The Causes of Autism

May 8, 2026

Event ≠ Not Event
Event-Probability = Event / Events + Not Events
therefore, Event ≠ Event-Probability

Adverse-Event = Event
Adverse-Event Probability (AEP) = Event / Events + Not Events
This is true for one variable that impacts adverse-event probability.

Variable 1(V1)
Variable 2(V2)

However, if…
“AEP(V1) = AEP(V2)”,
can the equation ever be truly equal (=) to arrive at conclusions regarding differences in adverse events for 2 different variables? I hypothesize this to be the equation that has been the source of much debate among vaccine-safety advocate criticisms regarding lack of saline solution placebo controlled clinical trials, where if, for example, the equation is 50% = 55%, this can result in reinterpreting/misinterpreting the data that established those probability results in the first place (e.g., 50 / 50 + 50 = 50%), where the dataset had 50 adverse events for Variable 1, and 55 adverse events for Variable 2 (e.g., 55 / 55 + 45 = 55%).

“AEP = AEP” vaccine clinical trials could be quantified as
55 / 55 + 45 (vaccine-1) = 50 / 50 + 50 (vaccine-2)
when in reality it should be…
55 / 55 + 45 (vaccine-1) ≠ 50 / 50 + 50 (vaccine-2)
because…
Adverse-Event Probability(V1) ≠ Adverse-Event Probability(V2)
and not…
Adverse-Event Probability(V1) = Adverse-Event Probability(V2)
to reject the data and say there “is no relationship” between the variable that established those probabilities in the first place.

Variable(1) ≠ Variable(2),
because Variable(1) could be chocolate cake and Variable(2) could be chocolate ice cream and the Event could be blood sugar spikes, thus,
Blood Sugar Spike(Chocolate Cake) ≠ Blood Sugar Spike(Ice Cream)

In other words,
if 50 events happened in dataset 1 for Variable 1,
then 50 events happened in dataset 1 for Variable 1,
and this data is separate from dataset 2 for Variable 2.

Hypothesized Correction:

If…
“AEP(V1) ≠ AEP(V2)”,
then…
“AEP(V1) = E / E + Not-E ≠ AEP(V2) = E / E + Not-E”
where the two variables are treated as unequal because 1. each variable impacts its own probability, and 2. one variable is not the other variable, the two are mutually exclusive (e.g., V1 ≠ V2).

Another Hypothesized application:
Oxidative Stress (OxSt) from Organophosphates (V1) = Event
Oxidative Stress (OxSt) from Pyrethroid (V2) = Event
Organophosphates (V1) ≠ Pyrethroid (V2)
OxSt-Event = can be measured in a variety of ways (e.g., DNA/RNA damage)
Organophosphates OxSt Probability = E / E + Not-E
Pyrethroid OxSt Probability = E / E + Not-E
…although I admit I’m skeptical about there being any “Not-E” events for those variables, but data is data, and if in every single case of exposure that is an Oxidative Stress event as measured through whatever means, then that’s the data, and its indicating oxidative stress due to exposure to that variable -both organophosphates and pyrethroids have statistically significant odds ratios in the literature for risk of autism.

Organophosphates OxSt Probability ≠ Pyrethroid OxSt Probability,
because “Organophosphates ≠ Pyrethroid” due to being chemically unequal, thus they would have their own set of probability of oxidative stress, even if the data shows the levels may be similar.

I would happen to know about two groups being unequal to each other due to the design of a study, as for my master’s thesis I designed a control group in a manner I thought would have zero effects, but it did end up having effects and was not significantly different from the treatment group whose condition was designed based upon the literature, which pretty much messed up the entire study because both groups ended up having similar effects. I never officially published those findings, but in reflecting back on it, and in the context of these new equations I’m designing/hypothesizing, I actually now ask myself: did I accidentally discover a new ‘treatment’ group that had similar effects to the treatment group whose design was based on the literature, when I meant to design a neutral control group?

The moral of the story here is:
if the variables aren’t equal,
then they aren’t equal,
and two separate sets of data are produced.

Autism Librarian