cause coronavirus to spread? There is no plausible hypothesis for how that's even possible but it doesn't stop Linda Birnbaum and Fred vom Saal from writing it might be so.


I witnessed this first-hand recently when we finally were able to gather together to bury my mother-in-law who passed away in early March.  The lack of compliance was alarming. Like other public health experts, I fear that such behavior will spawn further waves of infection and a devastatingly higher than necessary death toll.

Some may want to place the blame for rampant misinformation solely at the feet of the media (and they certainly deserve their share of the blame), but some scientists and other experts are also an important source of the problem.  Of course, the most blatant example has been the sudden retraction of two studies from prominent medical journals due to possible fraud, but there are many others, as well. 

Perhaps you also read Tommaso Dorigo’s interesting article “The Virus That Turns Physicists Into Crackpots” on Science 2.0.  His conclusion, “…the scientific method is something to treasure and hold dear and protect” certainly resonated with me.  He further noted, “It is not something we can give for granted, or something that once you learn to apply will stick with you for the years to come. Scientific truths can only be learnt by a painful, slow process made of small uphill steps.” Unfortunately, too many scientists these days have forgotten this and, often motivated by personal agendas, continue to try to short-cut the process by jumping directly to policy-making from incomplete, inaccurate and uncertain science.

 

A closer look: Unfounded allegations about chemicals and COVID-19 outcomes

As another example, I recently came across a series of four Op-Eds one by Heindel and Birnbaum, a second by vom Saal and Cohen, a third by Myers and a fourth by Trasande and Ghassabian that allege that low level exposure to purported endocrine disrupting chemicals (EDCs) has contributed to high prevalence rates of selected chronic diseases in the US population that are alleged to have been shown to render Covid-19 patients more vulnerable to severe complications, including increased mortality. 

A small, yet clearly coordinated echo chamber

Although published separately by four groups, the articles should NOT be considered independent of one another as at least four of the authors (Birnbaum, Heindel, vom Saal and Myers) have worked closely together for more nearly 30 years to promote their unconventional views.  This fact, the overlapping messages, and the close timing of all of the four Op-Eds, strongly suggests the authors have collaborated in an attempt to amplify their message in the media.  They seem to operate most comfortably in an echo chamber they have created for themselves, and unsuspecting readers might be given the impression that their claims are more mainstream than they really are.  Further evidence of their collusion is the June 18 webinar organized and delivered by authors from three of the groups.

For many years, these authors have been pushing an unproven and dubious hypothesis, i.e., that exceedingly low level exposures (far below those of more potent natural exogenous or endogenous hormones) to certain synthetic chemicals are an important cause of many chronic diseases, and are now opportunistically exploiting the Covid-19 crisis to further their cause.  (In the case of vom Saal and Cohen, they are also trying to exploit the situation to selfishly promote sales of their forthcoming book, which they none to subtly advertise at the end of their article.) Trasande also recently published a book pushing the hypothesis and has been shamelessly traveling the country promoting sales. 

At best, their efforts should be considered a distraction that takes the public’s eye off of the most important things they can do at this time to protect themselves and their loved ones. At worst, it is nothing more than blatant opportunism exploiting a global tragedy that is reprehensible and should be called out for what it is by the entire scientific community.

A lack of restraint, a lack of evidence

On a positive note, I should start by acknowledging that, after many months of being bombarded with conflicting information about COVID-19, the public has not totally lost trust in public health experts. A recent survey conducted by the New York Times and Siena College shows a large majority of Americans registered to vote quietly trust the advice of medical experts.  But the research also shows two key factors tend to dilute trust in public health: political polarization and mixed and confusing messages.  Both are present in the current Covid-19 crisis and are trending in the wrong direction.  As a another sign of the fragility of public trust, consider that one-third of Americans said they would NOT get vaccinated against Covid-19, even if a safe and efficacious vaccine is widely available at low cost.  

