Welcome to the Schizophrenia Resource Centre

Welcome, this website is intended for healthcare professionals in EMEA with an interest in the treatment of schizophrenia. By clicking the link below you are declaring and confirming that you are a healthcare professional

You are here

The impact of cannabis use on age of onset and clinical characteristics in first-episode psychotic patients. Data from the Psychosis Incident Cohort Outcome Study (PICOS)

Journal of Psychiatric Research, 4, 47, pages 438 - 444


Cannabis use is frequent among first-episode psychosis (FEP) patients and has been associated with several clinical features. This study aimed in an FEP sample to determine whether cannabis use is associated with 1) a higher level of positive symptoms, a lower level of depression and a better premorbid adjustment, 2) an earlier age of onset, and a better premorbid IQ.

The study was conducted within the framework of the Psychosis Incident Cohort Outcome Study (PICOS), a multisite collaborative research on FEP patients who attended the psychiatric services in Veneto Region, Italy. Standardized instruments were used to collect sociodemographic, clinical, and drug use data.

A total of 555 FEP patients met the inclusion criteria, 517 of whom received an ICD-10 diagnosis of psychosis; 397 (55% males; mean age: 32yrs ± 9.5) were assessed. Out of these, 311 patients agreed to be interviewed on drug and alcohol misuse; 20.3% was positive for drug misuse: cannabis (19.0%), cocaine (3.9%), and hallucinogens (3.9%). Cannabis use was not associated with a higher level of positive symptoms, but correlated with less severe depressive symptoms. No relationship was observed between premorbid adjustment or IQ and cannabis use. FEP patients who used cannabis had an earlier age of onset than abstinent patients, even after adjusting for gender and diagnosis.

Our results suggest a possible causal role of cannabis in triggering psychosis in certain vulnerable subjects. Particular attention must be paid to this behaviour, because reducing cannabis use can delay or prevent some cases of psychosis.

Keywords: Psychosis, Age of onset, Cannabis, First episode, IQ.

1. Introduction

Cannabis use is frequent among first-episode psychosis (FEP) patients and has a significant impact on the onset, course, and treatment of this disease. Cannabis users are more likely to be male ( Kavanagh et al., 2004 ) and younger ( Mueser et al., 1992 ; Barnes et al., 2006 ). Moreover, lower educational level ( Mueser et al., 1992 ; Kavanagh et al., 2004 ; Barnes et al., 2006 ), higher rates of unemployment, less desirable living conditions ( Kavanagh et al., 2004 ), and unstable relationship status are common in these patients.

Cannabis use has been linked to a shorter duration of untreated psychosis (DUP) ( Morgan et al., 2006 ), more severe positive symptoms ( Wade et al., 2007 ), and less severe negative symptoms ( Pencer and Addington, 2003 ). Yet, the effects of regular cannabis use on mood vary: whereas one study reported less depressed mood in cannabis users than in nonusers ( Denson and Earleywine, 2006 ), another observed that cannabis use is associated with depression ( Degenhardt et al., 2003 ).

Among the clinical variables that characterize psychotic disorders ( Nasrallah et al., 2011 ), age of onset is widely accepted as bearing strong clinical and prognostic significance. Some studies ( Compton et al., 2009 ; De Hert et al., 2011 ; Galvez-Buccolini et al., 2012 ) have suggested that cannabis use decreases the age of onset of psychotic disorders, and a recent meta-analysis ( Large et al., 2011 ) reported that cannabis users have an earlier onset of psychosis than nonusers, even if the association between cannabis use and age of onset should be considered spurious as FEP users patients tend to be younger than nonusers ( Wade et al., 2007 ). Earlier age at onset is associated with poorer outcomes in schizophrenia and seems to reflect the severity of the underlying neuropathological process ( DeLisi, 1992 ).

In contrast, some authors ( Sevy et al., 2001 ) have reported higher IQ levels and a better premorbid adjustment in those cannabis users who have developed schizophrenia, which also have greater cognitive reserve and lower severity of illness ( Barnett et al., 2006 ). Taken together, these two observations suggest that cannabis might trigger psychosis earlier in individuals who otherwise have good prognostic features and that the age at onset may be due to the action of cannabis rather than reflect a more severe illness.

Based on this evidence, this study examined the clinical features and the age of onset of FEP patients who reported cannabis use versus nonusers. Specifically, we hypothesized that in a cohort of Italian FEP patients cannabis use is associated with:

  • 1. higher level of positive symptoms, lower level of depressive symptoms, and better premorbid adjustment at onset;
  • 2. earlier age of onset and better premorbid IQ.

2. Material and methods

2.1. Design of the study

Data were collected within the framework of Psychosis Incident Cohort Outcome Study (PICOS), a multisite collaborative research that examined incident cases of psychosis in contact with public mental health services in the Veneto Region, northeastern Italy ( Lasalvia et al., 2007 ; Bertani et al., 2012 ). Overall, 25 collaborating sites participated in PICOS, encompassing a catchment area of nearly 3.3 million inhabitants, one of the largest cohorts ever examined. Detailed information on the study design, sample recruitment, sample representativeness, and clinical assessment has been reported elsewhere ( Lasalvia et al., 2012 ).

