Since cannabis initiation can begin early, it is essential to have early interventions in place to prevent adverse outcomes in later adolescence and adulthood for those youth who do initiate early . Effective early interventions include multidimensional family therapy, motivational enhancement therapy, and cognitive behavior therapy . However, despite the potentially detrimental effects of cannabis use during early adolescence, very few young people receive treatment . Barriers to treatment include a lack of treatment seeking due to embarrassment or due to a lack of felt need to discuss these issues or motivation to work on them . Youth might be more inclined to seek treatment when substance use is interfering with their mental health, educational attainment, or family, peer and romantic relationships . Since parental encouragement and support can be a key factor in youth treatment seeking , awareness and promotion interventions should target parents. In addition, adolescents may be open to discussing cannabis use with health care practitioners , but many may choose not to discuss it without probing; thus the onus of early identification among adolescents may therefore fall on service providers, making systematic screening important. Several limitations should be kept in mind when interpreting the findings. Notably, the data were self-reported, with retrospective estimates of the age of onset, which may have affected their validity. In addition, the youth in this current sample only reported past year symptom endorsement on the GAIN-SS. It is unknown whether these symptoms existed at age of initiation or whether symptoms progressed or improved from age of initiation. In the absence of a time course analyses , the directionality or bidirectionality of these associations is unclear, i.e., it remains to be determined whether early cannabis use represents a unique risk factor for developing problematic mental health and substance use behavior or whether is a marker of shared  or common underlying risk etiology . These findings support the conclusion that early cannabis initiation is clinically meaningful marker of risk across mental and behavioral health. The initiation of cannabis use at an earlier age than typically considered is associated with particular risk for more complex, concurrent mental and behavioral health concerns.

Cannabis use and other substance use, as well as broader mental health challenges should be systematically assessed from an early age, while prevention initiatives designed for younger youth are also called for. Service providers and researchers are encouraged to consider the potential complex comorbidities and polysubstance use patterns of youth who initiate cannabis use early, mobile vertical rack as well as ways to address factors such as trauma and their role in the early initiation. Individuals suffering from schizophrenia and related disorders have a high prevalence of substance use disorders . Epidemiological studies consistently report a higher lifetime prevalence rate of smoking  and cannabis use in patients with schizophrenia compared to the general population . Patients with schizophrenia who use cannabis present higher relapse rates, longer hospital admissions, and more severe positive symptoms than individuals who discontinue cannabis use and those who are nonusers . The association of comorbid substance use with negative symptoms has received less interest, although negative symptoms have a strong impact on functional outcome and remain difficult to treat . The negative symptoms of schizophrenia include blunted affect, alogia, asociality, avolition, and anhedonia , which can be organized along the two dimensions diminished expression and apathy . Negative symptoms be classified as primary or secondary . Primary negative symptoms are thought to be intrinsic to schizophrenia, while secondary negative symptoms can be caused by positive symptoms, depression, side effects and substance abuse. The hypothesis of cannabis as a potential cause for secondary negative symptoms was initially based on observations of an amotivational syndrome in otherwise healthy individuals who are chronic cannabis users . Cannabis may also exert an amotivational effect in patients with schizophrenia and thus be a cause for secondary negative symptoms . Following this argument, one would expect more prominent negative symptoms in cannabis-using patients with schizophrenia than in nonusers. This pattern has yet to be confirmed by the literature. An earlier meta-analysis by Potvin and colleagues studied the association between substance use disorders and negative symptoms  and found a lower severity of negative symptoms across different substances of abuse. In a subgroup analysis, fewer negative symptoms were found in patients with cannabis use disorders than in those without cannabis use disorders. However, in a more recent meta-analysis including more studies and excluding patients with previous substance use, Large and colleagues did not find an association between current substance use  and negative symptoms . In summary, the existing meta-analyses do not suggest a higher level of negative symptoms in patients with schizophrenia who use cannabis, but they were based on a limited number of available studies that did not allow us to distinguish between patients using cannabis only and patients using cannabis as a main drug of choice. Moreover, while data suggest a limited effect of cannabis discontinuation on negative symptoms , to the best of our knowledge, there is no systematic review of negative symptoms in patients recently abstaining from cannabis use. It is important to note that individuals with schizophrenia consume cannabis almost exclusively in the context of nicotine use .

