That's still more hypothesis, not conclusion. "Self-completed questionnaire on use of cannabis before and during pregnancy." = Observational study = result can

That's still more hypothesis, not conclusion. "Self-completed questionnaire on use of cannabis before and during pregnancy." = Observational study = result can ONLY be a hypothesis. It's right there in the conclusion "may be associated" = "they do not know if this is true at all". Next step would be to run a double-blind clinical trial, and have an unknown-half of the subjects smoke pot and compare that to the control to test if their hypothesis is true or not. Until then, it's only conjecture.