The University of Sydney caps doctoral theses at 80,000 words (excluding references). The theory is that external reviewers don’t want to read more than that (true!). One can apply to the Dean to increase the word limit to 100,000, which is what I did. But my doctoral thesis, as initially written, was closer to 140,000 words. So I had to cut three chapters that I really liked — the political economy of theories of genetic causation, how evidence-based medicine was captured by Big Pharma, and the history of the regulation of mercury.

I believe that some of the information in those excised chapters would be useful to policymakers in Washington, D.C. trying to figure out how to deal with the epidemics of chronic disease in children. So today I am sharing my original (slightly updated), never-before-seen, chapter 6, which challenges the entire paradigm of genetic determinism in disease causation.

I. Introduction

In the first chapter, I showed that the rise in autism prevalence is primarily a story of environmental triggers (with some smaller percentage due to diagnostic expansion and genetics). The story of how genetic theories became the dominant narrative in the autism debate thus needs to be explained. The hegemony of genetic theories of disease causation comes at a tremendous cost to society because they crowd out more promising alternatives. This problem is particularly acute in connection with autism, where genetic research swallows up the vast majority of research funding — and has for more than twenty years. So, one of the keys to effectively addressing the autism epidemic will be to demonstrate the flaws in the genetic approach to disease causation and replace it with a more comprehensive ontology that has better explanatory power.

To put this debate in context, I want to recap the genetic argument in connection with autism as I have presented it thus far. In the 1990s, it was routine for scientists, doctors, and policymakers to assure worried parents that autism was genetic. To the extent that anyone ventured a guess, the explanation was that autism was 90% genetic, 10% environmental. Then the state of California commissioned 16 of the top geneticists in the country (Hallmayer et al. 2011) to study birth records of all twins born in the state between 1987 and 2004. Hallmayer et al. (2011) concluded that at most, genetics explains 38% of the autism epidemic, and they pointed out twice that this was likely an overestimate. Blaxill (2011) argues that the eventual consensus will be 90% environmental, 10% genetic. And in chapter 5, I showed a model from Ioannidis, (2005b, p. 700) that suggests that only 1/10th of 1% of “discovery oriented exploratory research studies” (which include nutrition and genetic studies with massive numbers of competing variables) are replicable.

And yet, a disproportionate share of federal research money in connection with autism is going to study genetic theories of disease causation. In 2013, the Interagency Autism Coordinating Committee spent $308 million on autism research across all federal agencies and private funders participating in research (IACC, 2013a). This is a shockingly low amount to spend on research given estimates that autism is currently costing the US $268 billion a year (Leigh and Du, 2015). …

Hubbard (2013) points out that lost amidst the exuberance of the discovery of DNA and the double helix and the mapping of the human genome lies the potential for unintended consequences. Biological systems are more complex than the monogenic theory of disease causation suggests. This means that one simply cannot know how genetically engineered interventions will turn out.

Biotechnology — the industry of “genetic engineering” — is built on the pretense that scientists not only understand but also can anticipate and direct the functions of the DNA sequences they isolate from organisms or manufacture in the laboratory. The industry cheerfully promises that it can foresee the potential effects of transferring specific DNA sequences, wherever and however obtained, into bacteria, plants, or animals, including humans, and thus improve targeted characteristics. In reality, such operations can have three possible outcomes: (1) in the inhospitable environment of the cells of the host species, inserted DNA sequences do not succeed in specifying the intended proteins, so nothing new happens; (2) the inserted sequence mediates the synthesis of the desired protein product in the right amounts and at the right time and location; and (3) unpredicted and unintended consequences follow because the inserted DNA gets spliced into the wrong place in the genome of the host organism and disrupts or adversely alters one or more of its vital functions.

The first alternative wastes time and money, the second is the hope, and the third spells danger. Yet which of them happens cannot be predicted a priori, or from one genetic manipulation to another, because the conditions within and around the host organisms are likely to change over time.

If Hubbard is correct — that one cannot predict ahead of time how a genetically modified organism will impact its host — that potentially has profound implications for the autism debate. That is because one of the changes that followed the passage of the 1986 National Childhood Vaccine Injury Act was the introduction of genetically engineered vaccines — starting with the Hepatitis B vaccine in 1987. Four genetically engineered vaccines are currently on the CDC’s recommended schedule for the whole population: Hepatitis B, human papillomavirus (HPV), influenza, and Covid-19. Since 2006, the MMRII has been grown in a medium that includes recombinant (genetically engineered) human albumin (Wiedmann, et al. 2015, p. 2132). …

In the 1990s and 2000s government and industry had a theory of the case — that genes are responsible for disease — that has now been largely refuted. In the meantime an entire industry and public health infrastructure was built around this idea. So when the underlying theory was discredited, proponents simply modified the theory (to the search for the “missing dark matter”) so that the industry could keep going and continue to receive government funding. When this evolving research agenda produces profitable corporations and well-paid scientists but little to nothing that reduces human suffering it is an enormous problem for society.

The fact remains that Gilbert and Miller (2009), Landrigan, Lambertini, and Birnbaum (2012), the American College of Obstetricians and Gynecologists (2013), and Bennett et al. (2016) have all concluded that autism and other neurodevelopment disorders are likely caused by environmental triggers and are thus preventable through law and policy. Even if sophisticated genetic and genomic research is able to find ways to reduce symptoms and severity, it is still going to be orders of magnitude more cost-effective (not to mention more ethical) to prevent autism in the first place by keeping toxic chemicals out of children’s bodies.

Currently, genetic research is soaking up the vast majority of autism research funding and preventing more effective prevention strategies from emerging. This appears to be a reflection of the political power of biotech firms to shape the research agenda to serve their interests rather than a reflection of best practices in science or the best interests of society.

If Hubbard is correct — that one cannot predict ahead of time how a genetically modified organism will impact its host — that potentially has profound implications for the autism debate. That is because one of the changes that followed the passage of the 1986 National Childhood Vaccine Injury Act was the introduction of genetically engineered vaccines — starting with the Hepatitis B vaccine in 1987. Four genetically engineered vaccines are currently on the CDC’s recommended schedule for the whole population: Hepatitis B, human papillomavirus (HPV), influenza, and Covid-19. Since 2006, the MMRII has been grown in a medium that includes recombinant (genetically engineered) human albumin (Wiedmann, et al. 2015, p. 2132).