Testing hypotheses about the relationship between cannabis use and psychosis
Introduction
Clinical research has shown that high proportions of persons with schizophrenia report regular cannabis use and meet criteria for cannabis use disorders (Fowler et al., 1998, Mueser et al., 1990, Ziedonis and Trudeau, 1997). Epidemiological studies have also found an association between cannabis use and psychosis in the general population (Anthony and Helzer, 1991, Cuffel et al., 1993, Degenhardt and Hall, 2001, Tien and Anthony, 1990).
There has been considerable debate about the reasons for this association (Batel, 2000, Blanchard et al., 2000, Gruber and Pope, 1994, Hall, 1998, Hall and Degenhardt, 2000, McKay and Tennant, 2000, Mueser et al., 1998, Rosenthal, 1998, Thornicroft, 1990). Depending upon the nature of the relationship between cannabis use and psychosis, changes in the prevalence of cannabis use may potentially lead to changes in the incidence, prevalence or age of onset of psychosis.
In Australia, there has been a dramatic increase in the prevalence of cannabis use since the early 1970s (Degenhardt et al., 2000, Donnelly and Hall, 1994, Makkai and McAllister, 1998, McCoy, 1980). The present report assesses the evidence for four hypothesised relationships between cannabis use and psychosis, which would each predict different effects of increased cannabis use on the incidence, prevalence and age of onset of schizophrenia and the prevalence of chronic cannabis use among persons with the disorder.
According to this hypothesis there is a causal link between cannabis use and schizophrenia in the sense that cannabis use causes cases of the disorder that would not otherwise have occurred. This hypothesis has arisen from reports of ‘cannabis psychoses’ (Basu et al., 1999, Bernardson and Gunne, 1972, Carney et al., 1984, Chopra and Smith, 1974, Eva, 1992, Solomons et al., 1990, Tennant and Groesbeck, 1972, Wylie et al., 1995) and there is some evidence that cannabis users are more likely to report psychotic symptoms (e.g. Andreasson et al., 1987, Degenhardt and Hall, 2001, Tien and Anthony, 1990).
If this hypothesis is correct, then an increase in prevalence of cannabis use among young adults should increase the incidence and ultimately the prevalence of schizophrenia. Since there has been a dramatic rise in the prevalence of cannabis use in Australia, this hypothesis predicts an increase in the greater number of cases of schizophrenia among recent birth cohorts. Since the age of cannabis initiation has also declined, this hypothesis also predicts that the age of onset of schizophrenia would decline in recent birth cohorts. This hypothesis also predicts a rising prevalence of cannabis use among persons with schizophrenia.
A second hypothesis is that regular cannabis use precipitates schizophrenia among vulnerable individuals, that is, among persons who would have developed the disorder regardless of whether they used cannabis or not (Hall, 1998). This is supported by evidence that: (a) persons with first-episode schizophrenia who use cannabis are younger than those who do not (Linszen et al., 1994, Mathers et al., 1991, Rolfe et al., 1993); (b) cannabis use usually precedes the development of psychotic symptoms (Allebeck et al., 1993; Hambrecht and Haefner, 2000, Linszen et al., 1994); and (c) among first-episode cases of psychosis, those who used cannabis were more likely to have a family history of psychosis (McGuire et al., 1995).
According to this hypothesis, an increase in regular cannabis use in the general population would not affect the incidence of schizophrenia but it would reduce the age of onset of psychotic illness among those who used cannabis. That is, the incidence rates of persons using cannabis would be ‘brought forward’. If this led to more chronic psychotic disorders (e.g. because earlier onset cases are more likely to relapse) the prevalence of chronic cases of psychosis would increase. This would increase the prevalence of regular cannabis use among persons with schizophrenia.
