Or
3. Some of each.
4. Some unknown other factor causes both.
Yep, as is often said:
Correlation does not equal causation. Particularly where the correlation is not especially strong.
And without a proven mechanism, that’s all we’ve got - a statistical correlation. Just like those news stories you see all that time when they say that people who consume X or do Y have a 10% higher chance of developing disease Z - a 10% correlation is very weak, bordering on statistical noise. Having said that, it wouldn’t at all surprise me that long term use of any psychoactive substance could have deleterious effects on people’s mental state, particularly those already prone to mental instability.