Focus on Alternative and Complementary Therapies
www.pharmpress.com/fact
Focus Alternat Complement Ther©2005 Pharmaceutical Press
Focus Altern Complement Ther 2002; 7: 334–6
The discussion about the merits (or otherwise) of the therapeutic use of cannabis (Cannabis sativa) continues to be lively. Promising results have been obtained in the treatment of pain and wasting syndrome in patients with cancer or AIDS, spasticity in multiple sclerosis (MS) patients, movement disorders and glaucoma.1 Given this broad therapeutic potential and considering the fact that cannabis remains a popular recreational drug, it is important to determine its safety. The prevailing thinking is that adverse effects, such as an increased risk of depression,2 outweigh the potential therapeutic benefits of cannabis3 and that therefore the medicinal use of cannabis is justified only in highly selected patients.4
In 1894, the British House of Commons’ ‘Indian Hemp Commission Report’ concluded that, in excessive doses, cannabis was associated with poverty, bronchitis and insanity, especially in those that are ‘weak minded’.5 A short-lived psychotic episode after acute cannabis use is well documented.6 It is also likely that cannabis can aggravate the symptoms and course of schizophrenia.7 But does regular cannabis use cause schizophrenia?
In 1972, Bernhardson and Gunne described 46 Swedish cases of psychoses in cannabis abusers.8 Fourteen of these were deemed to have pre-existing psychoses but, in 30 cases, the authors considered that cannabis had caused the psychoses. The illness was usually self-limiting and resembled severe endogenous psychoses with aggressive and destructive behaviour.
Tsuang et al. compared 72 chronic drug abusers suffering from psychoses of long (n = 45) or short (n = 27) duration with 30 psychotic patients without a history of drug abuse.9 Ninety-three per cent of patients with short-term psychoses, 82% of patients with long-term psychoses and 40% of patients without a history of drug abuse had used cannabis in the past.
Magliozzi et al.10 screened 76 Californian psychiatric patients for cannabis use (urinary Δ-9-tetrahydrocannabinol-11-oic acid). The following subgroups tested positive: schizophrenia, 42%; primary affective disorder, 19%; post-traumatic stress disorder, 25%; borderline personality disorder, 25%; other personality disorders, 71%; and primary drug-abuse disorder, 35%.
Negrete et al.11 evaluated 137 schizophrenic patients admitted to a psychiatric unit in Montreal. The percentage of cannabis users in this patient sample was roughly equivalent to that of the general population. These authors found that schizophrenic cannabis users (cases) were associated with a higher degree of delusional and hallucinatory activity and more hospital visits than schizophrenic non-users of cannabis (controls).
Miller and Tanenbaum studied 55 male schizophrenic patients (DSM-III criteria) consecutively admitted to the psychiatric department of a New York general hospital.12 Thirty of these patients had a history of drug abuse: alcohol, 31%; cannabis, 24%; and cocaine, 8%.
Dixon and colleagues13 studied schizophrenic (n = 68), schizoactive (n = 12) or schizophreniform (n = 3) patients consecutively admitted to a New York hospital. Forty-eight per cent of these patients had a history of drug or alcohol abuse. The most frequently taken drug was cannabis (n = 26).
Mueser et al. categorised a group of 263 Swedish patients with schizophrenia or other major affective disorders according to specific types of substance abuse.14 Within the subgroup of schizophrenic patients (n = 85), the prevalence of cannabis use was 22% (schizoaffective disorder, 8%; major depression, 11%; bipolar disorder, 22%). These researchers also examined a group of 325 recently hospitalised psychiatric patients, of which 53% were diagnosed with schizophrenia or schizoaffective disorders.15 Twenty-six per cent and 29% of cannabis users had schizophrenia or schizoaffective disorder. For other psychiatric conditions the percentages were similar: bipolar disorder, 26%; major depression, 16%; other mental illnesses, 33%. Alcohol abuse was most commonly associated with any of these psychiatric diagnoses.
Collectively these cross-sectional data provide some, albeit inconsistent, evidence of an association between chronic cannabis use and schizophrenia. However, they shed little light on the nature of the association: cross-sectional studies cannot establish causality.