2.2. Study settings

The participating PICOS sites were routine public community-based mental health services, operating within the Italian National Health Service (NHS). Psychiatric care in Veneto is delivered by the NHS through its Departments of Mental Health (DMHs), each of which has a geographically defined catchment area. Multidisciplinary teams that operate the DMHs provide a wide range of comprehensive and integrated programs for the local adult population, including inpatient care, day care, rehabilitation, outpatient care, home visits, 24-h emergency services, and residential facilities for long-term patients.

In Veneto, the majority of psychotic patients is treated within the public sector: a negligible fraction of psychotic patients in treated in private hospitals or in private practice alone, and it is standard practice for general practitioners to refer all psychosis cases to public mental health services ( Lasalvia et al., 2012 ). Standard care for FEP patients generally comprises personalized outpatient psychopharmacological treatment, combined with nonspecific supportive clinical management at the Community Mental Health Center level or, when required, at the patient's home ( Lasalvia et al., 2007 ).

2.3. Participants

Based on the screening method ( Jablensky et al., 1992 ), the initial target group for the study comprised people aged 15–54 years who were residents in the Veneto Region and had had first contact with any mental health service in the PICOS area during the index period (Jan 1, 2005–Dec 31, 2007), with evidence of the following: delusions, hallucinations, thought disorder, or negative symptoms of schizophrenia, irrespective of cause. The primary exclusion criterion was any previous presentation or treatment for psychotic illness, other than initiation of treatment for the current episode at an earlier stage—for example, initiation of antipsychotic therapy by a health professional before referral to a participating mental health service provider, usually within the past 3 months. Patients who refused to participate in the study were contacted an additional 3 times at monthly intervals.

Written informed consent, including permission to contact their relatives, was obtained by participant after receiving a complete description of the study.

2.4. Measures

Information on drug and alcohol use in the year prior to the first contact with the service for psychosis was collected using the CDAUS (Clinical Drug and Alcohol Use Scale, Mueser et al., 1995 ) (information was gathered from the subject and his relatives). A set of standardized instruments was used to collect other clinical information: an ad hoc schedule to collect patients' sociodemographic characteristics; the PANSS (Positive and Negative Symptoms Scale, Kay et al., 1987 ) for symptoms; the HAM-D (Hamilton Rating Scale for Depression, Hamilton, 1960 ) for depressive symptoms; the GAF (Global Assessment of Functioning, APA, 1994 ) for global functioning; the DAS-II (Disability Assessment Schedule-II, WHO, 1988 ) for disability in the social roles; the TIB (Test d’Intelligenza Breve, Sartori et al., 1997 ) (the Italian version of the NART, Nelson, 1982 ) for premorbid IQ; and the PSA (Premorbid Social Adjustment Scale, Foerster et al., 1991 ) for premorbid adjustment.

Duration of untreated psychosis (DUP) was defined as the interval (in months) between the onset of the first psychotic symptoms and the initiation of adequate psychopharmacological treatment ( Norman and Malla, 2001 ) (information of psychosis onset was obtained from interviews with the patient or a close relative and from clinical notes).

The diagnosis was confirmed 6 months after inclusion into the study using the IGC (Item Group Checklist) of SCAN (Schedule for Clinical Assessment in Neuropsychiatry, WHO, 1992 ), which allows one to rate information from case records, integrated with interviews with the patient case manager if needed.

All investigators underwent specific training on the use of the standardized instruments, and an interrater reliability session was organized to determine the consistency of evaluations between investigators. The interrater reliability for PANSS was 0.902 (Cronbach's alpha).

2.5. Statistical analysis

Categorical variables were compared by chi-square test or Fisher's exact test if cell frequencies <5. Continuous variables were compared between independent groups by Mann–Whitney (2 groups) or Kruskal–Wallis (>2 groups) test. All tests were bilateral at p < 0.05.

ANCOVA, adjusted for gender, was performed to examine differences in age of onset between cannabis users and the abstinent group. Subsequently, the same model was applied to adjust for the effect of the diagnosis. Finally, both covariates were included in the model. All analyses were performed using SPSS 17.0 for Windows.

3. Results

Five hundred fifty-five patients were referred to the participating services during the recruitment phase, 517 of whom received an ICD-10 diagnosis of psychosis after 6 months. Of these patients, two could not be approached due to severe psychopathology, one died before being interviewed, 48 were untraceable, and 10 had moved after first service contact (most of these had been seen in hospital emergency wards only). Among the remaining patients, 27 declined to be interviewed; treating clinicians did not permit another 17 to be interviewed, due to the patient's severe psychopathology, and 15 were not referred to the research staff, due to lack of clinicians' cooperation in the study (for more details, see Bertani et al., 2012 ). In case of refusal or untraceability, three attempts to interview the subjects were made, at a 3–4 weeks interval. Three hundred and 97 subjects (76.8%) completed the PICOS baseline assessments.

Information on illicit drug and alcohol use was collected at the baseline interview on drug and alcohol abuse habits in the 12 months prior to first contact with the psychiatric service; the reliability of these data was assessed using all available sources, including relatives and key professionals. Of the 397 subjects who completed the baseline clinical assessments, 311 patients agreed to be interviewed on drug and alcohol misuse (78.3%). No significant differences were noted with regard to sociodemographic or clinical characteristics between subjects who completed the CDAUS and those who did not (data available from the authors).

3.1. Drug and alcohol misuse in FEP patients

Sixty-one subjects (20.3%) reported having used at least one illicit drug in the year prior to first contact with the service. The most commonly used drug was cannabis: 57 subjects had used cannabis (19.0%), 43 of whom (75.4%) had used cannabis alone; the remaining 14 subjects (24.6%) used cannabis together with other substances.