A significant irony exists when one contrasts the meticulous, science-driven approach taken by Dr. Anthony Fauci, Director of the National Institute of Allergy and Infectious Diseases, when he discusses the Covid-19 outbreak, with the more freewheeling, speculative approach taken by the authors whose claims are the subject of this analysis. Fauci sticks to what has been firmly established through the scientific method (e.g., he insists on well-designed and conducted clinical trials to evaluate potential preventative as well as therapeutic treatments and weight of the evidence evaluation) and refuses to engage in speculation.  By contrast, the authors of the four Op-Eds employ much less rigorous scientific standards and seem to revel in speculation.

Although I was initially reluctant to write anything about these articles for fear of giving them more attention than is warranted, I’m convinced that, if left unchallenged, their messages will ultimately do more public harm than good.  One reason for my concern? The authors are clever – they skillfully weave well-established facts and mere speculation seamlessly together to make their points.  Below, I will try to disentangle the two to bring greater clarity to the situation.

Facts

 

  • Although it is true that the prevalence of certain chronic health conditions (e.g., obesity, and type 2 diabetes) have been rising in the U.S, the CDC and most scientists believe this is largely or wholly attributable to: high-calorie, low-nutrient foods and beverages; not getting enough physical activity; sedentary activities such as watching television or other screen devices; medication use; and poor sleep routines.  Most authoritative reviews of the causes for obesity and type 2 diabetes do not even discuss a possible role for purported EDCs.

 

  • The trends in the prevalence of other diseases that were lumped together by the authors of the four op-eds are more complex than they acknowledge as is explained below.  Once again, most authoritative reviews of causes for these diseases do not even mention a role for purported EDCs.
    • Chronic Respiratory Disease:  A 2017 published study found mortality from chronic respiratory disease increased in the U.S. from 1980 to 2002, but has been declining since.  Chronic respiratory disease includes: chronic obstructive pulmonary disease (COPD), interstitial lung disease and pulmonary sarcoidosis, asthma, and all other chronic respiratory diseases.  The op-ed articles highlight asthma, however, the asthma mortality rate actually declined by 46.5% between 1980 and 2014.  Nearly three-quarters (73.4%) of deaths due to chronic respiratory disease in the United States have been attributed to tobacco smoke, including 81.8% of deaths due to COPD.  Occupational exposure to certain dusts (asbestos, coal dust and silica) is a significant risk factor for mortality from chronic respiratory diseases, particularly pneumoconiosis; however, long-term declines in mortality attributed to these dusts, follows the establishment and enforcement of exposure limits in the 1960s and 1970s, as well as declining participation in occupations associated with exposure.
    • Cardiovascular Disease (CVD) – From 1970 to 2010, the death rate from CVD dropped in half in the US and this has been attributed to improved pharmacological management, advances in surgical techniques, and healthier lifestyles.  Since 2010, the pace of decline in the death rate from CVD has slowed for unknown reasons.  According to the CDC, the most important risk factors for CVD are high blood pressure, high cholesterol and smoking.  Other contributing factors include: diabetes, obesity, unhealthy diet, physical inactivity, and excessive alcohol use.
    • Autoimmune System Diseases/Dysfunction: A 2015 systematic review found the following:
      • Multiple publications exist, describing past or actual incidences/prevalence of individual autoimmune diseases, however, long term studies on selected populations are scarce.
      • Results: The means ± s.d. of the net % increased /year incidence and prevalence of autoimmune diseases worldwide were 19.1 ± 43.1 and 12.5 ± 7.9, respectively. Rheumatic, endocrinological, gastrointestinal and neurological autoimmune diseases revealed the following annual % increases per year: 7.1, 6.3, 6.2, and 3.7, respectively.
      • In all of these, differences between old vs new frequencies were highly significant (p< 0.0001). Comparing various autoimmune diseases, celiac disease increased the most and the highest increase in incidence, comparing old to new surveys is allocated to myasthenia gravis. Despite considerable variations between the countries, celiac, type 1 diabetes and  myasthenia gravis frequencies increased the most in Canada, Israel and Denmark, respectively.
      • Frequencies of the autoimmune diseases increased significantly in the West and North when compared to East and South, respectively. Rheumatic, endocrinological and gastrointestinal autoimmune diseases in Israel, Netherlands, USA and Sweden increased the most.
      • These observations point to a stronger influence of environmental factors as opposed to genetic factors on autoimmune disease development; however, the definition of environmental factors should be interpreted in its broadest sense (i.e., including lifestyle factors) and the precise factors that contribute are unknown.

