Evidence table for systematic reviews

Transcription

1 Evidence table for systematic reviews Topic: synthetic Cannabinoids Reviewer: CMF, MS Abbreviations: CB cannabis, ER emergency room, etoh ethanol, f female, m male, SC synthetic cannabinoid, SR systematic review, y years Reference Research Parameters Population Outcomes Funding Additional comments Bibliographic reference Research question Theoretical approach Data collection Method and process of analysis Population and sample collection Gender Age Key themes Source of funding Limitations Evidence gap OCEBM level of evidence [1] [2] [3] [4] [5] [6] [7] [8] [9] [10] [11] [12] [13] [14] Brewer et al. SR > 350 cases; 12- (2014) 25y Castaneto et al. (2014) in adolescents and young adults, how does SC compared to CB affect psychosis and tachycardia within 24h of use epidemiology, pharmacodynamic profiles and clinical implications of SC use SR CINAHL, PubMed, PsychInfo, manual publication tracing, Google search PubMed, Embase, Web of Science, Scopus, Chochrane, biol.+chem. Abstracts via STN, SciFinder, Google, Google Scholar, manual publication tracing inclusion: English, peerreviewed publication, case reports; 24 case reports 215 articles split in epidemiology (9), human pharmacodynamics (71), receptor interactions (42) acute SC-intox: >250 cases (from EmergRoom) acute intox: 202m:38f (5.3:1) 12-25y; some caseseries with older patients (- 67y) acute intox: 13-59y, mean 22y > common SC-user in Emerg.Room: male, 12-29y > unpredictable effects of SCs due to higher affinity/potency of SCs and unknown content of plant material > negative SC effects: delusion, hallucination, anxiety, paranoia, agitation, seizure, dystonia, tremors, dyspnea, nausea/vomiting, palpitations, tachycardia, diaphoresis > mortality: 2 cases (coronary ischemia, suicide) > renal: acute kidney failure (19 cases) > lacking routine toxicol. testing for SCs > SCs are legal, perceived as save making them attrictive to adolescents > epidemiology: no population/community-based data, last year prevalence: %, male: %, concurrent CB use: 84-98%, etoh/polydrug: 64-92% > adverse clinical effect: 60% moderate or major toxicity symptoms after SC; neurological 61.9%, cardiovasc. 43.5%, gastrointest. 21.1%, respiratory 8%, ocular 5%, dermal 2.6%, renal 0.9%, hematolog. 0.4%, miscell.(acidosis, hyperglycemia, diaphoresis..) 25.9% > mortality: ( confirmed methoxetamine+scs) interagency agreement Dept. of Defense Counter Narcotics Program and Chemistry and Drug Metabolism Section, IRP, National Institute on Drug Abuse, NIH > research question not answered;lacking comparison SC vs. CB > included older cases not explicitely characterized > SC epidemiology data limited, estimates from cross-sectional studies/small convenience samples > undetectable/unidentifie d SC complicate evaluation of SC clinical effects und development of treatment > pharmacokineti c, tissue distribution, elimination, bioactivity of SC metabolites, drug drug interactions of SCs >longtermeffects, in utero effects and abuse liability of SCs Methodological rating SIGN 4 > incomplete methods (SIGN+) 4 > very good methods (SIGN++)

