Elsevier

Schizophrenia Research

Volume 113, Issues 2–3, September 2009, Pages 123-128
Schizophrenia Research

Assessing the impact of cannabis use on trends in diagnosed schizophrenia in the United Kingdom from 1996 to 2005

https://doi.org/10.1016/j.schres.2009.05.031Get rights and content

Abstract

A recent systematic review concluded that cannabis use increases risk of psychotic outcomes independently of confounding and transient intoxication effects. Furthermore, a model of the association between cannabis use and schizophrenia indicated that the incidence and prevalence of schizophrenia would increase from 1990 onwards. The model is based on three factors: a) increased relative risk of psychotic outcomes for frequent cannabis users compared to those who have never used cannabis between 1.8 and 3.1, b) a substantial rise in UK cannabis use from the mid-1970s and c) elevated risk of 20 years from first use of cannabis. This paper investigates whether this has occurred in the UK by examining trends in the annual prevalence and incidence of schizophrenia and psychoses, as measured by diagnosed cases from 1996 to 2005. Retrospective analysis of the General Practice Research Database (GPRD) was conducted for 183 practices in England, Wales, Scotland and Northern Ireland. The study cohort comprised almost 600,000 patients each year, representing approximately 2.3% of the UK population aged 16 to 44. Between 1996 and 2005 the incidence and prevalence of schizophrenia and psychoses were either stable or declining. Explanations other than a genuine stability or decline were considered, but appeared less plausible. In conclusion, this study did not find any evidence of increasing schizophrenia or psychoses in the general population from 1996 to 2005.

Introduction

The risk of developing psychoses as a result of cannabis use has been the topic of extensive research. A systematic review concluded that people who have ever used cannabis have a significantly higher risk of developing psychoses, with those using cannabis more frequently being at particularly high risk. The meta-analysis indicated an odds ratio of 1.41 (95% CI 1.20–1.65) for ever use and 2.09 (95% CI 1.54–2.84) for frequent use compared to never use (Moore et al., 2007).

More recently Zammit et al. (2008) conducted a systematic review to look specifically at whether cannabis use leads to worse outcomes in people with psychotic disorders, measured by relapse or rehospitalisation, symptom severity, and response and adherence to treatment. They found that few studies adjusted for factors known to influence mental health outcomes, such as alcohol or other drug use (Jane-Lopis and Matytsina, 2006, Regier et al., 1990) and socio-demographic characteristics (Lauronen et al., 2007, Munk-Jorgensen and Mortensen, 1992). Furthermore, an important limitation of many studies is that they have failed to distinguish the direction of association between cannabis use and psychosis; although using cannabis is associated with a greater risk of developing psychosis, there is also an evidence of increased cannabis use following psychosis onset (Hides et al., 2006, Ferdinand et al., 2005). This is consistent with higher rates of substance use in general among psychotic patients (Gregg et al., 2007), and in psychiatric illness overall (Frisher et al., 2005). It is therefore difficult to establish a causal relationship between cannabis use and the development of psychotic disorders.

An alternative approach to investigating this link is to examine population rates in psychosis and schizophrenia and to compare these to known trends in cannabis use. It is analogous to investigating changes in the incidence and prevalence of lung cancer following changes in smoking trends (see Doll and Peto, 1976). A UK study reported that cannabis use increased fourfold between 1972 and 2002, increasing 18-fold among under-18s (Hickman et al., 2007). The study then considers a scenario whereby the risk of schizophrenia is elevated 1.8-fold among ‘light or short-term’ users and 3.1 among ‘heavy or long-term’ users. These risks affect only those starting cannabis use when aged under 20, but elevated risk lasts for 20 years. Under this scenario, increases in cannabis would lead to increases in schizophrenia incidence and prevalence of 29% and 19% respectively, between 1990 and 2010. The corresponding figures for women are 24% and 14%. More recent data from the British Crime Survey (BCS: Roe and Man, 2006) report that cannabis use has started to decrease during the last decade, particularly among young people aged 16–24; 21.4% of this age group reported having used cannabis in the last year during the 2005/06 survey, compared with 26.0% in 1996.