 

 

 

 

Speculation

 

  • As evidence of the type of speculation in which these authors engage, consider the following quote from vom Saal and Cohen:
    • “In the U.S., these chronic diseases have been steadily increasing over the past 50 years, associated with the dramatic increase in chemical production for use in plastics, construction materials, pesticides, personal care products, furniture, cookware, food packaging, textiles, and many other products that are steadily infiltrating every aspect of human life.”
    • First, as discussed above, the trends in prevalence of these diseases has not been uniformly increasing during the past 50 years as is alleged.  Also, the correlation they hypothesize is really quite crude, and, at the risk of insulting readers its worth remembering that correlation does NOT equal causation.  One needs to only consider the many other changes that have taken place in society in the past 50 years (dietary changes, less physical activity, computer and internet usage, rate of consumption of alcohol, volume of air travel, etc.) to understand how over-simplified the authors’ logic is.
  • The authors of the Op-Eds claim that causal links have been established between exposures to EDCs and the chronic health conditions they list; however, this is mere speculation on their part and there is not a consensus among scientists that this is true. 
  • In prior publications by these authors and their collaborators they have been guilty of cherry-picking the scientific literature, including only the studies that support their hypothesis and excluding those that don’t.  They have often relied heavily on novel toxicology studies that failed to conform to international guidelines, and which have not been independently replicated. These studies have often employed small numbers of uncommonly used strains of animals, questionable dosing and flawed statistical methods.  In formulating their allegations, these authors have also relied heavily on observational epidemiology studies that cannot, in and of themselves, reliably establish causal relationships between putative EDCs and health effects.  Such studies are often limited by small population sizes, poor assessment of exposures, and suffer other flaws which make it impossible to rule out bias, confounding or chance as alternative explanations for the associations they report.
  • The following quote from the Heindel and Birnbaum op-ed is telling:
    • “While we can't pin the elevated U.S. disease numbers solely on exposures to endocrine disrupting compounds, it is clear that all of the disease and health conditions listed above (diabetes, obesity, heart disease, immune system diseases/dysfunction and respiratory diseases) have been linked to exposure to a variety of endocrine disrupting compounds in animal models and human epidemiology studies.”
    • The use of the phrase “…linked to…” seems to be a tacit acknowledgement that causation has NOT been firmly established.  As noted above, most authoritative reviews by scientists who study those chronic diseases do NOT even acknowledge EDCs among the risk factors they identify.

 

  • And according to Trasande and his colleague:
    • “Is the evidence perfect? Hardly. And we have to rely on observational studies – you can’t run a randomized controlled trial of potentially toxic mixtures of virus and chemical exposures. There are ethical and logistical challenges to running these kinds of studies. But absence of evidence doesn’t mean absence of harm.”
    • No one could disagree that the evidence these authors rely on is far from perfect – in fact, as Trasande acknowledges it is largely absent.

 

  • Heindel and Birnbaum also include the following quote:
    • “Improving our diet and nutritional status and reducing our exposure to endocrine disrupting chemicals are pivotal changes that will profoundly benefit our health and welfare.”
    • They focus exclusively on diet and EDCs to the exclusion of many other far more important risk factors such as: tobacco use, alcohol and drug abuse, untreated high blood pressure, untreated high cholesterol, lack of physical activity, sedentary lifestyles, etc.  This does NOT constitute sound public health advice.

 

  • Similarly, vom Saal and Cohen state:
    • “Individual lifestyle changes, which include increasing intake of nutrient-rich, unprocessed fresh or frozen foods that do not contain endocrine disrupting chemicals, drinking water without pollutants, improved sleep, stress management, and regular aerobic exercise, are regarded as cornerstones in reducing risk for developing most chronic diseases.”
    • They fail to mention tobacco use, alcohol and drug abuse, untreated high blood pressure, and untreated high cholesterol and thus are also guilty of miserably failing to educate their readers on the most important risk factors.

