**Dual Diagnosis Patients First Admitted to a Psychiatric Ward for Acute Psychiatric Patients: 2-Year Period 2003– 2004 versus 2013–2014**

[96] Kelly, J.F., Green, M.C., Bergman, B.C. (2016). Recovery benefits of the "therapeutic alliance" among 12-step mutual-help organization attendees and their sponsors. *Drug*

[97] Humphreys, K., Blodgett, J.C., Wagner, T.H. (2014). Estimating the efficacy of alcohol‐ ics anonymous without self-selection bias: an instrumental variables re-analysis of randomized clinical trials. *Alcoholism: Clinical and Experimental Research*, 38, 2688–2694.

*and Alcohol Dependence*, 162, 64–71. doi:10.1016/j.drugalcdep.2016.02.028.

doi:10.1111/acer.12557.

168 Recent Advances in Drug Addiction Research and Clinical Applications

Carla Gramaglia, Ada Lombardi, Annalisa Rossi, Alessandro Feggi, Fabrizio Bert, Roberta Siliquini and Patrizia Zeppegno

Additional information is available at the end of the chapter

http://dx.doi.org/10.5772/63102

#### **Abstract**

Dual diagnosis (DD) is the coexistence of severe mental illness (SMI) and substance use disorder (SUD). The increase of DD observed in recent years has important implications for mental health services organization. The aim of this study is to assess the prevalence and features of DD over a decade, comparing the periods 2003–2004 and 2013–2014. We performed a retrospective study retrieving sociodemographic and clinical data from the medical records of patients at their first admission to the Psychiatric Ward of University Hospital "Maggiore della Carità" in Novara, Italy. Patients with SMI and comorbid SUD (SMI-SUD) and patients with SMI without comorbidity (SMI) were compared in the two periods, 2003–2004 versus 2013–2014. SMI-SUD patients in both 2-year periods were more likely to be male, younger, unemployed, living with parents (or alone, for the 2013–2014 period) rather than with a family of their own, and single (or divorced, in 2003–2004). The 2003–2004 patients were more frequently diagnosed with a personality disorder, whereas the 2013– 2014 patients had mixed diagnoses. We have found differences in the possible predictors of substance abuse in the two periods as well: in both periods, male gender was associated with an increased risk of DD, whereas age >61 years was associated with decreased risk. Only in the first period (2003–2004) was having a university degree associated with a decreased risk of DD, whereas the diagnosis of a personal‐ ity disorder was associated with an increased risk of DD; on the contrary, in the second period (2013–2014), living in a protective environment was associated with a decreased risk of DD. The identification of changes in the prevalence of first admission DD patients and their clinical and sociodemographic features may help to highlight an evolving pattern of substance use and to identify possible risk factors that may be the target of prevention and treatment approaches.

© 2016 The Author(s). Licensee InTech. This chapter is distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/3.0), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.

**Keywords:** dual diagnosis, first admission, inpatients, substance use disorder, addic‐ tion

### **1. Introduction**

#### **1.1. Addiction**

The pathway that leads to addiction is characterized by specific steps: the initiation of sub‐ stance use, the established use, and, finally, the development of addiction. Several factors are involved in this process, including genetic and environmental ones. The availability of the substance may play a role in each stage in the development of addiction, whereas the accessi‐ bility of a substance seems relatively more important in the initiation of substance use [1]. Evidence from family, adoption, and twin studies converges on the relevance of genetic factors in the development of addiction according to a complex model of inheritance and clinical and genetic heterogeneity [2–10]. The role of genetic, sociocultural, biological, and other factors, including drug availability, peer influence, social support, and type and psychoactive proper‐ ties of the drug, varies across the lifespan and in different stages of the addiction process. Briefly, it seems that environmental factors (such as peer influences and family environment) have a stronger effect on exposure and initial pattern of use, whereas genetic factors play a major role in the transition from regular use to the development of addiction [11,12]. Moreover, it should not be overlooked that genetic factors underlying addiction may overlap to various degrees with those underlying other psychiatric disorders. For instance, studies focusing on the role of the COMT gene polymorphism in the genetic predisposition to mental disorders [including severe mental illness (SMI), such as schizophrenia, bipolar disorder, obsessive-compulsive disorder, anorexia nervosa, and attention-deficit hyperactivity disorder] have found that the same polymorphism may be involved in the pathogenesis of addiction and substance use disorder (SUD). Probably COMT increases susceptibility to mental disorders in general, whereas other genetic or environmental factors may influence the development of specific disorders, including SUD [13].