2 Until 12/2013 Yielded 3161 papers > acute SC effects: altered mood/perception, conjunctival injection, xerostomia, tachycardia, sedation, thought disruption, loss of concentration, mild anxiety, motoric impairment; resolved within few hours; prolonged feeling of sedation and exhaustion (6-24h) > acute SC intox. symptoms mainly transient (<24h), more severe symptoms (e.g. psychosis, cardiac/renal failure) require longer hospitalization (<2weeks) > acute SC intox. / mental: agitation/irritability/ restlessness, anxiety, confusion, STMemory/cognitive impairment, psychosis > acute SC intox. / physical: dilated pupils, conjunctival injection, nausea/vomiting, slurred speech, dyspnea, hypertension, tachycardia, chest pain, muscle twitches, sweating, skin pallor > psychosis: de-novo psychosis after SC (14, 11 resolved within 2weeks, 3 >5month), transient psychotic symptoms, psychotic relapse in vulnerable persons > SC withdrawl: similar symptoms as CB withdrawl; sleep disturbances, anxiety, craving, nausea, muscle twitching/ cramping, chills; after <1week of abstinence from chronic SC Gunderson et al. (2012) potential clinical effects and biopsychosocial correlates of SC use and initiation SR and case series Medline, PsychInfo; just peerrevied studies 5 case reports, 1 semistructured patient interview, 3 toxicological studies, 3 cases of SC for substitution of CB withdrawl 28 cases some with existing psychiatric disorders; 3 cases of CBdependence (patients/study attendants) 21+ male: 5+ female; 3 male cases 17-47y; mean: 26.1y > SC effects: CB-like; altered mood/ perception, conjunctival injection, xerostomia, increased pulse > severe SC effects: acute anxity, psychotic reactions, hypertension, hyperventilation, diaphoresis, numbness, tingling, nausea/vomiting, tremors, muscle twitching, seizure > case reports: 3 CB-users (>4y) with CB-dependence; positive + stronger effects of SC similar to CB, substituted withdrawl symptoms National Institute on Drug Abuse > lacking empirical knowledge on SC effects in humans > low level of evidence for reported SC effects (largely anecdotal and case resports) > unknown effects of herbal mixtures 4 > incomplete methods (SIGN+)

3 Papanti et al. (2015) range of psychopathol. issues related to SCs in present litereture SR and case report LILACS, Medline, Toxnet, Scopus inclusion: all languages, studies on psychotic symptoms; 41 studies (9 retrosp. toxicol.surveys, 25 case reports/series, 4 laborat. studies acute SC effects, 3 self-reports/ interviews) 2207 cases; 11 studies psychiatric environment, 15 studies EmerRoom, 9 studies poison center calls + 1case report 3.16male:1f emale; case report: 1m mean: 22.97y; case report: 18y > SC and psychosis: acute transient psychosis, persistent psychotic disorder (de-novo), relapse/worsening of pre-existing psychosis, psychotic disorders in vulnerable persons (prediagnosed for ADHD, PTSD, Depression, polysubstance abuse[cb, alcohol]) > reporting bias for psychotic disturbances: poison calls (9.4-11%) vs. EmerRoom + analyt. confirmed SC ( %) > mental: paranoia/irritability, altered perception/mental status, thought disorganization, confusion, agitation/ anxiety/ panic attacks/ restlessness, depression/suicidal thoughts > treatment: not well documented; often untreated (assuming transient psychotic effects), benzodiazepins commonly for anxiety/restlessness/agitation > CB-asso. psychopath. syndrom after CB: 2%, after SC 11.2% > case report: unexperienced user, tachycardia/ agitation/ confusion/ anxiety 1h after SCuse (transient, several hours); 2nd EmerRoom admission anxiety/dyspnea/tachycardia/ muscular pain, persistent anxiety/insomnia/palpitation/ ideas of reference/hallucination/autoscop y ammeliorated after 4 weeks of psychopharm. treatment > association but no causality of SC use and psychotic disturbances > typical polysubstance use in SC users confounding factor > strong structural heterogeneity of SC compounds complicates interpretation > multiple SCs not detectable; association to clinical presentation impossible 4 > very good methods (SIGN++) Papaseit et al. (2014) trends of new psychoactive substances in Europe SR Pubmed 6 case-series, 2 selfreports 24 confirmed cases of SC use; Europeans 22 male:2 female 14-31y; mean: 19.6 y > prevalence in sub-group of clubbers: 17% (lifetime); age first use: 21y; 39% last-year use, 14% last month use > self-reported adverse effects: panic and anxiety (82.6%), paranoia (56.5%), breathing difficulties (56.5%), hangovers > common symptoms: tachycardia, mydriasis, drowsiness, seizure Instituto de Salud Carlos III-FEDER, European Commission, Rio Hortega Fellowship > European samples > difficulties to detect new SCs with analytical tests > very limited number of studies/cases regarded by this SR 4 > insufficient/ missing methodologi cal description (SIGN -)

