The general logic/key points seem to be as follows:
- Some autism cases look like a form of inflammatory diseases with the immune system overactive (inflammation on high, anti-inflammation on low, or some combination thereof)
- Infection of a mother during pregnancy increases the risk of having a child with autism.
- In animal models, inducing inflammation in the mother (even without an infection) leads to an increased risk of behavioral "problems" in her offspring
- Inflammatory and/or autoimmune diseases (e.g., asthma) have increased in incidence along with autism.
- If a mother has automimmune or inflammatory diseases such as rheumatoid arthritis celiac disease she has a higher risk of having a child with autism. Similarly if a mother has allergies or asthma during the second trimester, there is a higher risk of having children with autism.
- Many automimmune and inflammatory disorders and autism are all more prevalent is the developed world.
- The developed world is generally cleaner that the developing world.
- There are many fewer parasites in people in the developed world.
- Parasites are known to suppress inflammation.
- Therefore, we may be able to stop/limit autism, asthma, and other inflammatory diseases by purposefully infecting people with parasites from our evolutionary past.
UPDATE - Emily Willingham has written a VERY detailed critique of the article that I think everyone interested in anything related to this topic should read: Emily Willingham: Autism, immunity, inflammation, and the New York Timeswww.emilywillinghamphd.com.