#### **1.2. Dual diagnosis (DD)**

The World Health Organization defines DD as the co-occurrence, in the same person, of a severe mental health condition (SMI) with a drug abuse or dependence disorder (SUD).

In 1993, First and Gladis [14] proposed to classify DD patients as follows: (1) main psychiat‐ ric disorder and secondary drug dependence, (2) main SUD and secondary psychiatric disorder, and (3) main psychiatric disorder and drug dependence. Discriminating among these three options may be particularly challenging in clinical settings, where it can be hard to understand whether it is the SMI that induced drug consumption or the drug consumption that induced or worsened the SMI.

**Keywords:** dual diagnosis, first admission, inpatients, substance use disorder, addic‐

The pathway that leads to addiction is characterized by specific steps: the initiation of sub‐ stance use, the established use, and, finally, the development of addiction. Several factors are involved in this process, including genetic and environmental ones. The availability of the substance may play a role in each stage in the development of addiction, whereas the accessi‐ bility of a substance seems relatively more important in the initiation of substance use [1]. Evidence from family, adoption, and twin studies converges on the relevance of genetic factors in the development of addiction according to a complex model of inheritance and clinical and genetic heterogeneity [2–10]. The role of genetic, sociocultural, biological, and other factors, including drug availability, peer influence, social support, and type and psychoactive proper‐ ties of the drug, varies across the lifespan and in different stages of the addiction process. Briefly, it seems that environmental factors (such as peer influences and family environment) have a stronger effect on exposure and initial pattern of use, whereas genetic factors play a major role in the transition from regular use to the development of addiction [11,12]. Moreover, it should not be overlooked that genetic factors underlying addiction may overlap to various degrees with those underlying other psychiatric disorders. For instance, studies focusing on the role of the COMT gene polymorphism in the genetic predisposition to mental disorders [including severe mental illness (SMI), such as schizophrenia, bipolar disorder, obsessive-compulsive disorder, anorexia nervosa, and attention-deficit hyperactivity disorder] have found that the same polymorphism may be involved in the pathogenesis of addiction and substance use disorder (SUD). Probably COMT increases susceptibility to mental disorders in general, whereas other genetic or environmental factors may influence the development of specific disorders,

The World Health Organization defines DD as the co-occurrence, in the same person, of a severe mental health condition (SMI) with a drug abuse or dependence disorder (SUD).

In 1993, First and Gladis [14] proposed to classify DD patients as follows: (1) main psychiat‐ ric disorder and secondary drug dependence, (2) main SUD and secondary psychiatric disorder, and (3) main psychiatric disorder and drug dependence. Discriminating among these three options may be particularly challenging in clinical settings, where it can be hard to understand whether it is the SMI that induced drug consumption or the drug consumption

tion

170 Recent Advances in Drug Addiction Research and Clinical Applications

**1. Introduction**

including SUD [13].

**1.2. Dual diagnosis (DD)**

that induced or worsened the SMI.

