Humm, I would be interested in hearing Nikki's take on this.
You're lucky I'm a night owl, and popped by here after a particularly grueling study session... Perhaps not surprisingly, I sometimes find the conversations here more intellectually stimulating than the drivel they're ramming down my throat in school (of course, the reverse is often true as well). Ironically, I'm going on TERB hiatus for a few months because I've been neglecting that drivel in favour of board banter, and my GPA has taken bit of a hit because of it. But I'm still here for a few days, so I'll chime in with my late-night musings.
Disclaimer - my knowledge of the neuropharmacological data regarding Prednisone is limited, as I haven't researched it too closely (yet); however, I'm quite sure I know more about the field of behavioural neuropharmacology than any writer on the Toronto Star's staff. My expertise on this issue is relative, not absolute.
As a scientist, I recognize that correlational data can neither prove nor disprove a hypothesis. You can't simply give the drugs to a population and see if they develop clinical psychopathologies (good luck getting that through ethics approval!) so you have to rely on information collected from patients who experience adverse reactions. The rationale is as follows: if enough patients from sufficiently-diverse populations experience similar reactions, that suggests the drug
may be a "common denominator" in their reactions, and
may bear some responsibility for the reactions. But this does not preclude other potential influences, any of which could be more responsible, nor does it preclude the possibility of combinatorial influence - drugs that are harmless on their own, but deadly together. It's really, really hard to draw strong inferences from this kind of limited data, so it's even more important to search for disconfirming evidence, instead of just trying to confirm our prior suspicions.
For example, the article mentions Prednisone as being taken by a 14-year-old cancer patient who later developed what appear to be symptoms of schizophrenia. What they didn't mention is that schizophrenia most often develops during the teenage years, often without any prior symptoms, and many different hallucinogenic drugs have been implicated in its onset. It's possible that the kid was predisposed to schizophrenia, and took another drug (such as marijuana, which is definitely implicated in schizophrenia and also used by cancer patients), but just didn't report it. Or maybe he just developed it because the hormonal changes caused by puberty altered his brain chemistry, and triggered what was already about to happen. Or maybe one of the other drugs he was taking for chemo helped trigger the symptoms; the point being, in this case, the presence of Prednisone might be purely coincidental, rather than influential.
Again, it's all correlational data so there's no way to say for certain, but considering the weight of evidence against the hypothesis, I'd say it's pretty unlikely that Prednisone could be causally responsible for Williams' actions. More to the point, he hasn't shown any of the other symptoms associated with schizophrenia, psychopathy, or antisocial behaviour; so, why would Prednisone only trigger one of the supposed reactions but not the others? I suspect this is more a reporter grasping at straws than a genuine explanation.
As an aside, even though the evidence to support my hypothesis isn't yet available, I personally think that Williams knows a lot more about his motivations than he's admitted to. The thing is, nobody's pushing him to reveal any more details, because they'd rather go with the simpler "crazy monster" explanation, for reasons I've previously stated and won't repeat. Articles like this supposedly support that, but the overwhelming amount of disconfirming evidence to the crazy-monster premise makes most of the medication arguments moot. Evidence supporting a false premise doesn't magically make a false premise valid.
***
I also want to point out (and avoid) an equivocation - calling Williams a "Homicidal Maniac." In clinical terms, this label makes absolutely no sense, because he hasn't shown any real symptoms of mania. He has admitted to homicide, no question, but to call him a "homicidal maniac," or "serial killer" for that matter, is an inappropriate description. The use of these terms invoke the common-use terms, which are roughly equivalent to "crazed killer." It's important not to confuse the two, especially considering all the so-called experts saying we have to "rewrite the book on serial killers" to accommodate Williams into the paradigm. A simpler possible explanation is that he
isn't a serial killer, but the public's continued use of the common-use term is actually convincing the so-called experts to rewrite the term to fit the public perception (which is pretty ridiculous in my opinion).
He's a murderer, no question, but the other labels are unproven and speculative, and largely the result of media generalizations purely for emotional effect.