4 Tait et al. (2015) 1. identification typical signs/symptoms of SC; idiosyncratic presentation 2. summary of interventions/ treatments SR PubMed, Medline, PsychInfo, Embase, Google Scholar, manual publication tracing 106 included studies; 29 case series (10+), 77 case reports (<10) > 4000 cases; from Emergency departments, poison control centers; adverse effects recorded by medical stuff (no self-reports); self-reported or analytically confirmed exposure to any SCs; no control for further drug exposure > 80% male mean age range: y > prevalence (last year use): 0.2%- 1.2% in general population(eur;aus); 2.5% <25y (AUS), 11.4% 17-18y (USA) > prototypic presentation: male (59-100%), tachycardia (37-77%), agitation (16-41%), nausea (13-94%) > most cases transient symptoms (<8h); some severe cases (mostly not analytically confirmed SC use) > mortality: cases direct or indirect after SC > cardiovascular: tachycardia (most prevalent), hypertension, chest pain, single cases of cerebral haemorrhage/occlusion, myocardial infarction/cardiac arrest > renal: <1% SC calls, acute kidney injury reported with SC intoxication (16cases/9month, 22 case reports), protein- /haematuria, enhanced white blood cell count > seizures: %, 15% <19y; prominent feature in case reports with analytically confirmed SC > nausea/vomiting: conspicious for SC (13-94%), reports on cannabinoid hyperemesis after SC > psychiatric: agitation (often), de-novo psychosis, facilitation of psychotic symptoms, panic attacks, anxiety, paranoia, hallucination, withdrawl symptoms (similar to CB) Curtin University Research Fellowship, ACT Health, WA Health, National Poison Information Service UK, Australian government fund > no exact number of cases due to overlapping data from hospitals / poison control centers > underestimation of SCrelated presentations/problems due to lacking routinely SC-screening methods > new formulations of SCs produce different /more severe clinical symptoms 4 > very good methods (SIGN++) van Amsterdam et al. (2015) 1. adverse health effects of SCs 2. psychosis-like effects SR PubMed, manual publication tracing Yielded 250 relevant papers 37 case reports/series included (number was not specified in the SR) Study quality not assessed and reported male > female 60-75% < 29 years > strong variation of SCs (amount, compound, pharmacokinetic) in herbal mixtures > toxicity profil of SC similar to overdosed THC; more severe / frequent side effects with SC; no letal cases with confirmed SC overdose (10 susceptible) > most frequent side effects: tachycardia, extreme agitation/irritability, drowsiness, hallucination > neurol./mental: euphoria, drowsiness, paranoia, panic attacks, anxiety, agitation, delusion, hallucination, dizziness none > limited number / quality of studies on SCs > insufficient reports on medical history / preexisting pathology / polydrug use which might moderate SC effects 4 > insufficient methodologi cal description of SR (SIGN:+/-)

5 and seizure, nausea, vomiting, STMemory+cognitive deficits > cardiovascular: tachycardia, hypertension, hyperglycemia, hypokalemia > renal: acute kidney injury > psychosis: 10 de-novo psychosis after SC (7 transient); 18-41% psychotic disorders in EmergencyRoom studies vs. 9-11% in interview studies; CBassoc. psychopathological syndrom more frequent after SC (11%) than CB (2%) >SC overdosing more frequent in drug-naive/unexperienced users; SC-patients in EmergRoom mean age:23/24y vs. 30y (CB)

6 S I G N Methodology Checklist 1: Systematic Reviews and Metaanalyses SIGN gratefully acknowledges the permission received from the authors of the AMSTAR tool to base this checklist on their work: Shea BJ, Grimshaw JM, WellsGA, BoersM, Andersson N, HamelC,. et al. Development of AMSTAR: a measurement tool to assess the methodological quality of systematic reviews. BMC Medical Research Methodology 2007, 7:10 doi: / Available from [cited 10 Sep 2012] Study identification (Include author, title, year of publication, journal title, pages) Brewer & Collins 2014 A review of clinical manifestations in adolescent and young adults after use of synthetic cannabinoids. J Spec Pediatr Nurs Guideline topic: Synthetic Cannabinoids Key Question No: Before completing this checklist, consider: Is the paper relevant to key question? Analyse using PICO (Patient or Population Intervention Comparison Outcome). IF NO reject. IF YES complete the checklist. Checklist completed by: Section 1: Internal validity In a well conducted systematic review: 1.1 The research question is clearly defined and the inclusion/ exclusion criteria must be listed in the paper. Does this study do it? 1.2 A comprehensive literature search is carried out. 1.3 At least two people should have selected studies. 1.4 At least two people should have extracted data. 1.5 The status of publication was not used as an inclusion criterion. x 1.6 The excluded studies are listed. x 1.7 The relevant characteristics of the included studies are provided.