 

  • The authors allege there are numerous EDCs in commerce, yet they often confuse chemicals that are merely endocrine active with those that are true EDCs.  The WHO defines an EDC as “an exogenous substance or mixture that alters function(s) of the endocrine system and consequently causes adverse health effects in an intact organism, or its progeny, or (sub)populations.”  By the WHO definition, there are far fewer EDCs in commerce than has been alleged.

 

 

  • The authors also speculate a role for some EDCs in causing inflammation without citing a convincing body of scientific evidence.  They further speculate on possible synergy between poor nutrition and exposure to EDCs to create an abnormal inflammatory response, but once again there is little to no scientific evidence to support this.

Just last week, The Intercept published an article that picked-up on the hypothesis from these authors and included reference to some additional research that they and their colleagues are planning.  For instance, Philippe Grandjean, a Danish scientist and frequent collaborator with this group, is reportedly in the process of collecting blood samples of people who were hospitalized with Covid-19 , analyzing them for PFAS levels, and comparing them with PFAS levels from the blood of people who were infected with the coronavirus but not hospitalized.  Presumably, he will be employing a cross-sectional study design with a single-point-in-time measurement of exposure after disease has already been diagnosed.

I’ve previously discussed the severe inherent limitations of such an approach to identifying potential causal relationships between purported EDCs and disease and have argued that it is time to move beyond cross-sectional studies to more reliable epidemiology study designs. Supporting my contention, just recently, the European Food Safety Authority opined that they considered the evidence from cross-sectional studies as too unreliable to give them much weight in their safety assessments.

Others are using animal experiments to explore how chemical exposure affects the impact of the coronavirus. Paige Lawrence, a professor of environmental medicine at the University of Rochester School of Medicine and Dentistry, plans to infect mice with a mouse-adapted human coronavirus that was built from the 2003 SARS pandemic and study how exposure to PFAS alters the course of the viral infection.  As others have pointed out, the viruses that cause SARS and Covid-19 resemble one another at some level, nevertheless there are distinct features that will inevitably call into question the generalizability and relevance of the results (e.g., SARS had a much higher case fatality rate, whereas Covid-19 is far more infectious).

When the results of these planned studies are published readers would be advised to treat them skeptically, at least until other scientists have had a chance to independently replicate them using stronger and more relevant methods.  One of my favorite quotes is from an article by Julia Belluz and Steven Hoffman:

 

“That science can fail, however, shouldn’t come as a surprise to anyone. It’s a human construct, after all. And if we simply accepted that science often works imperfectly, we’d be better off. We’d stop considering science a collection of immutable facts. We’d stop assuming every single study has definitive answers that should be trumpeted in over-the-top headlines. Instead, we’d start to appreciate science for what it is: a long and grinding process carried out by fallible humans, involving false starts, dead ends, and, along the way, incorrect and unimportant studies that only grope at the truth, slowly and incrementally.”

 

“A theory that explains everything, explains nothing”

In conclusion, after conducting a detailed review of the articles in question, I can say with confidence that the authors have blatantly mischaracterized the weight of scientific evidence linking chemicals to chronic disease by weaving together a narrative that tries to blur the lines between fact and speculation. They implausibly link EDC exposure to a broad range of chronic diseases, which calls to mind for me the quote from the brilliant science philosopher Karl Popper, “A theory that explains everything, explains nothing.”  At risk is a further undermining of public trust in science and public health advice that are critical to controlling the spread of this novel, and dangerous virus.

Unfortunately, given the significant flaws in the evidence cited to support their hypothesis, and the previous track records of the authors in question, readers are unfortunately left to deduce that the authors may in fact be shamelessly exploiting a global tragedy to promote their own personal agendas.

By publishing this piece, my hope is that other scientists with similar concerns will also join in calling them out for it.