**1.1. Addiction**

Consistent with the first class proposed by First and Gladis (psychiatric disorder first and subsequent drug dependence), the study of the psychological attitudes in addiction disor‐ ders has lead to the self-medication hypothesis (SMH) proposed by Khantzian [15]. The SMH primarily derives from clinical observations and posits substance dependence as a compen‐ satory means to modulate emotions and aimed at self-soothing in response to distressing psychological states. According to this hypothesis, drugs would become addicting because of their power to alleviate, counteract, or modulate psychological suffering; there would also be a considerable degree of specificity in a person's choice of drugs because of unique psycho‐ logical and physiological effects [15]. Hence, emotional states and distress, as well as expect‐ ancy of positive affective modifications, would be associated with substance use or relapse in people with SUD [16]. The SMH has received a variable empirical support, particularly as far as drug specificity is concerned [17]. The choice of type of drug according to the SMH is sometimes counterintuitive. For instance, while DD bipolar patients seem to use substances to maintain a euphoric state or soothe a depressive suffering, depressed patients, who might be expected to choose stimulants as well, often turn to depressants such as alcohol. Schizophren‐ ic patients are often strongly nicotine addicted, and their heavy smoking may be an attempt to alleviate cognitive deficits and to reduce extrapyramidal side effects induced by antipsy‐ chotic medication, through the effect of nicotine on dopaminergic activity [18]. Briefly, the choice of a particular type of substance could depend on the symptoms that patients wish to relieve, as well as on substance availability, and on patients' basic personality traits [19]. Moreover, clinical experience in the last years has highlighted clear and ongoing changes in the choice of type of drug [from heroin and lysergic acid diethylamide (LSD) to the current socalled "smart drugs", vegetable or synthetic origin compounds with psychoactive effects that are not yet considered illegal]. For instance, changes in the prevalence of commonly used substances in people with DD have been discussed in two studies that estimated the preva‐ lence of SUD in patients with SMI in Philadelphia [20,21]. These studies found a shift from cannabis being the most commonly used drug to cocaine and associated changes in demo‐ graphic correlates. Although these data date back to more than 20 years ago, they highlight the changing pattern of SUD [22,23].

Consistent with these results, in Italy, in the last two decades, significant changes in the main drug used by patients attending addiction services have been noticed. Although the use of alcohol has remained relatively stable, there has been a significant decrease in the use of heroin and an opposite tendency regarding the use of cocaine; moreover, the use of "smart drugs" and polyabuse (i.e., the abuse of several substances simultaneously) has increased in an alarming way. The reasons underlying the use of new drugs may be different from those guiding the first heroin addicts. For instance, ecstasy has been generally chosen by adoles‐ cents and young adults for its entactogenic properties (the stimulation and enhancement of feelings of empathy, love, and presumed emotional closeness to others) [24,25].

As already suggested, personality may play a relevant role in the choice of the drug and expectations concerning the desired effect of the drug itself. The motivation leading to drug abuse may span from alleviating boredom and active search for pleasure, improvement of attention and performance, to reducing tension and decreasing mental illness symptoms. Hence, personality traits should be assessed to improve the understanding of the complex relation between patients, drugs, and environment [26].

According to Cloninger's model [27,28], personality consists of temperament and character traits. Temperament is defined as a biological disposition reflected by relatively stable features related to mood, attitudes towards the environment, and reactivity to external and internal stimuli, including variability and intensity in emotional dispositions [29,30], whereas charac‐ ter is based on mechanisms that are developed through life experience. Although only a few studies have investigated personality dimensions in DD patients with the model developed by Cloninger, most of these found an association with high scores on the temperamental dimension of "novelty seeking" [31,32] as well as with high scores on "harm avoidance" [31]. According to Cloninger's descriptions of these temperament dimensions, we may expect patients with high scores on "novelty seeking" using drugs to search pleasure and to escape boredom and patients with high scores on "harm avoidance" using substances to achieve relief from tension and unpleasant or painful emotional states [33].