Most research on the association between symptoms and nicotine use has not focused on negative symptoms but on cognitive dysfunction . One recent metaanalysis reported more severe positive symptoms but similar levels of negative symptoms in smokers compared to nonsmokers . Of importance, no distinction was made in this study between patients using only nicotine and patients using nicotine along with other drugs. It has been suggested that negative symptoms are associated with reduced dopaminergic neurotransmission in the mesolimbic and mesocortical systems . The acute administration of cannabis, nicotine and other drugs of abuse can increase dopaminergic neurotransmission in these systems and could thus improve negative symptoms in the short term . However, chronic substance abuse has been suggested to impair dopaminergic neurotransmission and could have deleterious effects on negative symptoms . To our knowledge, no joint meta-analysis of the association of cannabis and nicotine with negative symptoms has been conducted thus far. The previous meta-analyses examining substance use disorder applied restrictive criteria, and therefore, only a few studies specific on cannabis use were retained . Furthermore, in these previous studies, no distinction could be made between populations consuming only cannabis and/or nicotine and populations using other substances such as alcohol, stimulants and opioids. Therefore, this systematic review investigates the relationship of cannabis and nicotine, used alone or mixed with other substances, with negative and positive symptoms of schizophrenia. In addition, we aimed to perform an analysis of negative symptom severity in persons recently abstaining from cannabis or nicotine use. Studies that met all the following criteria were included: a) available in English; b) included inpatients or outpatients aged greater than 18 years with a DSM  or ICD  diagnosis of schizophrenia, schizoaffective disorder, or related psychotic disorder; patients considered stable ; c) an observational study  or a clinical trial reporting a baseline measurement; d) reported current nicotine or cannabis use either alone or as main drug of choice; e) specified whether patients were abstinent at the time of evaluation; and f) reported a baseline negative symptom score. Studies that met any of the following criteria were excluded: a) meta-analyses, reviews, case reports, posters; b) reported populations with other diagnoses  and patients with first-episode psychosis; c) articles with overlapping datasets ; and d) articles with nonexploitable datasets . In case of exclusion, reasons were reported. Two authors  independently applied the inclusion criteria to the identified studies.To achieve a high standard of reporting and to assess study comparability, we performed a full-text review with a detailed analysis of each included study . Two authors  independently performed data extraction. Discrepancies between the two authors were resolved upon a joint full-text analysis with the third author . All extracted data were independently crosschecked by two authors  before the calculation of outcome variables. If necessary, additional or missing information was obtained from the study authors. For the primary outcome, we extracted the mean value and standard deviation at baseline  of negative symptom severity.

A broad range of assessment instruments was considered: the negative subscale of the Positive and Negative Syndrome Scale  , the Scale for Assessment of Negative Symptoms  ,vertical grow rack the withdrawal/retardation factor of the Brief Psychiatric Rating Scale   or any other validated scale for the assessment of negative symptoms. In a secondary analysis, we considered the negative symptom dimensions amotivation and diminished expression separately; amotivation was calculated as the sum of the global ratings of avolition/apathy and anhedonia/asociality, and diminished expression was calculated as the sum of blunted affect and alogia . The SANS dimension scores were imputed using the correlation matrix for the SANS global domain scores from a dataset reported in a previous study . We considered positive symptoms at baseline  as secondary outcomes. Positive symptoms were assessed with the positive subscale of the PANSS , the Scale for the Assessment of Positive Symptoms  , or the sum of the thinking/disturbance and hostile/suspiciousness factors of the BPRS. Additionally, we extracted all baseline values for scales reporting depression scores and specific variables such as demographic information , disease information , study design , and information on substance use . The included studies were assessed for methodological quality using the Newcastle-Ottawa Scale   modified version for each type of observational study retrieved . The NOS is based on three subscales concerning the selection of cases, their comparability with the controls and the ascertainment of the exposure. A star system is used to assess each quality item; the highest-quality studies are awarded up to nine stars. For each study type, we defined the mean score of all included studies as a cutoff to identify high-quality studies. Two authors  independently performed assessments. All statistical analyses were performed using Cochrane Collaboration software, Review Manager  and Stata software . Our primary goal was to examine the association between cannabis and nicotine use and negative symptoms. We distinguished studies where nicotine alone  or cannabis and nicotine were used by patients from studies where other drugs were used or combined with these drugs , as these drugs have different effects on the brain. Accordingly, we defined four different subgroups of drug use: ‘cannabis and nicotine’, ‘cannabis as a main drug of choice’, ‘nicotine only’ and ‘nicotine as a main drug of choice’.

We defined the ‘main drug of choice’ as either the drug the user reported as preferred or as the drug considered to be the main drug by the clinician rater. In addition, in almost all studies dependence criteria were fulfilled only for the main drug of choice. In the few studies including patients fulfilling dependence criteria for more than one substance, only a minority of patients was concerned. For a detailed description of drug use patterns in each study see supplementary table S1. An additional subgroup consisted of studies including patients with ‘recent cannabis abstinence’, which corresponds to the absence of cannabis consumption for at least the past 21 days . We intended to create an equivalent group for recent nicotine abstinence, but only two studies addressed the question, and the available data did not allow us to perform a meta-analysis . In a first main analysis, we focused on cannabis use and included the subgroups ‘cannabis and nicotine’ and ‘cannabis as a main drug of choice’; then, we compared these results with those of the ‘recent cannabis abstinence’ group. In a second main analysis, we focused on nicotine use and included the subgroups ‘nicotine only’ and ‘nicotine as a main drug of choice’. Since all included studies reported a baseline or a single score for our outcomes of interest, we did not separate studies on the basis of their design. For the primary outcomes, different scales or versions of scales were used across studies, and we therefore used standardized mean differences of cross-sectional measurement . There were clear a priori reasons for assuming heterogeneity because studies differed in design, population and substance use. Therefore, we used a random-effects model for all analyses.