Substance Use Disorders and Schizophrenia

Public Health and Clinical Significance of Comorbid Schizophrenia and Substance Use Disorders

Estimates of the lifetime prevalence of substance use disorders among people with schizophrenia range from 47% to 70% (1–3), and are as high as 80% or more when tobacco use is included (4). The most commonly used substances among patients with schizophrenia are tobacco, alcohol, cannabis, and cocaine (5).

Substance use disorders cause significant morbidity and mortality in those with schizophrenia. Patients with schizophrenia who also have alcohol-related and other substance use disorders tend to have significantly more positive and fewer negative symptoms than patients without substance use disorders (6). People with schizophrenia and co-occurring substance use disorders have worse outcomes in terms of course of illness, engagement in treatment, violence, homelessness, legal problems, life functioning, and physical illnesses (7). In addition, patients with schizophrenia and co-occurring substance use disorders may experience more frequent and longer hospitalizations (8) and increased suicidal ideation and behaviors (9).

It is difficult to differentiate substance-induced psychiatric disorders from primary mental illnesses. Traditionally, it was thought that treatment of the substance use disorders had to be completed before treatment of the psychiatric illness. The theory was that, after a period of abstinence, psychiatric symptoms would remit and perhaps obviate the need for psychotropic medications. However, the wisdom of this approach has since been challenged and refuted. Many studies have shown that for patients with diagnoses of a serious mental disorder plus a substance use disorder, treatment of the psychiatric disorder improves the outcome of the psychiatric illness and sometimes that of the substance use disorder as well (10). It is also true that treatment of the substance use disorder can decrease the severity of the comorbid psychiatric illness (11). Because treatment of substance use disorders can affect outcomes, accurate diagnosis and initiation of treatment are especially important for people with schizophrenia.

Understanding Substance Use Disorders

Neurobiology of Substance Use Disorders

Over the past 30 years, our understanding of the neurobiology of addiction has expanded significantly. The basic brain reward circuitry and pathways have been described, and understanding of this system continues to be refined. This reward pathway presumably evolved to encourage the preservation of the species by stimulating positive emotional experiences in response to the intake of food and engaging in sex, social interactions, and other evolutionarily advantageous behaviors. Multiple substances of abuse can directly or indirectly stimulate this pathway. Specifically, natural rewards and drugs of abuse both cause stimulation of neurons in the ventral tegmental area, which causes the release of dopamine in the nucleus accumbens. More recently, brain circuitry involved in motivation, learning, decision making, and impulse control has been implicated in drug use and addiction (12, 13). As our understanding of these systems has evolved, we have begun to identify targets for pharmacological intervention and treatment of substance use disorders.

Diagnostic Criteria for Substance Use Disorders

The fourth edition of the Diagnostic and Statistical Manual of Mental Disorders (DSM-IV) made the distinction between substance abuse and substance dependence as separate disorders, with substance abuse consisting of the continued use of a substance despite problems in one or more domains of life, whereas substance dependence emphasized loss of control, pervasive use, and physiological tolerance and withdrawal phenomena (14). However, DSM-5 removed this distinction and instead defined substance use disorder as a single entity combining the criteria of both abuse and dependence, with a range of severity. The DSM-5 Substance-Related Disorders Work Group also removed the abuse criterion involving legal problems and added the new criterion of craving. As such, a DSM-5 diagnosis of substance use disorder requires two or more of 11 criteria, with two to three criteria indicating a mild disorder, four to five indicating a moderate disorder, and six or more criteria indicating a severe disorder (15, 16).

Understanding Comorbidity: Why Are Substance Use Disorders so Common in Schizophrenia?