Cocaine was the second most frequently used illicit drug (3.9%), and hallucinogens were used by 2.9% of patients. All other classes of drugs (opiates 1.3%, stimulants 0.3%, phencyclidine 0.0%, sedatives 0.3%) were not used frequently enough to warrant separate analyses.

Regarding alcohol, 22.8% of cannabis users were teetotallers, 40.4% consumed alcohol without problems, and 36.9% abused alcohol compared with 58.6%, 37.2%, and 4.2% of cannabis nonusers, respectively (p = 0.000).

3.2. Association between sociodemographic and clinical characteristics and cannabis use

The main sociodemographic characteristics of patients who completed the assessment on cannabis are reported in Table 1 . The proportion of males was significantly higher among cannabis users (89.5%) compared with abstinent patients (44.4%). The mean age at onset was 26.12 years (sd 6.18) among cannabis users, which was significantly lower versus nonusers (33.54, sd 10.07). There was a higher proportion of people aged under 25 years (57.1%) in the cannabis user group versus nonusers (25.9%) (Table 1 ).

Table 1 Socio-demographic and clinical characteristics of the study sample stratified by cannabis use (abstinent patients vs users) (n = 296).

  Abstinent, N = 239 N (%) Cannabis – some evidence of use (N = 57), N (%) p-Value
Male 106 (44.4%) 51 (89.5%) 0.000a
Female 133 (55.6%) 6 (10.5%)
Age at onset (years) Mean (sd) 33.54 (sd 10.07) 26.12 (sd 6.18) 0.000b
Age of onset (years)   Missing = 1 0.000a
16–25 62 (25.9%) 32 (57.1%)
26–35 78 (32.9%) 17 (30.4%)
36–54 99 (41.4%) 7 (12.5%)
Italian 195 (81.6%) 52 (91.2%) Nsa
Non-Italian 44 (18.4%) 5 (8.8%)
Educational level Missing = 26 Missing = 4 Nsa
Medium-low 107 (50.2%) 24 (45.3%)
High 106 (49.8%) 29 (54.7%)
Employment status Missing = 28 Missing = 5 Nsa
Employed 93 (44.1%) 19 (36.5%)
Unemployed 74(35.1%) 25 (48.1%)
Economically inactive 44 (20.9%) 8 (15.4%)
Living condition Missing = 21 Missing = 6 0.000a
Alone or with partner 97 (44.5%) 3 (5.9%)
With the family of origin 115 (52.8%) 43 (84.3%)
Other 6 (2.8%) 5 (9.8%)
Relationship status Missing = 34 Missing = 11 0.001a
Single 132 (64.4%) 42 (91.3%)
Stable relationship 58 (28.3%) 3 (6.5%)
Other 15 (7.3%) 1 (2.2%)
Schizophrenia 49 (20.5%) 12 (21.1%) Nsa
Non-affective psychosis 132 (55.2%) 39 (66.7%)
Affective psychosis 58 (24.3%) 7 (12.3%)
DUP (Months) Missing = 31 Missing = 10 Nsb
Mean (sd) 7.70 (sd 18.38) 4.64 (sd 9.68)
Median 1.00 1.00

a Chi-square of Fisher test.

b M–W test.

Among cannabis users, 84.3% was still living with the family of origin, a proportion significantly higher than in the abstinent group (52.8%). Not being involved in a stable relationship was the most common hallmark (91.3%) of cannabis users compared with the abstinent group (64.4%). The majority of cannabis users was Italian (94.2%). Educational level and employment status did not differ significantly between cannabis users and nonusers ( Table 1 ).

There were no significant differences in diagnosis or DUP between cannabis users and nonusers.

3.3. Association between psychopathology, premorbid adjustment, and IQ and cannabis use

No significant differences between cannabis users and nonusers were found with regard to the various PANSS dimensions. On examination of individual PANSS items, however, we observed that cannabis users had more severe poor impulse control and disturbances of volition score and lower levels of depression with respect to nonusers, the latter of which was confirmed by the HAM-D score, which was significantly lower in cannabis users (16.73, sd 7.56 vs 19.25, sd 8.54; Mann–Whitney, p = 0.033) ( Table 2 ).

Table 2 Clinical characteristics of the study sample stratified by cannabis use (abstinent patients vs users) (N = 296).

  Abstinent (N = 239) mean (sd) Cannabis – some evidence of use (N = 57) mean (sd) p-Value M–W test
PANSS dimensions
Positive 3.05 (1.00) 3.29 (1.07) ns
Negative 2.45 (1.34) 2.43 (1.29) ns
General Psychopathology 2.63 (0.73) 2.59 (0.71) ns
Total 2.69 (0.71) 2.72 (0.72) ns
PANSS items
Depression 3.16 (1.77) 2.51 (1.56) 0.013
Impulsivity 2.00 (1.47) 2.59 (1.70) 0.007
Disturbance of volition 1.85 (1.43) 2.20 (1.31) 0.004
Hamilton Missing = 4 Missing = 1 0.033
19.25 (8.54) 16.73 (7.56)
IQ Missing = 109 Missing = 22 ns
103.36 (6.40) 103.71 (5.74)
PSA Missing = 79 Missing = 15  
Social accessibility/childhood 0.32 (0.22) 0.28 (0.19) ns
School childhood 0.27 (0.17) 0.29 (0.16) ns
Social accessibility/adolescence 0.33 (0.20) 0.29 (0.18) ns
School adolescence 0.32 (0.18) 0.38 (0.19) ns
GAF Missing = 1   ns
39.15 (10.88) 38.60 (11.10)
DAS Missing = 33 Missing = 9 ns
2.03 (1.21) 2.00 (1.18)

No differences in premorbid adjustment (social and school functioning, both considered during childhood and adolescence) were noted between cannabis users and abstinent subjects. School functioning during childhood and adolescence was significantly worse in males versus females (Mann–Whitney test; childhood p = 0.049; adolescence p = 0.008). Due to the gender effect on this variable, the sample was stratified by gender and analyzed to identify differences in premorbid adjustment between cannabis users and nonusers in each group; no association was found (data available from the authors).

