"First, the use of cannabis and rates of psychotic symptoms were related to each other, independently of observed/non-observed fixed covariates and observed time dynamic factors (Table 2). Secondly, the results of structural equation modeling suggest that the direction of causation is that the use of cannabis leads to increases in levels of psychotic symptoms rather than psychotic symptoms increasing the use of cannabis. Indeed, there is a suggestion from the model results that increases in psychotic symptoms may inhibit the use of cannabis."

Source

Fergusson, D. M., Horwood, L. J., & Ridder, E. M. (2005). Tests of causal linkages between cannabis use and psychotic symptoms. Addiction (Abingdon, England), 100(3), 354–366. doi.org/10.1111/j.1360-0443.2005.01001.x