Epidemiologic comorbidity is a nonrandom association (greater or less than that expected due to chance) between two disorders in a defined population, and it can arise through one or a combination of the following three processes: first, the two disorders share common etiologic factors (including other disorders) or common diatheses (including genetic risk factors); second, one disorder causes the other; and third, the relationship is mediated by another condition or disorder (17). When we consider the relationships between schizophrenia and substance use disorders, the first two processes are most compelling. It is important to remember that the relationships between the two disorders may differ between substances and at various stages of each illness. Substance use disorders are not a single disorder but a group of disorders describing multiple patterns of misuse of a variety of pharmacologically distinct compounds. For instance, different subtypes of alcoholism have been demonstrated, which have different patterns of inheritance, comorbidity, and response to pharmacotherapy (18). Although it is not clear that patients with schizophrenia are predisposed to misuse of any one class of drug over another, it remains a possibility that the observed comorbidity is due to an association between schizophrenia and particular types of substance use disorders. Schizophrenia is also a heterogeneous construct, so it is also possible that different variants of schizophrenia may have disparate relationships to substance use disorders.

Common Etiologic Factors?

If schizophrenia and substance use disorders were different aspects of the same underlying process (disease or disorder), there would be a common etiology or pathophysiological mechanism. There is considerable overlap in the regions of the brain shown to be involved in these disorders. Neurotransmitter systems, including dopamine, serotonin, and glutamate (N-methyl-d-aspartate [NMDA] receptors, in particular) have also been implicated in both addictive disorders and schizophrenia (19). Although genetic studies have failed to demonstrate any common genetic basis for schizophrenia and certain substance use disorders such as alcohol use disorder (20), genetic linkages have been identified for psychosis and use of other specific substances, especially cannabis (21).

Ways in Which Schizophrenia May Cause or Exacerbate Substance Use Disorders

Self-medication.

Self-medication is an attractive and frequently invoked theory stating that individuals with psychiatric disorders use substances in an effort (successful or unsuccessful) to alleviate their psychiatric symptoms (22). It follows that there should be specificity in the relationship between the psychiatric symptoms and the particular substance(s) used. In general, research has not been able to confirm such specific relationships, but it remains possible that self-medication is driving drug use in certain psychiatric diagnoses or subpopulations within diagnoses.

Although patients with schizophrenia have much higher rates of substance use disorder than people without psychiatric diagnoses, the substances commonly used by patients with schizophrenia appear to be the same as those commonly used by others. There have been reports of a selective elevation in the rate of stimulant use disorders among people with schizophrenia (23). Although it may seem counterintuitive that people with psychosis would preferentially use psychotogenic drugs, it has been proposed that stimulants help patients with schizophrenia by ameliorating negative symptoms such as anhedonia and anergia. Although this may happen in some cases, more recent research has shown high amounts of variation in studies assessing rates of stimulant use among people with psychosis (23). The case for self-medication with nicotine is more persuasive. People with a diagnosis of schizophrenia have strikingly high rates of tobacco use disorders compared with both the general population and people with other psychiatric disorders, and nicotine appears to ameliorate cognitive and sensory gating deficits in schizophrenia (24). However, it is difficult to prove that this is actually why people with schizophrenia smoke.

Use of drugs because of abnormal functioning of brain reward systems.