7 1.8 The scientific quality of the included studies was assessed and reported. 1.9 Was the scientific quality of the included studies used appropriately? 1.10 Appropriate methods are used to combine the individual study findings The likelihood of publication bias was assessed appropriately. x 1.12 Conflicts of interest are declared. x SECTION 2: OVERALL ASSESSMENT OF THE STUDY 2.1 What is your overall assessment of the methodological quality of this review? 2.2 Are the results of this study directly applicable to the patient group targeted by this guideline? 2.3 Notes: Clear research question (PICOS) Specific for adolescent users High quality (++) Acceptable (+) Low quality (-) Unacceptable reject 0

8 S I G N Methodology Checklist 1: Systematic Reviews and Metaanalyses SIGN gratefully acknowledges the permission received from the authors of the AMSTAR tool to base this checklist on their work: Shea BJ, Grimshaw JM, WellsGA, BoersM, Andersson N, HamelC,. et al. Development of AMSTAR: a measurement tool to assess the methodological quality of systematic reviews. BMC Medical Research Methodology 2007, 7:10 doi: / Available from [cited 10 Sep 2012] Study identification (Include author, title, year of publication, journal title, pages) Castaneto et al Synthetic Cannabinoids: Epidemiology, Pharmacodynamics, and Clinical Imlications. Drug Alcohol Depend. Guideline topic: Synthetic Cannabinoids Key Question No: Before completing this checklist, consider: Is the paper relevant to key question? Analyse using PICO (Patient or Population Intervention Comparison Outcome). IF NO reject. IF YES complete the checklist. Checklist completed by: Section 1: Internal validity In a well conducted systematic review: 1.1 The research question is clearly defined and the inclusion/ exclusion criteria must be listed in the paper. Does this study do it? 1.2 A comprehensive literature search is carried out. 1.3 At least two people should have selected studies. 1.4 At least two people should have extracted data. 1.5 The status of publication was not used as an inclusion criterion. 1.6 The excluded studies are listed. x 1.7 The relevant characteristics of the included studies are provided.

9 1.8 The scientific quality of the included studies was assessed and reported. 1.9 Was the scientific quality of the included studies used appropriately? 1.10 Appropriate methods are used to combine the individual study findings The likelihood of publication bias was assessed appropriately Conflicts of interest are declared. SECTION 2: OVERALL ASSESSMENT OF THE STUDY 2.1 What is your overall assessment of the methodological quality of this review? 2.2 Are the results of this study directly applicable to the patient group targeted by this guideline? 2.3 Notes: High quality (++) Acceptable (+) Low quality (-) Unacceptable reject 0

10 S I G N Methodology Checklist 1: Systematic Reviews and Metaanalyses SIGN gratefully acknowledges the permission received from the authors of the AMSTAR tool to base this checklist on their work: Shea BJ, Grimshaw JM, WellsGA, BoersM, Andersson N, HamelC,. et al. Development of AMSTAR: a measurement tool to assess the methodological quality of systematic reviews. BMC Medical Research Methodology 2007, 7:10 doi: / Available from [cited 10 Sep 2012] Study identification (Include author, title, year of publication, journal title, pages) Gunderson et al Spice and K2 Herbal Highs: A Case Series and Systematic Review of the Clinical Effects and Biopsychological Implications of Synthetic Cannabinoid Use in Humans. Am J Addict Guideline topic: Synthetic Cannabinoids Key Question No: Before completing this checklist, consider: Is the paper relevant to key question? Analyse using PICO (Patient or Population Intervention Comparison Outcome). IF NO reject. IF YES complete the checklist. Checklist completed by: Section 1: Internal validity In a well conducted systematic review: 1.1 The research question is clearly defined and the inclusion/ exclusion criteria must be listed in the paper. Does this study do it? 1.2 A comprehensive literature search is carried out. 1.3 At least two people should have selected studies. 1.4 At least two people should have extracted data. 1.5 The status of publication was not used as an inclusion criterion. x x 1.6 The excluded studies are listed. x 1.7 The relevant characteristics of the included studies are provided.