There were no differences between cannabis users and abstinent patients regarding premorbid IQ, global functioning, or disability at onset ( Table 2 ).

3.4. Effect of gender and diagnosis on the association between age of onset and cannabis use

Gender appeared to be a confounder of cannabis use and age of onset, because males were more frequently cannabis users ( Table 1 ) and had a significantly lower age of onset versus females (30.22 sd 8.94 vs 34.28 sd 10.25; Mann–Whitney, p = 0.000). By using gender-adjusted ANCOVA (reference category: male), we confirmed the significantly lower age of onset in cannabis users compared with the abstinent group [F(1,292) = 17.18, p < 0.001].

Diagnosis had a significant effect on the age of onset (schizophrenia 31.05 sd 10.90, non affective non schizophrenia psychosis 31.19 sd 9.35, affective psychosis 35.58 sd 9.61; ANOVA, p = 0.006), but by Bonferroni's post hoc comparison, age of onset for schizophrenia patients did not differ from non affective, non schizophrenia psychosis patients (p > 0.05). Consequently, the diagnosis was dichotomized as affective versus non affective psychosis. By using ANCOVA, adjusted for dichotomized diagnosis (reference category: non affective psychosis), we detected a significantly lower age of onset in cannabis users compared with abstinent subjects [F(1,292) = 24.95, p < 0.001].

Finally, both covariates were introduced into the ANCOVA model, and the main effect of cannabis use on age of onset remained significant [F(1,291) = 16.29, p < 0.001] ( Table 3 ).

Table 3 ANCOVA adjusted means for gender, diagnosis (NAP vs AP) and both covariates.

Age of onset Never used (N = 239) Cannabis – some evidence of use (N = 57) p-Value ANCOVA
Adjusted means (95% CI) for gender 33.30 (32.09–34.52) 27.13 (24.52–29.74) <0.001
Adjusted means (95% CI) for diagnosis 34.42 (33.03–35.81) 28.28 (24.55–32.00) <0.001
Adjusted means (95% CI) for gender and diagnosis 34.13 (32.66–35.60) 26.57 (21.17–31.97) <0.001

4. Discussion

4.1. Strength and limitations of the study

This study has several strengths over previous reports. The research was conducted on an epidemiological catchment-area based cohort of FEP patients, reducing the likelihood of selection bias. Further, the study used standardized assessments on drug use, sociodemographics and psychopathology, and operational diagnostic criteria for clinical diagnoses. Finally, the sample size allowed us to analyze cannabis-only users, eliminating the confounding effect of polydrug use.

However, our findings should be interpreted with regard to certain methodological limitations. This study only included subjects who had contact with psychiatric services, excluding all subjects who did not, such as those with access to addiction services. Also, the data on substance use relied on patients' self-reports and were not corroborated by biological markers of drug use; although earlier studies have suggested that there is a tendency to underreport in substance use studies, participants are generally willing to give accurate accounts ( Harrison et al., 1993 ).

Further, although information on the specific diagnosis (e.g., use, abuse, or dependence) was available, the subgroup of FEP patients with abuse or dependence was too small for a separate analysis; consequently, we could not examine the differences between sporadic and heavy substance users even if a dose response effect has recently been described ( Moore et al., 2007 ). We also observed that cannabis users have higher rates of alcohol consumption than nonusers, limiting the significance of models of drug-induced psychosis. Moreover, nicotine use data was unavailable in our sample, the use of which is highly prevalent in people with psychosis compared with the general population ( Volkow, 2009 ). Finally, the instrument used to collect information about the use of drug did not allow to collect information on frequency and length of use of cannabis beyond the previous year.

4.2. Prevalence of cannabis use

This study is the first Italian survey on illicit drug use in a population of FEP patients. In our sample, nearly one-quarter of FEP subjects used an illicit drug in the 12 months prior to first contact with a psychiatric service. As expected, cannabis was the most widely used illicit drug, followed by cocaine and hallucinogens.

The annual prevalence of cannabis use was more than 2 times higher in FEP subjects than in the general population of the same catchment area (19% versus 9.1%, p < 0.0001), as shown by data from the IPSAD-Italia 2005 study ( Italian Population Survey on Alcohol and Drugs, EMCDDA, 2006 ). Moreover, male psychotic patients used all substances significantly more often than their healthy counterparts—cannabis 31.7% versus 10.5% (p < 0.0001), cocaine 6.8% versus 2.0% (p < 0.0001), and hallucinogens 5.6% versus 0.6% (p < 0.0001). This behaviour was not observed in the female cohort of the FEP sample.