Because of the lack of specificity of substances used by people with schizophrenia, more general hypotheses have been developed to attempt to explain why these patients use drugs. One of the most persuasive explanations invokes abnormalities in neural reward circuitry to explain the propensity of patients with schizophrenia to misuse substances. Because of abnormalities in the reward system, patients with psychotic disorders are hypothesized to experience less reward in response to naturally occurring reinforcement. Hence, the rewarding effects of drugs are more salient to people with schizophrenia because of what has been termed reward deficiency syndrome (25), which may make people with schizophrenia more sensitive to the rewarding effects of substances and less able to control impulses to use drugs (19). All common drugs of abuse reliably cause the release of dopamine in the nucleus accumbens, which is thought to be responsible for at least some of the rewarding and pleasurable effects of drug use. Since antipsychotic drugs block dopamine receptors, using these drugs could presumably also cause reward deficit and heighten the risk of addiction. If this were true, one might expect lower rates of substance use among patients on lower potency or second-generation antipsychotic medications than on higher potency, first-generation antipsychotic medications. While it is true that all antipsychotic medications (with the possible exception of clozapine) work by blocking dopamine receptors, there are differences between agents in both affinities for dopamine-2 receptors and potentials to mitigate the effects of dopamine blockade via serotonin receptor antagonism. Indeed, uncontrolled studies of patients on clozapine have shown decreases in substance use—including tobacco—over time on this medication (10, 26, 27). However, there have not been any large randomized trials designed to compare changes in drug or alcohol use among patients with both schizophrenia and substance use disorders on different antipsychotic medications, and results from smaller studies—although often favoring second-generation antipsychotics—have been inconsistent (10, 27). In a secondary analysis of data from the Clinical Antipsychotic Trial of Intervention Effectiveness (CATIE) Study contrasting the effects of four second-generation antipsychotics and one first-generation medication, substance use decreased significantly over the 18 months of the trial, but substance use outcomes were not significantly different between the various antipsychotics (28).

Assessment of Substance Use Disorders

Substance use disorders are underdiagnosed among patients with schizophrenia, so all patients in treatment for schizophrenia should be screened for co-occurring substance use disorders. Screening tools can be an effective first step in determining whether a patient has a problem with substances. Multiple tools exist to screen for substance use disorders, and a recent analysis of the literature has shown that the CAGE Questionnaire, the Michigan Alcoholism Screening Test, the Alcohol Use Disorders Identification Test (AUDIT-C), and the 10-item Drug Abuse Screening Test (DAST-10) all have fairly good sensitivity and reliability when used in this population (29, 30). The Alcohol, Smoking and Substance Involvement Screening Test screens for all substances of abuse and has been shown to be reliable and valid among first-episode psychosis patients (31). The AUDIT-C (32) and the DAST-10 (33) are particularly efficient screening tools. These and many other screening tools can be accessed online on the Substance Abuse and Mental Health Services Administration’s website: http://www.integration.samhsa.gov/clinical-practice/screening-tools.

A positive screen must be followed by a more thorough assessment of the severity of the substance use problem. A good substance use history includes the following elements: substances used, route of administration, age at first use, period of heaviest use, frequency of use and amount used at heaviest use, frequency of use and amount used at present, date of last use, and any periods of abstinence. A thorough assessment will obtain these parameters for each substance used. Because patients will often minimize and deny substance use, it is important to ask about alcohol and use of illicit substances in a nonjudgmental way. In addition, the assessment should utilize information from multiple sources, such as family members, therapists, and case managers. Random urine drug and alcohol screens may provide important additional information.

Treatment of Substance Use Disorders and Schizophrenia

Models of Treatment

Traditionally, separate systems have existed in the United States for treatment of mental illnesses and addictions. This has greatly complicated the treatment of individuals with co-occurring disorders. Patients with psychiatric and substance use disorders were often excluded from mental health programs, and those seeking drug and alcohol treatment were frequently not allowed into addiction treatment programs if they had a serious mental illness. Since the 1980s, there has been increasing awareness of the frequency of co-occurring disorders and a corresponding bolstering of efforts to modify treatment systems to better accommodate those who have them.

At the most general level conceptually, there are three possible models of treatment for co-occurring disorders: the sequential, parallel, and integrated treatment models. In the sequential model, the disorders are treated one at a time in separate programs. This is typically how treatment was provided in the past. It was frequently reinforced by a lack of training and experience on the part of program staff in both settings. For those with a serious mental illness, this model is problematic. Because schizophrenia is usually a chronic condition, treating it first would result in never treating the addiction. Treating the addiction first would result in frequent exacerbations of and relapses from untreated schizophrenia.

The parallel model recognizes that both disorders should be treated simultaneously but keeps treatment of the mental disorder and the substance use disorder separate. Although this is a major improvement over the sequential model, splitting the treatment can result in poor coordination, inconsistency in the treatment, and logistical burden on patients, particularly if the two treatments are at separate locations. However, with careful coordination, parallel treatment can be effective.