According to this hypothesis, cannabis use would worsen the prognosis of schizophrenic persons by increasing relapse to schizophrenia. It is supported by evidence that persons with schizophrenia who use cannabis are more likely to suffer a relapse (Jablensky et al., 1991, Linszen et al., 1994). This hypothesis does not predict an increased incidence of schizophrenia among regular cannabis users. Instead, it predicts that persons with schizophrenia who are regular cannabis users will be more likely to have a relapse after their initial episode. This could increase the number of persons in the population with chronic schizophrenia. It would not affect the age of onset of psychosis. The prevalence of cannabis use among persons with schizophrenia would increase because there would be more cannabis users among chronic cases.
According to this hypothesis, persons with schizophrenia are more likely to become regular cannabis users, if they use the drug (Mueser et al., 1998). There is no causal relationship between cannabis use and psychosis, so increasing rates of cannabis use will have no effect upon the incidence or prevalence of schizophrenia and there would not be a change in age of onset. There would be an increased prevalence of cannabis use among persons with psychosis.
The predictions generated from each of these four hypotheses are summarised in Table 1; notably, each hypothesis has a unique set of predictions. We used mathematical modelling to assess the plausibility of these four hypotheses. We combined empirically derived information about the epidemiology of cannabis use and psychosis to predict trends in incidence, prevalence and age of onset of schizophrenia according to each of these hypotheses. These predicted trends were compared with evidence on observed trends in schizophrenia and cannabis use.
Section snippets
Incidence
It was assumed that schizophrenia does not develop before the age of 15 years (Goldstein et al., 1984), and that new cases do not occur after the age of 54 years (Goldstein et al., 1984). Separate specific incidence rates were used for males and females because males have an earlier onset of schizophrenia on average than females (Jablensky et al., 1991, Jones and Cannon, 1998). Estimates of the average incidence rate of schizophrenia per 100 000 population per year by age and gender were
Modelling the natural history of cannabis use
Fig. 2 shows the estimated natural history of cannabis use in each of the birth cohorts. The peak prevalence of regular cannabis use occurs earlier in recent birth cohorts while peak prevalence of weekly use was higher for earlier birth cohorts.
Modelling the prevalence of schizophrenia
Fig. 3 shows the estimated prevalence of schizophrenia among Australian males and females according to age. The prevalence of schizophrenia by age 54 was 1.17% for males, and 1.08% for females. This is at the higher end of the estimated prevalence of
Does cannabis use cause psychosis?
The hypothesis that cannabis causes schizophrenia was not supported by the data on trends in the incidence of this psychosis in Australia. There was no evidence that there has been an increase in incidence over the past 30 years of the magnitude predicted by the hypothesis. This suggests that cannabis use has not caused cases of psychosis that would not otherwise have occurred. Even if regular cannabis use did double the risk of users developing schizophrenia (the ‘doubling’ of risk being the
Conclusions
This study has used modelling (incorporating data-based parameters) to predict what changes we would expect to see in the incidence and prevalence of schizophrenia if each of four hypotheses about the relationships between cannabis use and psychosis were true. The claim about cannabis and psychosis is widely understood in the popular media and public debate in Australia to imply that cannabis use has increased the number of cases of psychosis in the population (in the sense of causing cases of
References (72)
- et al.
Cannabis and schizophrenia: a longitudinal study of Swedish conscripts
Lancet
(1987) Addiction and schizophrenia
Eur. Psychiatry
(2000)- et al.
Substance use disorders in schizophrenia: review, integration, and a proposed model
Clin. Psychol. Rev.
(2000) - et al.
Cohort trends in the age of initiation of drug use in Australia
Aust. N. Z. J. Public Health
(2000) - et al.
Substance abuse and suicidality in schizophrenia: a common risk factor linked to impulsivity
Psychiatry. Res.
(2001) - et al.
The THC content of cannabis in Australia: evidence and implications
Aust. N. Z. J. Public Health
(2000) - et al.
The new epidemiology of schizophrenia
Psychiatr. Clin. North Am.
(1998) - et al.
Morbid risk of schizophrenia for relatives of patients with cannabis associated psychosis
Schizophr. Res.
(1995) - et al.
Dual diagnosis: a review of etiological theories
Addict. Behav.
(1998) Decreasing rates of incident schizophrenia cases in psychiatric services: a review of the literature
Eur. Psychiatry
(1995)