We observed a significant difference between FEP patients and the general population among subjects aged 16–25 years (this age range is considered at higher risk for drug use) with regard to cannabis (33.3% vs 22.9%; p < 0.05) and hallucinogen use (7.3% vs 1.5%; p < 0.01) but not cocaine. The high rate of comorbidities in FEP patients highlights the need to ascertain to which extent drug use might be a precipitating factor of psychosis and how much it is influenced by local social contexts ( Carrà and Johnson, 2009 ). A comparison between studies in various countries and contexts (urban, rural, or mixed) is necessary to clarify these aspects.

Studies sample selection methods also vary: very few studies have considered FEP patients ( Barnes et al., 2006 ; Mazzoncini et al., 2009 ) and few have compared the prevalence rates of drug use in FEP with the general population, thus do not allowing an analysis of the relative risk for psychotic patients using illicit drugs ( Mazzoncini et al., 2009 ).

For example, the relative risk of an FEP patient to use cannabis is 4.6 in the UK (data from the AESOP study, Mazzoncini et al., 2009 ) versus 2.1 in this study cohort, supporting the hypothesis that although higher rates of illicit drug use appear to be a constitutive clinical feature of psychosis, social determinants might also influence this behaviour.

The context differences in the AESOP study (urban area) and the PICOS study (mixed area) and the additional differences in social structure and illicit drug availability between the two studies might account for the increased relative risk in the UK. No conclusions can be drawn, and more comparative research is needed to clarify this issue.

4.3. Clinical correlates of cannabis use

As expected, a strong association between male gender and cannabis use was noted. Younger subjects were more likely to use any illicit drug and cannabis, a finding that is also consistent with the international literature ( Mueser et al., 1992 ; Barnes et al., 2006 ). We were unable to detect any significant difference between cannabis users and nonusers with regard to the other demographic characteristics, global functioning and disability at onset, indicating a lack of association between a poor social adjustment at the time of psychosis onset and drug use among psychotic subjects.

Moreover, in contrast with previous findings ( Sevy et al., 2001 ; Leeson et al., 2012 ), there was no correlation between cannabis use and premorbid functioning. Subjects who used cannabis still lived with their family of origin more frequently than nonusers. Thus, the inability to gain independence from the family of origin does not appear to be attributable to a social disadvantage but possibly to the younger age of cannabis users and to needs that are unmet by psychiatric services, resulting in an adjunctive burden for familial care givers.

Considering the psychopathological dimensions, cannabis users and nonusers did not differ in terms of negative or positive symptoms, contrary to what has been previously reported ( Wade et al., 2007 ). According to the “self-medication hypothesis” ( Khantzian, 1997 ), patients use cannabis to relieve psychiatric symptoms, such as positive symptoms, and to ameliorate negative symptoms. Notably, cannabis users had a lower level of depressive symptoms and higher scores on certain PANSS items, such as “disturbance of volition” and “impulsivity,” than nonusers. Additionally, negative symptoms cause hedonic and motivational deficits that can interfere with the drive and desire to obtain and consume illicit drugs.

A variation of the traditional self-medication hypothesis, the “alleviation of dysphoria model” ( Kolliakou et al., 2011 ), proposes that cannabis use is favoured by the pressure to alleviate dysphoric experiences, such as boredom and depression. The low level of depressive symptoms in our patients with substance comorbidities seems to confirm this model.

4.4. Age of onset and premorbid IQ

We found that cannabis users had an earlier age of onset, which remained significant after controlling for the effect of confounders, such as gender and diagnosis. Based on our findings and the existing literature ( Large et al., 2011 ), we may speculate that the earlier age of psychosis onset among cannabis users may be caused by a “toxic effect” of cannabis rather than by an intrinsically more severe illness.

These data appear to support a causal role of cannabis in precipitating psychosis, at least in individuals who otherwise are not at risk, perhaps through an interaction with genetic factors, such as the COMT gene ( Caspi et al., 2005 ), or by a disruption in brain development, which might especially take place during neurological maturation in early adulthood ( Trezza et al., 2008 ). Exposure to Δ9-tetrahydrocannabinol (THC), the principal psychoactive component of cannabis, can alter the dopamine system by stimulating the cannabinoid neurotransmission system, specifically in the prefrontal cortex and mesolimbic pathway. Dopamine sensitization can thus constitute a possible mechanism by which environmental factors, such as stress and cannabis, contribute to the risk of psychosis ( Collip et al., 2011 ).

4.5. Clinical implications and future directions for research

Considering the high rate of drug use among FEP patients and the possible negative consequences related to drug exposure (e.g., reduced compliance to treatment, worse clinical and social outcome, higher rates of relapse) ( De Girolamo et al., 2012 ), detecting and treating this dysfunctional behaviour should be a public health priority and an essential task for health services. To pursue this aim, prevention strategies and greater collaboration between psychiatric and addiction services are needed, such as the inclusion of drug abuse specialists within early intervention teams.

The high percentage of cannabis users among FEP patients indicates that specific attention should be paid to this behaviour, because reducing cannabis use can delay and prevent cases of psychosis. Patients who use cannabis have an earlier age of onset, suggesting a causal role of this substance in triggering psychosis, at least in some vulnerable subjects. If an onset of psychosis is inevitable, an additional two or three years of psychosis-free functioning could allow some given patients to achieve the developmental milestones of late adolescence and early adulthood (e.g., completing school, finding a job, being involved in a stable relationship, gaining independence from the family of origin), consequently limiting the long-term disability that is associated with psychosis.

