This contribution reviews the international literature about dual diagnosis, meaning patients who have simultaneously mental health problems and substance use disorders and discusses epidemiology, clinical characteristics, but primarily etiopathogenesis and different treatment models and interventions. The epidemiological data coming from large-scale studies in the general population in USA, Australia and UK demonstrate the close relationship between mental health problems and substance use disorders. Also, the results from Greek research projects support this close relationship, but their research designs have significant limitations. Multiple and high risks are common in this population, like violent or suicidal behavior, self-harm, physical problems, while they appear less responsive to treatment. Subsequently, different models for etiopathogenesis of dual diagnosis have been suggested: (a) Causal relationship: secondary substance use disorder is subsequent of primary mental illness (self-medication hypothesis, supersenstivity model) or vice versa (alcohol, cannabis, and cocaine use trigger or contribute to development of mental illness). (b) Third factor as the cause of both mental and substance use disorders (genetic factor, neuropathology, traumatic experience, personality characteristics, multiple factors). (c) Comorbidty is due to chance. (d) Each disorder mutually exacerbates the other, regardless the cause. Here, the relationship between alcohol and depression is discussed further as example. The ideas and the research-evidence which support each of these models are presented. Also there is an overview of different treatment models: (a) Consecutive treatment: mental health treatment and substance misuse treatment are provided consecutively. (b) Parallel treatments: the patient attends programs of both mental health and substance use services simultaneously. (c) Integrated treatment: the same clinical team addresses both mental health issues and substance use disorders. The first two models have significant weaknesses due to lack of focus on the interaction of the different disorders or to administrative and managerial barriers. On the other hand, integrated programs appear to overcome  these  limitations.  They  use modified  interventions  provided  by  the  same  team  in  order to address the multiple needs of the patients. Different types of integrated programs are discussed. More  pragmatist  inter ventions  should  also  be  taken  into  consideration.  It  is expected  that  some direction for research and clinical practice to Greece will emerge from this contribution.