3 ms·
If the effect size were much larger, like an OR of 3, that would be suspicious as well, since the ORs for suicide/accident/mortality/schizophrenia and Toxoplasm
by gwern 10y ago
If the effect size were much larger, like an OR of 3, that would be suspicious as well, since the ORs for suicide/accident/mortality/schizophrenia and Toxoplasma also tend to be in the 1-2 range. I would consider this as part of a broader picture showing a pervasive small-medium correlation of toxoplasma infections & bad things involving the central nervous system.
It's interesting because we know that a lot of other infectious agents can really screw up the nervous system, and psychiatric disorders repeatedly show genetic correlations with the immune system.
- et2o 10y agoVery happy to have your comment, Gwern. I am a big fan of some of your work. I certainly agree with you in principle. I just am not sure the sample sizes here are truly large enough to capture an effect on the order of 1.25x for such a complex phenotype. It's not necessarily a criticism of the author's work, which appears to be well-done. I just updated my prior and remain not yet convinced. I also agree that many psychiatric disorders certainly have integral immune components, but on the other hand this is also true for most chronic diseases and pathologic processes (in my opinion).