Intended for healthcare professionals


Severe mental illness and substance misuse

BMJ 1999; 318 doi: (Published 16 January 1999) Cite this as: BMJ 1999;318:137

Research is needed to underpin policy and services for patients with comorbidity

  1. Tim Weaver, Research fellow.,
  2. Adrian Renton, Reader in social medicine,
  3. Gerry Stimson, Professor of sociology of health behaviour,
  4. Peter Tyrer, Professor of community psychiatry
  1. Department of Social Science and Medicine, Imperial College of Science, Technology, and Medicine, London SW6 1RQ
  2. Department of Psychiatry, Imperial College of Science, Technology, and Medicine, London W2 1PD

    Evidence from the United States suggests that half of all patients with schizophrenia also have a substance misuse disorder.1 This comorbidity is associated with poor prognosis and heavy use of expensive inpatient care through recurrent “revolving door” admissions.2 The phenomenon has only recently been recognised in the United Kingdom, but one survey of psychotic patients in an inner London district found that 36% misused drugs or alcohol. The same survey observed inpatient admission rates among comorbid patients that were almost double those of patients with psychosis alone.3 This high prevalence, the problems of clinical management,4 and a continued rise in the general rate of drug misuse make comorbidity a major public health issue, and the Department of Health is currently inviting applications for research into the prevalence and pattern of comorbidity.

    The term “dual diagnosis” is used increasingly in psychiatric practice to describe this combination of severe mental illness (mainly psychotic disorders) and substance misuse. Unfortunately the term is imprecise and its use seems only to confirm the inadequacy of current classification systems in describing certain complex presentations.5 We prefer the term “comorbidity” (the simultaneous presence of two or more disorders), though even this may fail to capture potential causal interactions between psychosis and substance misuse. Patterns of substance misuse vary considerably, but use as well as dependency may be problematic among people with psychosis. Whether there is any causal relation between substance misuse and psychotic disorders remains controversial. However, some types of substance misuse, particularly alcohol, cannabinoids, hallucinogens, and stimulants (such as amphetamines), can produce psychotic symptoms directly without mental illness. They may also precipitate psychotic disorders among people with a predisposition.

    Services for drug misusers and mental health services have tended to develop in ways determined more by public anxiety and political ideology than by research evidence.6 Although it is important that this history should not be repeated in relation to comorbidity, negotiating an evidence based path through existing policies may be difficult. For example, programmes to improve liaison between psychiatric and substance misuse services may appear attractive. Improved communication is never a bad thing, but there is a fashionable belief that it may be the crucial element missing.7 The medical model of psychiatric services, with their recourse to legal compulsion to treat those incapable of making rational health choices, contrasts sharply with the psychosocial orientation of substance misuse services. Thus the common language required for successful communication may not exist. Moreover, as both services often operate referral criteria that specifically exclude comorbid patients, liaison alone may be a recipe for buck passing.

    Policy makers may also find liaison models attractive because they mesh with the brokerage models of case management adopted in the United Kingdom. Under these arrangements case workers assume responsibility for assessing clients' needs and managing packages of appropriate services. Unfortunately, evidence supporting the effectiveness of UK case management is minimal.8 Assertive community treatment, which involves more intensive team based interventions, is now favoured in the United States. Such treatment has a secure evidence base8 but one derived almost exclusively from trials that have specifically excluded comorbid patients. The few trials of assertive community treatment specifically targeted at comorbidity have shown little benefit.9 Hence a growing body of opinion now argues that integrated treatment specifically for comorbidity must underpin an approach based on assertive community treatment and that clinical teams must be able to implement motivational therapy and treatment for both types of disorder without cross referral to other agencies.4 Recent quasiexperimental studies of integrated treatment teams suggest that the approach does have benefit over multiagency treatment involving separate teams.10

    Although the literature suggests that these broad principles should inform our response to comorbidity, evidence for the efficacy of interventions is limited. This makes the implementation of treatment for comorbid patients difficult for the NHS. Moreover, owing to an almost complete absence of evidence based service development in the recent past, current community care arrangements provide shaky foundations on which to develop appropriate services. There are three urgent priorities. Firstly, investment in an appropriately skilled workforce, trained in a range of treatment modalities, must be regarded as a basic building block of any future service. Secondly, we need research which investigates the extent and nature of comorbidity in the UK. Thirdly, new models for the delivery of services need to be developed and tested before widespread implementation. The policy mistakes forced on us by expediency in recent years must not be repeated.


    View Abstract