So the other day I was having a conversation with a friend of mine where he was arguing in support of the use of genetically modified organisms (GMO). While he did grant my point that there can be financially driven bias amongst the scientists studying the harmful effects of GMOs, he dismissed the idea that Monsanto and other companies seeking FDA approval for the latest GMO technology could be influencing the regulatory process in their favor. The basis for this denial was that independent scientists had reached the same conclusions as the corporate sponsored scientists.
As I went over the conversation in my mind later, I realized that I had neglected to mention a major reason for my skepticism of the scientific studies that are used to demonstrate that GMOs and vaccines are safe. Of course, the primary reason is the aforementioned fraud and corruption where regulators are often hired by the industry they regulate once they are done working for the government, but there is a reason that even well-meaning scientists can underestimate the risks associated with various compounds. This has to be with the simplistic model that is used by scientists studying toxicity. This model is as follows:
Compound A —> Condition B
How would you test the idea that compound A causes condition B? You would introduce compound A to a research animal and measure the resulting frequency of condition B. The problem is that real toxicity is almost never as simple as this simple mono-modal causality suggests. Rather, toxicity is caused by an extremely complex set of factors including the reactions of Toxin A with other compounds in the organism, available transport mechanisms across biological membranes, the rate of elimination of the compound by biological filtering mechanisms, genetic predisposition, cellular damage caused by other toxins, stress level, developmental level and other factors. This multitude of factors is why compounds that sometimes show promise as a treatment for various conditions in a test tube(in vitro) are ineffective in a living organism (in vivo). It is also why toxins affect various people differently with different dosage levels. If we modify the simple toxicity model above to a more realistic model by writing out various possible toxicity scenarios, a very much more complex picture emerges:
Compound A + Compound C —> Condition B
Compound A + Genetic Condition D —> Condition B
Compound A + Environmental Factor E —> Condition B
Compound A + Environmental Factor E + Genetic Condition D —> Condition B
So the job of someone testing a new GMO or a vaccine to ensure that it is safe to use in the general population is enormous. Compound A must be monitored for a sufficient period of time for all possible conditions B while varying all possible environmental factors E all possible genetic conditions D and all possible compounds C. As anyone with even basic understanding of mathematics will know, this kind of testing would be extremely prohibitive. The only way to approach this problem is with statistical methods.
Unfortunately, the complexity of biological systems defeats what statistical methods can accomplish in this regard. What if the mechanism of toxicity requires a genetic condition that has a frequency of 1/100 with an environmental factor that has a frequency of 1/20? In this case, to see one case of condition B from this mechanism would require a sample size of 2000 people. How many toxicity trials study this many people? Even if the study saw a single case of condition B, would it not be considered statistical noise and ignored?
Now for the case of drugs, the simplistic models that are used for toxicity analysis are fine. If people are dying of cancer or have some other serious medical condition, do we really want to deny them the chance of a cure because there are some rare cases where there will be side-effects to the medication? Probably not. The case is entirely different, however, for GMOs and vaccines where otherwise healthy people are going to be given a compound that could have these kind of rare side-effects. A side-effect that has a frequency of 1/10,000 in a vaccine or a GMO will appear 30,000 times in a population of 300 million. This is a much bigger hazard than occurs with drugs and a different standard of testing is needed for compounds that are going to be introduced into the general population.
Now I do not have a problem with people eating GMOs or getting their children vaccinated if this is what they choose to do. The problem that I have is simple-minded people who grab on to science that they don’t understand and say, “Anyone who doesn’t agree with the science is stupid and ‘anti-science’. They must be forced to comply with the science for the good of the children.” I have met simple-minded people who actually think they are on the side of right and truth because they want the health gestapo to kick down the doors of concerned parents and force them to be vaccinated at gunpoint. I have likewise met people who want to make it illegal to label GMO foods so that it is impossible to choose to avoid the risks associated with them. Such people need to understand the limitations of science before they force their views on other people.
I have discussed other aspects of this issue in my post “Exaggerated Science“.
It will, of course, occur to every reader that experts in toxicity must be aware of the possibility of multi-factor toxicity. This is surely true, but it does not mean that they incorporate such a possibility into their methodology. Why? There are a few reasons. First, that the primary toxic effects will be direct single factor effects for most cases of interest. Second, that there really is not much that can be done about multi-factor toxicity as it is impossible to test for every possible combination of factors. Third, because the most prominent effect of including multi-factor toxicity into the FDA approval process would be greatly increasing the cost and amount of time that was required to test various products. Now some might ask, “do we even know that multi-factor toxicity exists?” The answer is obviously yes. Whenever you see the commercial for the latest pharmaceutical, the warnings against taking the medication with other substances all come from the FDA trials. The problem is that these are only the effects that had a high enough probability to show up in the trials. Also, the “Exaggerated Science” post linked above contains one documented case of multi-factor toxicity and a discussion of the toxicity of Agent Orange which I suspect to be another example.