Adolescence is a time of significant brain development, and this maturation process is thought to represent a time of heightened vulnerability to the adverse effects of environmental risk factors such as cannabis. While the association of adolescent cannabis use and subsequent psychosis and depression has received considerable attention in the existing literature, very few prospective studies have focused on the association of early cannabis exposure with severe self-harm or suicide, bipolar disorder, or anxiety disorders. Moreover, though there are reports focusing on the prognostic effect of cannabis use in patients at high risk for psychosis, populationbased studies examining the prognosis of non-treatment-seeking adolescents with both psychotic-like experiences (PLEs) and cannabis exposure are scarce.

The aim of this thesis was to examine the association of early cannabis use with severe self-harm requiring medical attention, bipolar disorder, and depressive or anxiety disorders (studies I–III). Secondly, the prognosis of adolescents with PLEs with or without cannabis exposure was examined with respect to several psychiatric sequelae (study IV). A prospective general population-based Northern Finland 1986 Birth Cohort (N=9432) was utilized. Data on substance use including cannabis use and early psychopathology including PLEs were gathered in a field study in 2001– 02 when the participants were aged 15–16 years. The participants were followed for 18 years (until the year 2018, when they were 33 years). Data on diagnoses made in clinical practice from nationwide registers were utilized as outcome measures.

Early cannabis use was found to be associated with severe self-harm (Study I). While an independent association of this exposure was seen with depressive/anxiety disorders (Study II), the association with a bipolar disorder attenuated to nonsignificance after adjusting for types of other substance use (Study III). In study IV, participants with both PLEs and early cannabis exposure were found to display a greater odds for adverse psychiatric sequelae than participants with only PLEs.

In summary, the findings support the proposal that early cannabis use is an adverse prognostic marker for other psychiatric sequelae in addition to psychosis.