#### **1.3. Epidemiology and clinical features of DD patients**

Comorbidity between drug and/or alcohol dependence and a SMI is highly prevalent, and clinicians should be aware that patients asking for a psychiatric advice are likely to conceal their problems related to use and/or abuse of substances, unless specifically asked about them [34,35], possibly leading to an underestimation of DD. Recent studies report that the prevalence of DD in patients attending mental health service and substance misuse services ranges between 55% and 85% [36,37], and an association between SUD and mood and anxiety disorders has been supported by epidemiological and clinical studies [38–41]. In 2014, in the United States, among adults with a past year SUD, 39.1% had a comorbid psychiatric disor‐ der, whereas, among adults without a past year SUD, only 16.2% had a SMI [42]. Moreover, epidemiological studies consistently describe a gender difference in DD patients as far as diagnosis is concerned; overall, DD is more common in males, and male patients usually suffer from psychotic and bipolar disorders, whereas depression and anxiety are more represented in women who, on the contrary, represent a smaller percentage of DD samples [43,44].

Many longitudinal or cross-sectional studies tried to identify recurrent and significant sociodemographic features and pattern/type of abuse typical of DD patients. The literature suggests that DD patients, compared to those with SMI but no comorbidity with SUD, are usually younger, males, with a lower level of education, often unemployed, still living with parents rather than with a family of their own, and with an overall lower social functioning [45–47]. DD is related to worst compliance to treatments, higher relapse rates and health services usage, more functional disability, and cognitive as well as psychological, physical, and social impairment [37,43,47,48]. Overall, DD patients show a poorer quality of life and reduced life expectancy [49] compared with patients with SMI or patients with SUD with no other comorbid psychiatric disorder.

The treatment of DD patients is particularly challenging because of the poor compliance and significant deterioration in social functioning that often occurs in these patients. They are more likely to suffer from comorbid medical conditions; as described above, they may experience more difficulties in family relationships and troubles maintaining a stable job and financial situation; moreover, they may have legal problems related to the behavioral consequences of substance use and/or to the illegal attempts to obtain the substance. Compared to schizophre‐ nia patients who do not use alcohol and drugs, patients with DD tend to have an earlier age of onset, more frequent and sometimes longer periods of hospitalization, more severe depressive and psychotic symptoms, more episodes of suicidal and violent behavior, more legal and financial problems, and higher mortality risk [50–56]. Likewise, bipolar patients with DD have 6.4-fold risk for violent crime compared with bipolar patients without comorbidity [57,58]. Overall, because they are less likely to adhere to their medication regimen, patients with DD also are at an increased risk of relapse and re-hospitalization [59,60]. However, even in those who do adhere to their medication, commonly abused substances can trigger or exacerbate psychiatric symptoms, eventually leading to relapse and need for inpatient treatment.

Hence, personality traits should be assessed to improve the understanding of the complex

According to Cloninger's model [27,28], personality consists of temperament and character traits. Temperament is defined as a biological disposition reflected by relatively stable features related to mood, attitudes towards the environment, and reactivity to external and internal stimuli, including variability and intensity in emotional dispositions [29,30], whereas charac‐ ter is based on mechanisms that are developed through life experience. Although only a few studies have investigated personality dimensions in DD patients with the model developed by Cloninger, most of these found an association with high scores on the temperamental dimension of "novelty seeking" [31,32] as well as with high scores on "harm avoidance" [31]. According to Cloninger's descriptions of these temperament dimensions, we may expect patients with high scores on "novelty seeking" using drugs to search pleasure and to escape boredom and patients with high scores on "harm avoidance" using substances to achieve relief

Comorbidity between drug and/or alcohol dependence and a SMI is highly prevalent, and clinicians should be aware that patients asking for a psychiatric advice are likely to conceal their problems related to use and/or abuse of substances, unless specifically asked about them [34,35], possibly leading to an underestimation of DD. Recent studies report that the prevalence of DD in patients attending mental health service and substance misuse services ranges between 55% and 85% [36,37], and an association between SUD and mood and anxiety disorders has been supported by epidemiological and clinical studies [38–41]. In 2014, in the United States, among adults with a past year SUD, 39.1% had a comorbid psychiatric disor‐ der, whereas, among adults without a past year SUD, only 16.2% had a SMI [42]. Moreover, epidemiological studies consistently describe a gender difference in DD patients as far as diagnosis is concerned; overall, DD is more common in males, and male patients usually suffer from psychotic and bipolar disorders, whereas depression and anxiety are more represented in women who, on the contrary, represent a smaller percentage of DD samples [43,44].