11 1.8 The scientific quality of the included studies was assessed and reported. 1.9 Was the scientific quality of the included studies used appropriately? 1.10 Appropriate methods are used to combine the individual study findings The likelihood of publication bias was assessed appropriately Conflicts of interest are declared. SECTION 2: OVERALL ASSESSMENT OF THE STUDY 2.1 What is your overall assessment of the methodological quality of this review? 2.2 Are the results of this study directly applicable to the patient group targeted by this guideline? High quality (++) Acceptable (+) Low quality (-) Unacceptable reject Notes: Small number of included cases (9 studies/ 28 cases; 3 case reports) SCs for substituting cannabis withdrawl symptoms

12 S I G N Methodology Checklist 1: Systematic Reviews and Metaanalyses SIGN gratefully acknowledges the permission received from the authors of the AMSTAR tool to base this checklist on their work: Shea BJ, Grimshaw JM, WellsGA, BoersM, Andersson N, HamelC,. et al. Development of AMSTAR: a measurement tool to assess the methodological quality of systematic reviews. BMC Medical Research Methodology 2007, 7:10 doi: / Available from [cited 10 Sep 2012] Study identification (Include author, title, year of publication, journal title, pages) Papanti et al Spiceophrenia: a systematic overview of Spice-related psychopathological issues and a case report.hum Psychopharmacol Clin Exp Guideline topic: Synthetic Cannabinoids Key Question No: Before completing this checklist, consider: Is the paper relevant to key question? Analyse using PICO (Patient or Population Intervention Comparison Outcome). IF NO reject. IF YES complete the checklist. Checklist completed by: Section 1: Internal validity In a well conducted systematic review: 1.1 The research question is clearly defined and the inclusion/ exclusion criteria must be listed in the paper. Does this study do it? 1.2 A comprehensive literature search is carried out. 1.3 At least two people should have selected studies. 1.4 At least two people should have extracted data. 1.5 The status of publication was not used as an inclusion criterion. x 1.6 The excluded studies are listed. x 1.7 The relevant characteristics of the included studies are provided.

13 1.8 The scientific quality of the included studies was assessed and reported. 1.9 Was the scientific quality of the included studies used appropriately? 1.10 Appropriate methods are used to combine the individual study findings The likelihood of publication bias was assessed appropriately Conflicts of interest are declared. SECTION 2: OVERALL ASSESSMENT OF THE STUDY 2.1 What is your overall assessment of the methodological quality of this review? 2.2 Are the results of this study directly applicable to the patient group targeted by this guideline? 2.3 Notes: High quality (++) Acceptable (+) Low quality (-) Unacceptable reject 0

14 S I G N Methodology Checklist 1: Systematic Reviews and Metaanalyses SIGN gratefully acknowledges the permission received from the authors of the AMSTAR tool to base this checklist on their work: Shea BJ, Grimshaw JM, WellsGA, BoersM, Andersson N, HamelC,. et al. Development of AMSTAR: a measurement tool to assess the methodological quality of systematic reviews. BMC Medical Research Methodology 2007, 7:10 doi: / Available from [cited 10 Sep 2012] Study identification (Include author, title, year of publication, journal title, pages) Papaseit et al Emerging drugs in Europe. Curr Opin Psychiatry Guideline topic: Synthetic Cannabinoids Key Question No: Before completing this checklist, consider: Is the paper relevant to key question? Analyse using PICO (Patient or Population Intervention Comparison Outcome). IF NO reject. IF YES complete the checklist. Checklist completed by: Section 1: Internal validity In a well conducted systematic review: 1.1 The research question is clearly defined and the inclusion/ exclusion criteria must be listed in the paper. Does this study do it? 1.2 A comprehensive literature search is carried out. 1.3 At least two people should have selected studies. 1.4 At least two people should have extracted data. 1.5 The status of publication was not used as an inclusion criterion. x x x 1.6 The excluded studies are listed. x 1.7 The relevant characteristics of the included studies are provided. x

15 1.8 The scientific quality of the included studies was assessed and reported. 1.9 Was the scientific quality of the included studies used appropriately? 1.10 Appropriate methods are used to combine the individual study findings The likelihood of publication bias was assessed appropriately. x 1.12 Conflicts of interest are declared. SECTION 2: OVERALL ASSESSMENT OF THE STUDY 2.1 What is your overall assessment of the methodological quality of this review? 2.2 Are the results of this study directly applicable to the patient group targeted by this guideline? High quality (++) Acceptable (+) Low quality (-) Unacceptable reject Notes: European data considering new psychoactive substances (NPS); SCs just partly touched (adverse effects small abstract) Methods description very sparse; lacks numbers of references retrieved/ included; publication period just 1 year