A recent study in Zurich drew comparisons between trends in cannabis use and the incidence of psychotic disorders in young people between 1977 and 2005 (Ajdacic-Gross et al., 2007). First admission rates for psychotic disorders remained constant for men and showed a downward trend for women; in the second half of the 1990s however there was a strong increase in the youngest age groups, particularly among males. This coincided with a distinct increase in cannabis availability (i.e. hemp shops) and consumption in the 1990s (Delgrande Jordan et al., 2004, Kuntsche, 2004). This is contrary to reports that increases in population cannabis exposure have not been followed by upward trends in the incidence of psychotic disorders (Macleod et al., 2006, Arsenault et al., 2004, Rey and Tennant, 2002). However, it is important to take into account socio-cultural differences in cannabis availability and consumption; for this reason, it is more appropriate to focus on trends in cannabis use in the UK when comparing these to schizophrenia/psychosis trends in UK general practice.

The present study uses the General Practice Research Database (GPRD) to determine the annual rate of new and existing diagnosed cases of schizophrenia/psychosis occurring in a population over the 10-year period from 1996 to 2005. Approximately 95% of the UK population is registered with a general practitioner, and age and sex distributions of patients within the GPRD are similar to those reported by the National Population Census (Garcia Rodriguez and Pérez Gutthann, 1998). Socio-economic status is indicated at practice level by Index of Multiple Deprivation (IMD: Office of the Deputy Prime Minister, 2004) score, which is used to assign the practice to one of five quintiles based on the spread of scores within each country; the distribution of GPRD practices across these quintiles reflects that of the general population. Furthermore, the quality of GPRD data has been found to be satisfactory for clinical research, with validation studies reporting high levels of concordance between clinical and computer records (Jick et al., 1991, Jick et al., 1992).

Based on literature suggesting a) an elevated risk of developing schizophrenia/psychosis among cannabis users, b) a substantial rise in cannabis use in the UK from the mid-1970s onwards and c) an assumed elevated risk of 20 years, this model would predict a corresponding increase in schizophrenia/psychosis during our study period. The paper focuses on the age specific rates for patients aged 16–44 for three reasons. First, rates for all ages could be influenced by an ageing population. Second, a 20 year old commencing cannabis use in 1980 would be 45 in 2005. Third, 16–44 is a conventional category used in other studies.

Section snippets

Sample

The data for this study comes from the GPRD, which is managed by the Medicines and Healthcare Products Regulatory Agency (MHRA). The data have been obtained under a license from the Medical Research Council (MRC). The study cohort comprises all patients aged 16–44 in 183 GPRD practices, with almost 600,000 patients within each study year. These practices were selected because they continuously submitted data from 1996 to 2005. The practices are drawn from the nine National Health Service (NHS)

Results

Fig. 1 shows the annual prevalence rates for schizophrenia and psychoses diagnoses. There was a significant decrease in the prevalence of schizophrenia diagnoses from 1996 to 2005 (Chi² for linear trend = 25.7 (1 DF), P < 0.0001). There was no significant change in the prevalence of psychoses diagnoses from 1996 to 2005 (Chi² for linear trend 0.345456 (1 DF), P = 0.7298) but there was a significant decrease in the prevalence of psychoses diagnoses from 1999 to 2005 (Chi² for linear trend = 42.76, (1

Main findings

The results of this study indicate that the incidence and prevalence of diagnoses of schizophrenia and psychoses in general practice did not increase between 1996 and 2005. As well as the GPRD indicating declining prevalence of schizophrenia/psychoses, Hospital Episode Statistics (HES) data for admissions relating to schizophrenia and psychoses also show a decline from 1998/99 to 2005/06 (HES online, 2008). While the latter data could be due to policy, e.g. less care for such patients in

Role of funding source

Funding for this study was provided by the North Staffordshire Primary Care Research Consortium and the Advisory Council on the Misuse of Drugs (ACMD); neither had any further role in study design; in the collection, analysis and interpretation of data; in the writing of the report; and in the decision to submit the paper for publication.

