In 2002, the National Advisory Committee on Drugs commissioned a team at Dublin City University to explore the management of individuals with a combination of mental illness and substance misuse in Ireland. The definition of dual diagnosis used for this study was ‘the co-existence of both mental health and substance misuse problems for an individual’. Two distinct groups of service providers treat patients with combinations of such illnesses, the mental health services and the addiction services.

The report1 was launched on 1 November 2004. In order to review the management of dual diagnosis in Ireland, the researchers:

• Reviewed relevant national and international literature to identify the most appropriate methods of assessment, treatment and management of such illnesses. 
• Organised an open forum in one geographical area with a representative group of stakeholders (n=58/60, 97%). This included service users, mental health and addiction service providers, social service providers and representatives from the police. Participants considered the findings of the literature review in the context of their personal experience of dual diagnosis. The proceedings were documented and the transcriptions were analysed and themes were identified. 
• Carried out a national postal survey in order to provide a national overview of service provision for dual diagnosis in Ireland. This survey also ascertained attitudes towards and opinions about the place of dual diagnosis in Irish health services. A stratified sample by occupation and county of employment was selected. The participants were managers, clinicians and other service providers (n=141/191, 74%). The questionnaire consisted of 35 questions.
• Completed face-to-face semi-structured interviews with 10 per cent of the respondents to the national survey (n=14) to explore the key findings. The responses were taped and transcribed. 

A number of key themes were identified during the open forum:

• Respondents found it difficult to conceptualise, define and assess the severity of dual diagnosis;
• Policy development to date did not recognise dual diagnosis;
• The general practitioner was the first point of contact and the main service provider;
• Clients experienced stigmatisation, discrimination and marginalisation at addiction and mental health services;
• Services were not always client centred, sometimes normal practices overrode client needs;
• It was difficult for clients to access services;
• Few formal structures and protocols were provided to guide staff;
• Application of evidence-based practice varied;
• There was inadequate communication and liaison between services;
• Clients were lost in the gaps between the addiction and mental health services;
• Professionals working in addiction services were educated separately to those working in mental health services;
• Differences in the professional cultures of the two services led to conflicting beliefs and practices;
• There were difficulties in respecting professional care and treatment provided by the ‘other’ service;
• Multi-disciplinary approaches were more effective than single-discipline approaches; 
• A variety of service models exist, mainly serial, sometimes parallel and occasionally integrated. 

These themes were used to develop the questionnaire for the national survey.

The main findings of the national survey were: 

Over one-fifth of service providers reported that policies to address dual diagnosis were available in their area. The examples cited by respondents indicated that these policies address aspects of dual diagnosis rather than dual diagnosis itself. None of the plans or service reviews submitted to the researchers addressed the specific issue of dual diagnosis. Over two-fifths of service providers reported that formal and informal structures existed in their area, with more in addiction services (56%) than in mental health services (33%). In theory, one-fifth of respondents thought refusing treatment to people with dual diagnosis was justified. In practice, a large proportion of providers in the addiction (58%) and the mental health (43%) services reported that exclusion criteria applied to people with a dual diagnosis. The source of referral differed between the mental health services and addiction services. The mental health services accepted referral through general practitioners only, while the addiction services accepted referrals from a wide variety of sources (including self-referrals).

In relation to assessment, 93 per cent of respondents thought routine screening should be in place and 66 per cent reported that they always assessed clients for dual diagnosis. Sixty-three per cent of respondents agreed with the statement, ‘Clinical staff in my service are adequately trained to assess dual illnesses’, and 71 per cent strongly agreed with the statement, ‘Our service identifies clients with dual diagnosis’. According to the service providers, dual diagnosis is recorded for 37 per cent of cases. No respondent used a validated tool to assess dual diagnosis.

There is some ambiguity in relation to the recognition and treatment of dual diagnosis, evidenced by the lack of service structures and the extent of exclusion criteria. For example, 92 per cent of respondents from mental health services reported that they treat people with substance misuse problems, yet 43 per cent of respondents reported exclusion criteria applied to people with a dual diagnosis. Seventy-one per cent of respondents reported that they do not follow recommended models of treatment and 39 per cent of respondents agreed with the statement, ‘Clinical staff in my service are adequately trained to treat dual diagnosis’.

Overall, the responses indicated that there was little systematic co-ordination of care for people with dual diagnosis evident in any health board area; only 18 per cent of services offered a specific service. There were at least three models of service provision in operation: a parallel model (52%), an integrated model (29%) and a serial model (16%), although three-quarters of survey respondents agreed with the statement, ‘A fully integrated service is the best way to help people with dual diagnosis’.

Respondents reported that 76 per cent of services had formal communication links with other services, while 54 per cent had informal communication links. Respondents were generally content with the level of communication between addiction and mental health services, though this was strongly influenced by those who had responsibilities across both service areas. According to the authors, a higher proportion of respondents from the addiction services disagreed with the statement ‘Communication between addiction and mental health services is adequate to treat dual diagnosis’ than the proportion of their counterparts in the mental health services, (p<0.03).2   

There was consensus throughout the study that GPs should be involved in the management of people with dual diagnosis.

It is clear that a published national strategy is required to deal with individuals who have both drug addiction and psychiatric illness. In practice, there is a need to:

• Formalise referral procedures between the mental health services and the addiction services;
• Reconsider exclusion criteria;
• Use valid assessment tools;
• Develop and expand the small number of evidence-based dual diagnosis services in existence;
• Provide appropriate treatment for psychiatric illness and problem drug use, regardless of the treatment provider or setting. 

In order to improve responses, further research is required to estimate the prevalence of dual diagnosis, to ascertain the needs of persons with dual diagnosis and define the role of primary care professionals in the management of these combined conditions. 

1. MacGabhann L, Scheele A, Dunne T, Gallagher P, MacNeela P, Moore G and Philbin M (2004) Mental health and addiction services and the management of dual diagnosis in Ireland. Dublin: Stationery Office.

2. A p-value is a probability value which measures the likelihood that the observed association occurred due to chance.  Probability is measured on a scale of zero to one.  By convention, a value of p <0.05 is considered statistically significant (for health-related studies).  A value of p <0.05 means that there was a less than one in twenty probability that the observed association occurred by chance alone.