Cannabis use is associated with increased risk for psychotic-like experiences (PLEs) and psychotic disorders. It remains unclear whether this relationship is causal or due to confounding.

A total of 1929 young adults aged 18–30 years participated in a nationwide internet-based survey in The Netherlands and gave information on demographics, substance use and parental psychiatric illness and completed the Community Assessment of Psychic Experiences (CAPE).

Cigarette smoking and cannabis use were equally strongly associated with the frequency of PLEs in a fully adjusted model (β = 0.098 and 0.079 respectively, p < 0.05). Cannabis use was associated with distress from PLEs in a model adjusted for an elaborate set of confounders excluding smoking (β = 0.082, p < 0.05). However, when cigarette smoking was included in the model, cannabis use was not a significant predictor of distress from PLEs. Cigarette smoking remained associated with distress from PLEs in a fully adjusted model (β = 0.107, p < 0.001).

Smoking is an equally strong independent predictor of frequency of PLEs as monthly cannabis use. Our results suggest that the association between moderate cannabis use and PLEs is confounded by cigarette smoking.

Although the association between cannabis use and a wide range of psychiatric symptoms is fairly well established, it is not clear whether cannabis use is also a risk factor for general mental health problems at secondary school.

A total of 10 324 secondary school children aged 11–16 years, participating in an ongoing Public Health Service School Survey, gave information on demographics, substance use, school factors and stressful life events and completed the Strengths and Difficulties Questionnaire (SDQ).

Cannabis use in the past month was associated with a clinically relevant score on the SDQ [unadjusted odds ratio (OR) 4.46, 95% confidence interval (CI) 3.46–5.76]. Other risk factors associated with poor psychosocial functioning were: a low level of education, alcohol use, cigarette smoking, hard drug use, frequent truancy, an unfavourable school evaluation, feeling unsafe at school, being victimized, frequent absence due to illness, a mentally ill parent, molestation by a parent, financial problems and feeling distressed by an adverse event. In a full model adjusting for these risk factors, cannabis was not significantly associated with mental health problems, although an association at trend level was apparent. Of these risk factors, regular alcohol use, cigarette smoking, hard drug use, frequent truancy, an unfavourable school evaluation and frequent absence due to illness were also associated with cannabis use.

The association between cannabis use and poor psychosocial functioning in adolescence is due, at least in part, to confounding by other risk factors. Thus, cannabis use can best be viewed as an indicator of risk for mental health problems in adolescence.

Cannabis use is associated with psychosis and a range of subclinical psychiatric symptoms. The strength of this association depends on dosage and age at first use. The current study investigates whether level of cannabis exposure and starting age are associated with specific profiles of subclinical symptoms.

We collected cross-sectional data from a young adult population sample by administering an online version of the Community Assessment of Psychic Experiences (CAPE). Cannabis exposure was quantified as the amount of Euros spent on cannabis per week and the age of initial cannabis use. The primary outcome measure was the odds ratio (OR) to belong to the highest 10% of scores on the total CAPE and the positive-, negative- and depressive symptom dimensions.

In 17 698 adolescents (mean age 21.6, s.d.=4.2 years), cannabis use at age 12 years or younger was strongly associated with a top 10% score on psychotic experiences [OR 3.1, 95% confidence interval (CI) 2.1–4.3] and to a lesser degree with negative symptoms (OR 1.7, 95% CI 1.1–2.5). The OR of heavy users (>€25/week) for negative symptoms was 3.4 (95% CI 2.9–4.1), for psychotic experiences 3.0 (95% CI 2.4–3.6), and for depressive symptoms 2.8 (95% CI 2.3–3.3).

Early start of cannabis use is strongly associated with subclinical psychotic symptoms and to a lesser degree with negative symptoms, while smoking high amounts of cannabis is associated with increased levels of all three symptom dimensions: psychotic, negative and depressive. These results support the hypothesis that the impact of cannabis use is age specific.