Role of funding source

This study was supported by the Ricerca Sanitaria Finalizzata 2004, Giunta Regionale del Veneto with a grant to Prof. Mirella Ruggeri, Ricerca Sanitaria Finalizzata 2005, Giunta Regionale del Veneto with a grant to Dr. Antonio Lasalvia, and by the Fondazione CariVerona, who provided a 3-year grant to the WHO Collaborating Centre for Research and Training in Mental Health and Service Organization at the University of Verona, directed by Prof. Michele Tansella. We would like to acknowledge Dr Sabrina Molinaro (researcher at the Institute of Clinical Physiology of the CNR) for having provided us the IPSAD-Italia study data.


ST conceptualized the aims, performed the literature search wrote the first draft of this paper and the consecutive drafts were reviewed by MR, MT, AL, RM. AL contributed to conceiving the overall design of the study and to the interpretation of the data; CB and DC contributed to define the procedures for data analyses and performed the statistical analysis; ST, RM, KDS, MB, LL, GM, SB, DL, RR, FG, AU participated in data collection. MR and AL directed the entire project. All the authors revised the draft and approved the final version.

Conflict of interest

None of the authors have a conflict of interest.



Coordinating staff: M. Tansella M. Ruggeri, A. Lasalvia, M. Bertani, C. Bonetto, P. Brambilla, S. Tosato, D. Cristofalo, G. Marrella, S. Bissoli, C. Perlini.

Collaborating Sites: Bassano del Grappa: P. Tito, M. Lunardon, F. Gava, E. Borso, L. Grandina, M. Paliotto, L. Roggia. Thiene: A. Danieli, C. Poloni, M.R. Altiero, F. Piazza. Monteccchio M.: E. Ceccato, C. Busana, A. Campi, A. Zanconato. Vicenza 1 UO:P. Zamorani, R. Binotto, A. Caneva, E. Lazzarin, G. Zordan. Vicenza 2 UO:C. Dolce, G.B. Fanchin, C. Negro. Vicenza 3 UO:F. Gardellin, M. Crestale, L. Paiola, A. Sale. Pieve di Soligo: I. Morandin, E. Biondi, A. Cordella G. Favaretto, S. Geatti, P. Urbani. Treviso: M. De Rossi, G. Zanatta, J. Spessotto, R. Penelope, L. Grando, M. Sgnaolin, C. Tozzini, G. Visentin, L. Schiavon. Portogruaro:B. Gentile, M.G. Bolacchi, L. Marzotto, F. Moni, L. Rossi. San Donà di Piave: I. Amalric, C. Miceli, M.R. De Zordo, L. Ramon, S. Russo. Venezia:R. Rossi, G. Casagrande, V. De Nardo, A. Facchetti, F. Ramaciotti. Mirano: V. Marangon, G. Coppola, A. Marcolin, P. Meneghini, F. Sbraccia, C. Segato. Camposampiero:R. Riolo, L. Cappellari, M. Cutugno, L. Meneghetti, L. Longhin, B. Paoleschi. Cittadella:D. Scalabrin, L. Antonello, A. Purgato, G. Santucci, C. Tosin, R. Volpato, R. Zurlo. Padova 2 Serv.: M. Zucchetto, M. Pedron, S. Pinton, M. Benetazzo. Padova 3 Serv.:C. Cremonese, L. Pavan, M. Semenzin, L. Sifari, F. Zorzi. Rovigo: M.M. Martucci, N. Magno, G. Meloni, E. Toniolo. Adria: M. Pavanati, E. Destro, L. Finotti. Verona 1 Serv.: R. Fiorio, A. Marsilio, N. Pedrocco, P. Pollola. Verona 2 Serv.: L. Lazzarotto, F. Nosè, P. Rossin, V. Vivenza. Verona 3 Serv.: A. Lasalvia, M. Bertani, S. Bissoli, S. Tosato, K. De Santi, G. Marrella, R. Mazzoncini, M. Ruggeri. Verona 4 Serv.: A. Urbani, L. Bianchi, G. Carcereri, L. Lunardi, G. Migliorini, G. Perdonà, C. Piazza. Legnago: D. La Monaca, G. D’Agostini I. Boggian, G. Piccione, E. Saladini. Domegliara:F. Gomez, S. Frazzingaro. Isola della Scala: S. Nicolaou, L. Cordioli, G. Bertolazzi, V. Pagliuca. Villa Santa Chiara: M. Abate, M. Bortolomasi, M. Giacopuzzi, M. Segala. Villa Santa Giuliana: F. De Nardi, F. Basetto, C. Bernardis, A. Bezzetto, M. Santi.