Many longitudinal or cross-sectional studies tried to identify recurrent and significant sociodemographic features and pattern/type of abuse typical of DD patients. The literature suggests that DD patients, compared to those with SMI but no comorbidity with SUD, are usually younger, males, with a lower level of education, often unemployed, still living with parents rather than with a family of their own, and with an overall lower social functioning [45–47]. DD is related to worst compliance to treatments, higher relapse rates and health services usage, more functional disability, and cognitive as well as psychological, physical, and social impairment [37,43,47,48]. Overall, DD patients show a poorer quality of life and reduced life expectancy [49] compared with patients with SMI or patients with SUD with no

The treatment of DD patients is particularly challenging because of the poor compliance and significant deterioration in social functioning that often occurs in these patients. They are more likely to suffer from comorbid medical conditions; as described above, they may experience

relation between patients, drugs, and environment [26].

172 Recent Advances in Drug Addiction Research and Clinical Applications

from tension and unpleasant or painful emotional states [33].

**1.3. Epidemiology and clinical features of DD patients**

other comorbid psychiatric disorder.

Although, in the last years, several studies have been performed about early detection and treatment of DD [59–64], there is a dearth of studies focused on the changes in the preva‐ lence of DD in psychiatric inpatients [65]. A study about comorbidity with SUD in psychiat‐ ric inpatients performed in Spain [66] found 24.9% of inpatients having a SUD as well as another psychiatric disorder. Consistent with the literature about DD, a statistically signifi‐ cant predominance of men was found in the DD group together with younger age at the time of their first psychiatric admission; the most common diagnoses in this group were schizo‐ phrenia or related psychoses, although patients with SMI only had mostly affective disor‐ ders. As described above for preferred substance of abuse, the most used was alcohol followed by cannabis and cocaine.

Another interesting study assessing the trends in the incidence and demographic and clinical correlates of DD among patients whose first psychiatric hospitalization occurred between 1996 and 2010 was performed in Israel by Ponizovsky et al. in 2015 [67]. Based on the literature and their clinical experience, they hypothesized an increase of the proportion of DD among all first psychiatric hospitalizations during the study period due to the increasing prevalence of substance-related disorders in the general population, an increased vulnerability to DD on behalf of specific population groups (e.g., new immigrants) [68–70], and, lastly, higher DD rates in involuntarily hospitalized patients because they may be more likely to show epi‐ sodes of suicidal and violent behavior [52] compared to voluntarily admitted patients. Over the study period, DD with drugs decreased from 1996 to 2010, whereas DD with alcohol and DD with both drugs and alcohol increased. The changing pattern of DD over time was supported as well as most findings concerning the sociodemographic features of DD pa‐ tients reported in the literature. The positive predictors of DD with alcohol were male gender, previous suicide attempt, compulsory hospitalizations, and marital status. DD with alcohol was found mainly in immigrants, whereas DD with drugs was more common in the native population.

#### **1.4. Challenges and perspectives in research**

Treatment of DD patients requires a thorough understanding of both mental illness and addiction and the consequent integration of the traditional treatment approaches in both the mental health and addiction treatment fields [71].

In the research field, a complicating issue for DD studies is that they may focus on different populations: the general public, the population of subjects referring to psychiatric services, and the population of people currently treated by addiction services [72]. This diversity affecting the research field may be a concrete, challenging reality from a clinical standpoint. In Italy, this is particularly important because the standard practice for patients with comor‐ bid psychiatric disorders and SUD is a parallel treatment. In our country, mental health and addiction facilities have different institutional cultures, etiological concepts, administrative arrangements, screening, and treatment approaches [73]. The problematic issues of such treatment approach include possible flaws in communication, collaboration, and linkage, which might significantly hinder or complicate comorbidity service delivery [74,75].