Contributors

Martin Frisher participated in conceptualising and developing the original research idea, designing the study, and analysing the data; Ilana Crome participated in conceptualising and developing the original idea for this research and in designing the study; Orsolina Martino participated in developing the study idea and analysing the data; Peter Croft participated in the developing the study idea and in designing the study. All authors contributed to the writing of the manuscript and have

Conflict of interest

None.

Acknowledgement

We are extremely grateful to the following people. Efrosis Setakis, Tim Williams and Jon Ford at the Medicines & Healthcare Products Regulatory Agency, (MHRA). The MHRA team provided the data for the study under an MRC license granted to the authors. The authors would also like to thank all contributing general practitioners and their patients. The authors are grateful to Peter Jones, Professor of Psychiatry at the University of Cambridge and also to members of the ACMD for comments on the

References (31)

  • EMCDDA
  • FerdinandR.F. et al.

    Cannabis use predicts future psychotic symptoms, and vice versa

    Addiction

    (2005)
  • FrisherM. et al.

    Substance misuse and psychiatric illness: prospective observational study using the general practice research database

    J. Epidemiol. Community Health

    (2005)
  • Garcia RodriguezL.A. et al.

    Use of the UK general practice research database for pharmacoepidemiology

    Br. J. Clin. Pharmacol.

    (1998)
  • HickmanM. et al.

    Cannabis and schizophrenia: model projections of the impact of the rise in cannabis use on historical and future trends in schizophrenia in England and Wales

    Addiction

    (2007)
  • Cited by (49)

    • Is there sufficient evidence that cannabis use is a risk factor for psychosis?

      2020, Risk Factors for Psychosis: Paradigms, Mechanisms, and Prevention
    • Co-occurring mental illnesses and substance use disorders: Evidence from Cyprus

      2019, Archives of Psychiatric Nursing
      Citation Excerpt :

      They find that only a few of these studies adjust for factors known to influence mental health outcomes, such as other substances abuse (Jane-Lopis & Matytsina, 2006) and socio-demographic characteristics (Lauronen et al., 2007). Furthermore, they infer that an important limitation of many related studies is that they fail to distinguish the direction of association between substance misuse and mental illness (Frisher, Crime, Martino, & Croft, 2009). Indeed, although misusing substances is associated with a greater risk of developing mental disorders, there is also evidence of increased substance use following psychosis onset (Gregg, Barrowclough, & Haddock, 2007).

    • Association between cannabis and psychosis: Epidemiologic evidence

      2016, Biological Psychiatry
      Citation Excerpt :

      Given that cannabis use has increased greatly since the 1960s, an argument made against a causal association between cannabis and schizophrenia is that a corresponding increase in schizophrenia diagnoses has not been observed. Some studies have found that incidence of psychotic outcomes has increased in recent decades (58,59), while others have found no change or a decrease (60,61). Ecological evidence such as this provides only very weak evidence for causality, as it cannot be ascertained whether individuals using cannabis are the same as those experiencing psychosis (the ecological fallacy), studies are unable to account for likely confounders, and studies do not account for other potentially competing risk factors for schizophrenia that may have declined over the same time period.

    • Effect of reclassification of cannabis on hospital admissions for cannabis psychosis: A time series analysis

      2014, International Journal of Drug Policy
      Citation Excerpt :

      One possible explanation is that the changing trend in cannabis psychosis reflects broader trends in the incidence of schizophrenia. However, Frisher, Crome, Martino, and Croft (2009) report that “Between 1996 and 2005 the incidence and prevalence of schizophrenia and psychosis were either stable or declining.” They found this was true for both general practice and hospital admission data.

    View all citing articles on Scopus
    View full text