  • American Psychiatric Association, 1994 American Psychiatric Association. Diagnostic and statistical manual of mental disorder. 4th ed. (, 1994) (DSM-IV). Washington DC
  • Barnes et al., 2006 T.R. Barnes, S.H. Mutsatsa, S.B. Hutton, H.C. Watt, E.M. Joyce. Comorbid substance use and age at onset of schizophrenia. British Journal of Psychiatry. 2006;188:237-242
  • Barnett et al., 2006 J.H. Barnett, C.H. Salmond, P.B. Jones, B.J. Sahakian. Cognitive reserve in neuropsychiatry. Psychologie Medicale. 2006;36:1053-1064
  • Bertani et al., 2012 M. Bertani, A. Lasalvia, C. Bonetto, S. Tosato, D. Cristofalo, S. Bissoli, et al. The influence of gender on clinical and social characteristics of patients at psychosis onset: a report from the Psychosis Incident Cohort Outcome Study (PICOS). Psychologie Medicale. 2012;42(4):769-780
  • Carrà and Johnson, 2009 G. Carrà, S. Johnson. Variations in rates of comorbid substance use in psychosis between mental health settings and geographical areas in the UK. Social Psychiatry and Psychiatric Epidemiology. 2009;44:429-447
  • Caspi et al., 2005 A. Caspi, T.E. Moffitt, M. Cannon, J. McClay, R. Murray, H. Harrington, et al. Moderation of the effect of adolescent-onset cannabis use on adult psychosis by a functional polymorphism in the catechol-O-methyltransferase gene: longitudinal evidence of a gene X environment interaction. Biological Psychiatry. 2005;57(10):1117-1127
  • Collip et al., 2011 D. Collip, R. van Winkel, O. Peerbooms, T. Lataster, V. Thewissen, M. Lardinois, et al. COMT Val158Met-stress interaction in psychosis: role of background psychosis risk. CNS Neuroscience & Therapeutics. 2011;17(6):612-619
  • Compton et al., 2009 M.T. Compton, M.E. Kelley, C.E. Ramsay, M. Pringle, S.M. Goulding, M.L. Esterberg, et al. Association of pre-onset cannabis, alcohol and tobacco use with age of onset of prodrome and age of onset of psychosis in first-episode patients. American Journal of Psychiatry. 2009;166(1):1251-1257
  • De Girolamo et al., 2012 G. De Girolamo, J. Dagani, R. Purcell, A. Cocchi, P.D. McGorry. Age of onset of mental disorders and use of mental health services: needs, opportunities and obstacles. Epidemiology and Psychiatric Sciences. 2012;21:47-57
  • De Hert et al., 2011 M. De Hert, M. Wampers, T. Jendricko, T. Franic, D. Vidovic, N. De Vriendt, et al. Effects of cannabis use on age at onset in schizophrenia and bipolar disorder. Schizophrenia Research. 2011;126(1–3):270-276
  • Degenhardt et al., 2003 L. Degenhardt, W. Hall, M. Lynskey. Exploring the association between cannabis use and depression. Addiction. 2003;98(11):1493-1504
  • DeLisi, 1992 L.E. DeLisi. The significance of age of onset for schizophrenia. Schizophrenia Bulletin. 1992;18:209-215
  • Denson and Earleywine, 2006 T.F. Denson, M. Earleywine. Decreased depression in marijuana users. Addictive Behaviors. 2006;31(4):738-742
  • EMCDDA, 2006 Osservatorio Europeo sulle Droghe e le Tossicodipendenze (EMCDDA). Studio IPSAD-Italia 2005 sull'uso di sostanze nella popolazione generale. (, 2006)
  • Foerster et al., 1991 A. Foerster, S. Lewis, M. Owen, R. Murray. Pre-morbid adjustment and personality in psychosis: effects of sex and diagnosis. British Journal of Psychiatry. 1991;158:171-176
  • Galvez-Buccolini et al., 2012 J.A. Galvez-Buccolini, A.C. Proal, V. Tomaselli, M. Trachtenberg, C. Coconcea, J. Chun, et al. Association between age of onset of psychosis and age at onset of cannabis use in non-affective psychosis. Schizophrenia Research. 2012;
  • Hamilton, 1960 M. Hamilton. A rating scale for depression. Journal of Neurology, Neurosurgery, & Psychiatry. 1960;23:56-62
  • Harrison et al., 1993 E.R. Harrison, J. Haaga, T. Richards. Self-reported drug use data: what do they reveal?. American Journal of Drug and Alcohol Abuse. 1993;19:423-441
  • Jablensky et al., 1992 A. Jablensky, N. Sartorius, G. Ernberg, M. Anker, A. Korten, J.E. Cooper, et al. Schizophrenia: manifestations, incidence and course in different cultures. A World Health Organization ten-country study. Psychological Medicine Monograph Supplement. 1992;20:1-97
  • Kavanagh et al., 2004 D.J. Kavanagh, G. Waghorn, L. Jenner, D.C. Chant, V. Carr, M. Evans, et al. Demographic and clinical correlates of comorbid substance use disorders in psychosis: multivariate analyses from an epidemiological sample. Schizophrenia Research. 2004;66(2–3):115-124
  • Kay et al., 1987 S.R. Kay, A. Fiszbein, L.A. Opler. The positive and negative syndrome scale (PANSS) for schizophrenia. Schizophrenia Bulletin. 1987;13(2):261-276
  • Khantzian, 1997 E.J. Khantzian. The self-medication hypothesis of substance use disorders: a reconsideration and recent applications. Harvard Review of Psychiatry. 1997;4(5):231-244
  • Kolliakou et al., 2011 A. Kolliakou, C. Joseph, K. Ismail, Z. Atakan, R.M. Murray. Why do patients with psychosis use cannabis and are they ready to change their use?. International Journal of Developmental Neuroscience. 2011;29(3):335-346