16 S I G N Methodology Checklist 1: Systematic Reviews and Metaanalyses SIGN gratefully acknowledges the permission received from the authors of the AMSTAR tool to base this checklist on their work: Shea BJ, Grimshaw JM, WellsGA, BoersM, Andersson N, HamelC,. et al. Development of AMSTAR: a measurement tool to assess the methodological quality of systematic reviews. BMC Medical Research Methodology 2007, 7:10 doi: / Available from [cited 10 Sep 2012] Study identification (Include author, title, year of publication, journal title, pages) Tait et al A systematic Review of Adverse Events Arising from the Use of Synthetic Cannabinoids and Their Associated Treatment, Clinic Toxicol Guideline topic: Synthetic Cannabinoids Key Question No: Before completing this checklist, consider: Is the paper relevant to key question? Analyse using PICO (Patient or Population Intervention Comparison Outcome). IF NO reject. IF YES complete the checklist. Checklist completed by: Section 1: Internal validity In a well conducted systematic review: 1.1 The research question is clearly defined and the inclusion/ exclusion criteria must be listed in the paper. Does this study do it? 1.2 A comprehensive literature search is carried out. 1.3 At least two people should have selected studies. 1.4 At least two people should have extracted data. 1.5 The status of publication was not used as an inclusion criterion. x x 1.6 The excluded studies are listed. 1.7 The relevant characteristics of the included studies are provided.

17 1.8 The scientific quality of the included studies was assessed and reported. 1.9 Was the scientific quality of the included studies used appropriately? 1.10 Appropriate methods are used to combine the individual study findings The likelihood of publication bias was assessed appropriately Conflicts of interest are declared. SECTION 2: OVERALL ASSESSMENT OF THE STUDY 2.1 What is your overall assessment of the methodological quality of this review? 2.2 Are the results of this study directly applicable to the patient group targeted by this guideline? 2.3 Notes: Best SR considering content and methods >4000 cases included High quality (++) Acceptable (+) Low quality (-) Unacceptable reject 0

18 S I G N Methodology Checklist 1: Systematic Reviews and Metaanalyses SIGN gratefully acknowledges the permission received from the authors of the AMSTAR tool to base this checklist on their work: Shea BJ, Grimshaw JM, WellsGA, BoersM, Andersson N, HamelC,. et al. Development of AMSTAR: a measurement tool to assess the methodological quality of systematic reviews. BMC Medical Research Methodology 2007, 7:10 doi: / Available from [cited 10 Sep 2012] Study identification (Include author, title, year of publication, journal title, pages) Van Amsterdam et al The adverse health effecs of synthetic cannabinoids with emphasis on psychosis-like effects. J Psychopharmacol Guideline topic: Synthetic Cannabinoids Key Question No: Before completing this checklist, consider: Is the paper relevant to key question? Analyse using PICO (Patient or Population Intervention Comparison Outcome). IF NO reject. IF YES complete the checklist. Checklist completed by: Section 1: Internal validity In a well conducted systematic review: 1.1 The research question is clearly defined and the inclusion/ exclusion criteria must be listed in the paper. Does this study do it? 1.2 A comprehensive literature search is carried out. 1.3 At least two people should have selected studies. 1.4 At least two people should have extracted data. 1.5 The status of publication was not used as an inclusion criterion. x x x 1.6 The excluded studies are listed. x 1.7 The relevant characteristics of the included studies are provided. x

19 1.8 The scientific quality of the included studies was assessed and reported. 1.9 Was the scientific quality of the included studies used appropriately? 1.10 Appropriate methods are used to combine the individual study findings The likelihood of publication bias was assessed appropriately. x 1.12 Conflicts of interest are declared. SECTION 2: OVERALL ASSESSMENT OF THE STUDY 2.1 What is your overall assessment of the methodological quality of this review? 2.2 Are the results of this study directly applicable to the patient group targeted by this guideline? High quality (++) Acceptable (+) Low quality (-) Unacceptable reject Notes: Methods underreported Short overview of longterm effects, psychosis and self-medication in schizophrenia