ORIGINAL PAPER Giuseppe Carra ` Æ Sonia Johnson Variations in rates of comorbid substance use in psychosis between mental health settings and geographical areas in the UK A systematic review Received: 19 March 2008 / Revised: 16 October 2008 / Published online: 13 November 2008 j Abstract Background Comorbid substance misuse in psychosis is associated with significant clinical, social and legal problems. An epidemiologically informed approach to planning service delivery requires an understanding of which clinical populations are at particularly high risk for such ‘dual diagnosis’. Evidence has now been accumulating in the UK since the early 1990s, and allows a relatively comprehensive compari- son of rates between service settings, geographical areas and social contexts in terms of ethnic background. Methods A literature search was carried out with the aim of investigating: (a) comorbid alcohol and drug misuse rates in people with established psychosis in different mental health and addiction settings in the UK, (b) variations in such rates between different population groups. Results There are wide variations in reported drug and alcohol misuse rates in psychosis. Most recent UK studies report rates between 20 and 37% in mental health settings, while figures in addic- tion settings are less clear (6–15%). Rates are generally not as high as in US studies, but appear to be especially high in inpatient and crisis team settings (38–50%) and forensic settings. In terms of geography, rates appear highest in inner city areas. Some ethnic groups are over-represented among clinical populations of people with dual diagnosis. Conclusions Rates of substance misuse in psychosis are likely to be influenced by service setting, population composition and geogra- phy. Acute and forensic settings are especially appro- priate for the development of targeted interventions. j Key words severe mental illness – substance abuse – prevalence – population characteristics – United Kingdom Introduction Comorbid drug and alcohol problems (‘dual diagno- sis’) are growing concerns among people with schizophrenia and other severe psychoses because of their association with poorer clinical [63] and social [46, 94] outcomes. Most evidence about the preva- lence of this comorbidity comes from the USA. The largest of these investigations, the Epidemiologic Catchment Area (ECA) study found that 47% of people with schizophrenia had a comorbid substance use disorder, with odds of meeting criteria for such a disorder 4.6 times higher for individuals with schizophrenia than for the rest of the US population [83]. In the National Comorbidity Survey (NCS) [58], half of those who met criteria for a mental disorder also met those for a lifetime substance use one, with a rate for people with non-affective psychosis of 45%. Nevertheless, even within the US, there is evidence that rates of comorbidity in fact vary considerably between settings [10, 22, 33, 44, 72, 92] and geo- graphical areas [75]. In particular, it has been sug- gested that prevalence is lower in rural than urban areas [75, 82]. However, the complexity of the US mental health care delivery system makes it difficult to draw clear conclusions about such variations. An important question regarding the epidemiology of dual diagnosis is how far high rates of comorbidity are an inherent clinical feature of psychosis and how far they are influenced by local social contexts. Until fairly recently little evidence has been available about the epidemiology of dual diagnosis in countries other than the USA and whether prevalence and patterns G. Carra `(&) Æ S. Johnson Dept. of Mental Health Sciences Royal Free and University College Medical School Charles Bell House 67-73 Riding House Street London W1W 7EY, UK Tel.: +44-20/7679-9428 Fax: +44-20/7679-9426 E-Mail: g.carra@ucl.ac.uk Soc Psychiatry Psychiatr Epidemiol (2009) 44:429–447 DOI 10.1007/s00127-008-0458-2 SPPE 458