As previously described, the clinical identification of MD in substance abusers constitutes a challenge for both medical care and research. Firstly, acute or chronic effects of substance use can mimic MD symptoms, making it difficult to differentiate the psychiatric symptoms that represent an independent (primary) MD from those related to an acute or chronic substance use or withdrawal. Furthermore, psychiatric diagnoses such as MD are syndromes rather than diseases with well-known pathophysiology and associated biological markers. The lack of biological markers has forced psychiatrists to develop operational diagnostic criteria, including the DSM and the ICD, and to design structured clinical diagnostic interviews in order to improve the validity and reliability of diagnoses. The use of standard criteria based on directly observable behavioral symptoms, and the incorporation of these into structured interviews, maximizes the extent to which identical information can be elicited and applied to the same criteria to achieve diagnosis. As mentioned before, methodological differences, particularly regarding the diagnostic criteria (e.g., DSM-III-R, DSM-IV, ICD-9, ICD-10) and assessment instruments used (e.g., the Structured Clinical Interview for DSM Disorders, SCID, the Psychiatric Research Interview for Substance and Mental Disorders, PRISM, and the Schedules for Clinical Assessment in Neuropsychiatry, SCAN), can also influence the prevalence rates of dual disorders (Torrens et al. 2006).

Among the assessment instruments available, the PRISM (Hasin et al. 2006) is a semi-structured interview that facilitates the distinction among independent (primary) disorder, substance-induced disorder, and the expected effects of the substance. The PRISM interview has demonstrated good psychometric properties in terms of test–retest reliability (Hasin et al. 2006), inter-rater reliability (Morgello et al. 2006), and validity (Torrens et al. 2004) to diagnose psychiatric disorders among substance users. That is to say, it is able to discern among MD independent from substance (when symptoms are substantially in excess of what would be expected given the type or the amount of the substance used or the duration of use; the onset of depressive symptoms precedes the onset of the substance use; or the symptoms persist for a period of time after the cessation of intoxication or acute withdrawal); substance-induced MD (when the episode occurs entirely during a period of heavy substance use or within the first 4 weeks after cessation of use, and the substance used is relevant to the disorder and the symptoms are greater than the expected effects of intoxication and/or withdrawal); and the expected effects (expected physiological effects of a substance, as a result of intoxication or withdrawal—e.g., insomnia—which may be identical to symptoms found in independent MD).

Distinguishing between independent and induced MD in a patient with a substance use disorder represents a challenge; nevertheless, we are able to highlight a few differential characteristics of the two forms that may be of help in clinical practice. On the one hand, a sudden change, either an increase or reduction in substance intake in the SUD, prior to the onset of depressive symptoms may indicate that the mood disorder was induced by the SUD. On the other hand, emergence of depressive symptoms during a period of stable consumption, or the persistence of depressive symptoms after clinically relevant withdrawal, probably suggests an independent MD disorder. Furthermore, in the absence of substance use, the presence of a previous history of depression or familial antecedents may indicate the existence of MD. In addition, patients with independent MD are more likely to have a history of good response to antidepressant treatments (Table 8.2). Some patients can present both independent and induced MD and undergo an increasing number of depressive symptoms throughout their lives. They are more frequently found with comorbid anxiety disorders, and are more likely to have attempted suicide (Torrens et al. 2011b).

It is important to highlight that the studies that distinguish between independent and induced MD have found a clearly higher prevalence of independent MD (Torrens et al. 2011a). Furthermore, recent data from one of the most representative epidemiological study in the United States, the National Epidemiologic Survey on Alcohol and Related Conditions (NESARC), are shedding light on a new perspective with respect to substance dependence and comorbid MD (Blanco et al. 2012). The subgroup of patients diagnosed at an earlier stage with induced MD or dysthymia together with SUD, when reevaluated at a second time point 3 years later, were reclassified as being affected by independent MD. Also, in a follow-up study of a Spanish cohort of ecstasy users, most of the induced MD diagnosed at the baseline t of the study were diagnosed as independent MD at a 3-year follow-up (Martin-Santos et al. 2010). This may have been due to a number of factors including a higher probability of being diagnosed with an induced affective disorder when severe drug dependence symptoms are present, or that the diagnosis of induced disorders captures subjects with a preexisting higher risk for MD whose symptoms are precipitated by substance use. It may also indicate that SUD had precipitated an MD whose erroneous diagnosis had been masked by the substance consumption (Magidson et al. 2013).

Clinical data indicate that people affected by MD present a higher vulnerability to developing a SUD, and individuals with SUD have a higher risk of developing MD when compared to the general population. Furthermore, co-occurrence of SUD and MD is a predictor of clinical severity: patients show a more severe clinical course, less response to treatment, and a poorer prognosis for both disorders overall (Boschloo et al. 2013; Conner et al. 2012; Samet et al. 2013). These dually diagnosed patients additionally present a higher prevalence of attempted/completed suicide than those with only one disorder (Conner et al. 2012; Marmorstein 2011; Blanco et al. 2012).

Besides MD, patients with comorbid SUD often manifest or develop other medical, psychiatric, and substance use comorbidities, thus making treatment even more challenging. As expected from such a severe clinical picture, dual disorder patients have considerable psychosocial disability and require an increased utilization of health care resources, including emergency rooms and psychiatric hospitalization (Mueller et al. 1994; Martín-Santos et al. 2006; Pettinati et al. 2013; Samet et al. 2013).

The main results coming from the systematic reviews and meta-analyses of comorbid MD and SUD randomized clinical trials (Nunes and Levin 2004; Torrens et al. 2005; Pani et al. 2010) indicate that (1) antidepressant drugs improve comorbid depression with alcohol dependence but not the depression that concurrently occurs with cocaine or opiate dependence. Furthermore, the improvement of depression, together with alcohol dependence, takes place only with imipramine, desipramine, and nefazodone; while selective inhibitors of serotonin reuptake (SSRIs) are not effective, (2) treating depressed substance-dependent patients with antidepressants does not directly improve substance use. When the antidepressants improve depressive symptoms, there is also a quantitative reduction in the use of the substance of abuse, but no effect on the acquisition of abstinence or total remission of these substances use. Thus a specific and concomitant treatment for SUD is required. In a recent trial for comorbid MD and alcohol dependence, a combined treatment of a medication for depression (sertraline) and another for alcohol dependence (naltrexone) was found to simultaneously reduce depressive symptoms and excessive drinking (Pettinati et al. 2010).

An additional concern when treating these dual disorder patients is the safety of the treatment itself due to the frequency of comorbid physical illness (e.g., HIV and/or HVC infections, hepatic cirrhosis) and the risk of interactions with other drugs that the person may be taking (e.g., risk of QTc prolongation in HIV-infected patient receiving methadone maintenance treatment and SSRI) (Funk and Bostwick 2013; Vallecillo et al. 2013). The main interactions and general recommendations about the clinical management of patients with MD and SUD are summarized in Table 8.3. Besides aspects of efficacy, safety of antidepressant use, and possible interactions with the consumption of various substances or other drugs, the potential for the abuse of the different drugs used for depression treatment should also be taken into account. A review conducted by Haddad suggested that antidepressants have no potential for dependence with the exception of tranylcypromine or amineptine for their dopaminergic effects and stimulant properties (Haddad 1999).