Andrèasson and colleagues studied 45 570 Swedish conscripts over a follow-up period of 15 years.16 Based on (non-anonymous) questionnaires, the authors evaluated the frequency of previous cannabis use. Taking into account 11 confounding variables related to social and psychological characteristics, the investigators calculated the relative risk of schizophrenia-related hospital admission during follow-up as a function of cannabis use. Individuals that had tried cannabis one to four times in the past were not associated with an increased risk. However, having used cannabis 5–10, 11–50 or more than 50 times was associated with an increased risk. A dose–effect relationship was apparent across these categories of cannabis use. The relative risk of schizophrenia for those individuals having consumed cannabis more than 50 times was 6.0 [95% confidence interval (CI) 4.0–8.9].
A small subsample of participants in this study was submitted to further analyses.17 Eight conscripts that had used cannabis on more than 10 occasions and were later diagnosed with schizophrenia were compared with 13 conscripts that had not used cannabis but who also developed schizophrenia. The authors’ aim was to investigate in detail the role of potential confounding factors. They found no significant impact of other narcotic drugs, prior mental disorder or family history of schizophrenia. However, negative social background factors were found to be more common among cannabis users.
Mathers and Ghodse conducted a prospective comparison of UK inpatients with psychotic symptoms plus cannabis-positive urine analysis (cases, n = 61) and concurrently admitted psychotic patients with negative urine analysis (controls, n = 43).6 At baseline, there were marginally more schizophrenic patients in the groups of cases than in that of the controls. As expected, a symptom cluster consistent with acute cannabis intoxication was prevalent in cases but not in controls. At 1-month follow-up, the two groups differed significantly only in terms of delayed sleep, which was more frequent in cases than in controls.
Allebeck et al. identified all patients registered in Stockholm County diagnosed with cannabis dependence and psychosis between 1971 and 1983.18 These investigators retrospectively verified the psychiatric diagnoses, the history of substance abuse and the clinical cause. Of 229 cases admitted into this study, 112 (49%) fulfilled the DSM-III-R criteria for schizophrenia. Sixty-nine per cent of the cases had a record of heavy cannabis abuse at least 1 year before the onset of the psychotic symptoms. The authors therefore concluded that ‘cannabis abuse may be a risk factor for schizophrenia’.
Even though the conclusiveness of the findings summarised above is highly variable, collectively they suggest a link between chronic cannabis use and schizophrenia. The most compelling evidence originates from longitudinal studies.6,16–18 Causality is, however, difficult to establish and even the longitudinal studies usually lack control for potentially aetiological factors (e.g. genetic background, multi-substance abuse and premorbid psychopathology). The findings of Andrèasson and colleagues16 seem to provide the soundest data but could still be confounded by a range of factors. For example, the reliability of non-anonymous questionnaires handed out to military conscripts might be questioned, the validity of the psychiatric diagnosis is debatable, the influence of other drugs or acute effects of cannabis is undetermined and it is unclear whether schizophrenia was caused by cannabis use or vice versa.19–21 Andrèasson and colleagues attempted to resolve some of these issues by analysing a subset of their data in greater detail.17 Unfortunately, the sample size of that study was minute and negative social background factors could not be ruled out as a common denominator causing or precipitating both schizophrenia and cannabis use. The case–control study by Allebeck et al.18 is similarly open to confounding and bias. In particular, the retrospective verification of diagnoses and substance abuse constitute methodological weaknesses of that investigation, which render its interpretation ambiguous.
Cannabis could therefore either directly cause schizophrenia or, more probably, precipitate it in schizophrenia-prone patients.16 Alternatively, cannabis use might be the expression of an attempt to self-medicate symptoms of pre-existing schizophrenia.22 However, it is conceivable that, in schizophrenic patients, cannabis use renders clinical symptoms more clear-cut or severe which, in turn, would make a positive diagnosis more likely.11 It is also possible that chronic cannabis use causes a psychotic condition which has some of the features of schizophrenia but which is unrelated to schizophrenia itself.20 Furthermore, repeated acute cannabis use could lead to recurring short episodes of psychosis, which might accumulate into a chronic psychotic state.23
Other data seem to refute the existence of a link between cannabis and schizophrenia altogether.6 Jablensky and colleagues showed a uniform incidence of schizophrenia across a range of cultures.24 There was no identifiable influence of the highly different levels of cannabis use in these cultures. These results imply that, if chronic cannabis use contributes to schizophrenia, its influence is too small to be relevant.
In conclusion, the question whether chronic cannabis use causes schizophrenia remains unresolved. An association is supported by much of the evidence available to date but the data to suggest that it is causal by nature is not compelling. The problem is extremely complicated and unlikely to be answered on the basis of the existing, mostly epidemiological investigations.