Even if it is clearly a changing and growing problem, the number of studies on DD preva‐ lence in patients admitted to psychiatric wards in general hospitals in Italy is still scant. A recent study [76] has focused on differences in the length of stay in first-hospitalization schizophrenic patients with and without comorbid SUD and found that the first showed poorer symptom improvement and required longer stays than the latter. The flaws of communica‐ tion and linkage between psychiatric and addiction services emerged from the study by Preti et al. [77], who reported that only approximately 30% of patients with SUD discharged from acute psychiatric inpatient facilities were referred to drug addiction services. Other issues that have been investigated in this field include SUD in emergency room settings [78], gender differences in DD patients [65], and attempts to understand whether SUD follows or pre‐ dates the psychiatric diagnosis [79,80].

Considering these premises, the aim of our study was to describe the sociodemographic and clinical features of DD patients at their first admission to the Psychiatric Ward of University Hospital "Maggiore della Carità" in Novara, Italy. With more detail, we collected data about all patients admitted for the first time during the 2-year periods 2003–2004 and 2013–2014 to (1) assess the extent of comorbidity with drug abuse in a sample of patients at their first admission to a psychiatric ward in a general hospital in Italy; (2) investigate whether there are differences between inpatients with and without comorbid SUD, focusing on sociodemo‐ graphic, clinical, and other background variables in both periods; (3) investigate the possible differences between patients with comorbid SUD in the two 2-year periods; and (4) identify the possible predictors of comorbidity with SUD and their changes over a decade.

## **2. Methods**

We performed a retrospective study reviewing the clinical charts of patients at their first admission to the Psychiatric Ward of University Hospital "Maggiore della Carità" in Novara, Italy. We assessed two 2-year periods: 2003–2004 and 2013–2014. We excluded the records of patients with a diagnosis at discharge of SUD with no comorbid psychiatric disorder. Because our interest was to focus on DD, we collected data about patients with comorbid psychiatric and SUD diagnosis (later on described as SMI-SUD) and about psychiatric patients without comorbid SUD (SMI).

The following information was retrieved from the clinical charts: (1) sociodemographic data, including age, sex, education, occupational status, living accommodation, marital status, and legal problems and (2) clinical and psychopathological history, information concerning drug use, history of self-harm (including suicidal and parasuicidal behaviors), and history of aggressive behaviors and acting out.

Psychiatric diagnoses were made during the hospital stay by experienced psychiatrists with the aid of the Structured Clinical Interview I [81] and II [82] for Axis I and Axis II disorders, respectively. According to the International Classification of Diseases [83], diagnoses were the following: schizophrenia and other psychoses, affective disorders, anxiety disorders, and personality disorders; disturbance of conduct, mental retardation, eating disorders, acute stress reaction, and adaptation reaction were grouped as "other diagnoses".

Information about the use of psychotropic drugs was collected by the treating psychiatrist during inpatient treatment, including age at first use and type of substance (alcohol, psychi‐ atric drugs, cannabis, heroin, and cocaine; methamphetamine, ketamine, phencyclidine, LSD, butyl nitrite, amyl nitrite, and γ-hydroxybutyric acid were grouped together as "other drugs"). As for diagnosis, these data were then gathered for research purposes from clinical charts. The research project was approved by the Institutional Review Board of Università del Piemonte Orientale.

Statistical analyses were performed using STATA 11 (Stata Corp., College Station, TX, 2011). Initial descriptive statistics included the *χ*<sup>2</sup> test to evaluate the differences in proportions between groups (SMI-SUD vs SMI patients in the two periods). Then, a multivariate analysis was performed using a logistic regression to assess the potential predictors of substance abuse. The covariates included in the final model were selected using a stepwise forward selection process, with a univariate p<0.25 as the main criterion [84]. Results are expressed as odds ratio (OR) with 95% confidence interval (CI). A two-tailed p<0.05 was considered significant for all analyses.
