@georgeshalgreen

The thing with this drug is it affects everyone but with varying impact – the damage appears to be based on historical data (within the family) or the personality of hte user playing an important role.

Undeniably, which is why the study in itself has merit.

However, that in itself doesn’t mean that we should draw full conclusions and thereby determine policy on the basis of a correlation without a clear distinction of the causal mechanism.

If we did things like that, we could (for example) see that interest rates and house prices move together and decide that the it is interest rates causing the house price change. This is of course why fully reducible theory is so important – it is impossible to look at data without using a framework to give it understanding, and just saying that the data “tells you something” without admitting the implicit framework you are using for causality is dishonest … it of course doesn’t help that A LOT of people do this.

I also have many friends who used a lot of drugs when young, or use even more recently. However, I can’t shake the feeling that these are the same people I knew before this happened – and that even if the drugs weren’t available their psychological profile would have ended up similar.

I find it hard to blame a substance when it is the individuals choices, social environment, and underlying physical and mental structure that help to determine outcomes. Drugs do have a feedback into this, but the question should be “how” – and the answer to this question is likely to be far more complicated than the aforementioned newspaper article suggests.

It’s difficult to blame cannabis or a correlation between the two but it makes you wonder…

Although we need a model of how psychiatric conditions develop before we can say anything there – just because people didn’t seem to have a condition when they were young doesn’t imply that they did not have an inherent propensity to develop a defined disorder.

The even harder thing is that “psychosis” is actually a continuous variable, but it is being modeled as a binary variable – hmmmm.

I think that is an interesting point, and I am glad that they used a time trial – so that they could look at the separate groups that used cannabis and those that didn’t before looking at the level of psychosis.

However, I’m not sure I really agree with that quote – would anyone with experience of cannabis really view it as self-medication? They are acting like it is either self-medication or that the drug causes it – but in reality there are other casual mechanisms.

The real alternate hypothesis for me is that people that are more likely to consume cannabis are more likely to suffer from psychosis later in life – fundamentally, the personality types and social situations of those people are causes, not the drug in itself. Testing this without doing personality tests and then splitting the groups means that we can’t differentiate though.

There was no evidence for self-medication effects since psychotic symptoms did not predict later cannabis use, they said.