  • Large et al., 2011 M. Large, S. Sharma, M.T. Compton, T. Slade, O. Nielssen. Cannabis use and earlier onset of psychosis: a systematic meta-analysis. Archives of General Psychiatry. 2011 Jun;68(6):555-561
  • Lasalvia et al., 2007 A. Lasalvia, B. Gentile, M. Ruggeri, A. Marcolin, F. Nosè, L. Cappellari, et al. Heterogeneity of the Departments of Mental Health in the Veneto Region ten years after the National Plan 1994-96 for Mental Health. Which implication for clinical practice? Findings from the PICOS Project. Epidemiol Psichiatr Soc. 2007;16(1):59-70
  • Lasalvia et al., 2012 A. Lasalvia, S. Tosato, P. Brambilla, M. Bertani, C. Bonetto, D. Cristofalo, et al. Psychosis Incident Cohort Outcome Study (PICOS). A multisite study of clinical, social and biological characteristics, patterns of care and predictors of outcome in first-episode psychosis. Background, methodology and overview of the patient sample. Epidemiology and Psychiatric Sciences. 2012;21(3):281-303
  • Leeson et al., 2012 V.C. Leeson, I. Harrison, M.A. Ron, T.R. Barnes, E.M. Joyce. The effect of cannabis use and cognitive reserve on age at onset and psychosis outcomes in first-episode schizophrenia. Schizophrenia Bulletin. 2012;38(4):873-880
  • Mazzoncini et al., 2009 R. Mazzoncini, K. Donoghue, J. Hart, C. Morgan, G.A. Doody, P. Dazzan, et al. Illicit substance use and its correlates in first episode psychosis. Acta Psychiatrica Scandinavica. 2009;121:351-358
  • Moore et al., 2007 T.H. Moore, S. Zammit, A. Lingford-Hughes, T.R. Barnes, P.B. Jones, M. Burke, et al. Cannabis use and risk of psychotic or affective mental health outcomes: a systematic review. Lancet. 2007;370(9584):319-328
  • Morgan et al., 2006 C. Morgan, R. Abdul-Al, J.M. Lappin, P. Jones, P. Fearon, M. Leese, et al. Clinical and social determinants of duration of untreated psychosis in the AESOP first-episode psychosis study. British Journal of Psychiatry. 2006;189:446-452
  • Mueser et al., 1992 K.T. Mueser, P.R. Yarnold, A.S. Bellack. Diagnostic and demographic correlates of substance abuse in schizophrenia and major affective disorder. Acta Psychiatrica Scandinavica. 1992;85(1):48-55
  • Mueser et al., 1995 K.T. Mueser, P. Nishith, J.I. Tracy, J. DeGirolamo, M. Molinaro. Expectations and motives for substance use in schizophrenia. Schizophrenia Bulletin. 1995;21(3):367-378
  • Nasrallah et al., 2011 H. Nasrallah, R. Tandon, M. Keshavan. Beyond the facts in schizophrenia: closing the gaps in diagnosis, pathophysiology, and treatment. Epidemiology and Psychiatric Sciences. 2011;20:317-327
  • Nelson, 1982 H.E. Nelson. National Adult Heading Test (NART): test manual. (NFER-Nelson, Windsor, 1982)
  • Norman and Malla, 2001 R.M. Norman, A.K. Malla. Duration of untreated psychosis: a critical examination of the concept and its importance. Psychologie Medicale. 2001;31(3):381-400
  • Pencer and Addington, 2003 A. Pencer, J. Addington. Substance use and cognition in early psychosis. Journal of Psychiatry & Neuroscience. 2003;28(1):48-54
  • Sartori et al., 1997 G. Sartori, L. Colombo, G. Vallar, M.L. Rusconi, A. Pinarello. Test di Intelligenza Breve per la valutazione del quoziente intellettivo attuale e pre-morboso. Giornale dell'Ordine Nazionale degli Psicologi Supplement. 1997;
  • Sevy et al., 2001 S. Sevy, D.G. Robinson, S. Holloway, J.M. Alvir, M.G. Woerner, R. Bilder, et al. Correlates of substance misuse in patients with first-episode schizophrenia and schizoaffective disorder. Acta Psychiatrica Scandinavica. 2001;104(5):367-374
  • Trezza et al., 2008 V. Trezza, V. Cuomo, L.J. Vanderschuren. Cannabis and the developing brain: insights from behavior. European Journal of Pharmacology. 2008;585(2–3):441-452
  • Volkow, 2009 N.D. Volkow. Substance use disorders in schizophrenia–clinical implications of comorbidity. Schizophrenia Bulletin. 2009;35:469-472
  • Wade et al., 2007 D. Wade, S. Harrigan, P.D. McGorry, P.M. Burgess, G. Whelan. Impact of severity of substance use disorder on symptomatic and functional outcome in young individuals with first-episode psychosis. The Journal of Clinical Psychiatry. 2007;68(5):767-774
  • WHO, 1988 WHO. Disability Assessment Schedule-II (DAS). (World Health Organization, Geneva, 1988)
  • World Health Organization, 1992 World Health Organization. Schedule for clinical assessment in neuropsychiatry, SCAN. (WHO, Geneva, 1992)


a Department of Public Health and Community Medicine, Section of Psychiatry, University of Verona, 37134 Verona, Italy

b Department of Mental Health, NHS Local Health Authority, Legnago (VR), Italy

c Department of Mental Health, NHS Local Health Authority, Camposampiero (PD), Italy

d Department of Mental Health, NHS Local Health Authority, Vicenza, Italy

e Department of Mental Health, NHS Local Health Authority, Dolo-Mirano (VE), Italy

Corresponding author. Tel.: +39 045 8124441; fax: +39 045 8027498.