Integrated treatment models were developed in the 1980s and 1990s and continue to be elaborated today. In integrated dual-diagnosis treatment (IDDT), psychiatric and substance use disorders are treated in a single program by a team of clinicians with expertise in both sets of disorders. IDDT models have typically emphasized outreach into the community through services such as case management. Substance use disorder treatments include both pharmacotherapies and psychosocial treatments, which typically use established, evidence-based addiction treatment models but may be modified for use with co-occurring disorders. Although outcomes have been reported for many such integrated treatment programs and high-quality parallel treatments (10, 34), it has been difficult to demonstrate the superior efficacy of any particular form of treatment, and there remains a need for high-quality trials to identify the most efficacious approaches (35).

Psychosocial Treatments for Substance Use Disorders and Co-Occurring Disorders

Before discussing specific psychosocial treatments, there are three important general points to keep in mind. First, the evidence base for these treatments in populations with co-occurring disorders is much thinner than in standard addiction treatment populations. Second, very few studies of psychosocial treatments have limited their samples to patients with schizophrenia—much more often, the sample is a mixed group of patients with serious mental illnesses. Third, the modalities discussed in the following text are, in practice, often used in combination rather than as stand-alone treatments. As an example, one of the most successful recent psychosocial treatment trials in a population with serious mental illness used an intervention combining three of the modalities described in the following text: motivational interviewing, contingency management, and skill training (36).

Cognitive-behavioral therapy.

Cognitive-behavioral therapy for substance use disorders has multiple components. Patients work with therapists, either individually or in groups, to identify the proximate causes and consequences of substance use (functional analysis) and to work on changing behaviors and learning the skills necessary to avoid relapse (37). Treatment providers often confer psychoeducation about the impact of drug use and generally focus some time on identifying triggers and behaviors that lead to relapse. Social skills training is incorporated to practice behaviors that reinforce sobriety, such as drink or drug refusal skills, asking for help, going to meetings, or calling a sober friend or sponsor. Obviously, there is considerable overlap between these methods and the social skills training methods used for the treatment of social deficits in schizophrenia (38).

Motivational interviewing and motivational enhancement therapy.

Motivational interviewing techniques, which focus on enhancing motivation to change rather than teaching specific skills, are effective in the treatment of substance use disorders in the general population (39). Motivational interviewing approaches are widely recommended and used in populations with serious mental illness, including schizophrenia, although there are relatively few studies in such populations. Adjustment of the interviewing techniques in the population with schizophrenia appears to be helpful. Some of these adjustments include using more structured activities, prompts, examples, and lists of options, as well as frequent review and reminders, active reflection, and periodic summarization to promote cognitive organization and processing (40). Focus on concrete issues such as the negative consequences of substance use disorders appears to be more beneficial than asking about the motivating factors that perpetuate use. These techniques have not been rigorously examined in well-designed studies but appear to be more effective than traditional motivational interviewing strategies.

Contingency management.

Contingency management is an effective substance use disorder treatment approach using operant conditioning principles to increase the probability of target behaviors during treatment (41). Typically, participants in substance use disorder treatment receive rewards (cash, vouchers, or prizes) for producing drug-free urine samples, participating in treatment, or both. Contingency management is usually an adjunct to other interventions rather than a stand-alone treatment. The effects of contingency management on treatment-related behaviors are quite robust, but they often attenuate fairly rapidly when the contingencies are removed. This approach has not been widely used in populations with serious mental illness, but it has shown some promise (42). Contingency management can be applied to a variety of target behaviors, including psychiatric medication adherence.

Residential treatment.

Residential treatment with a co-occurring disorder focus is associated with improved outcomes across a number of studies (43). Although substance use disorders and schizophrenia are both typically long-term illnesses, it appears that an acute episode of intensive treatment can affect the course of both disorders. It may be that a period of stabilization in a drug-free and medication-adherent state brings about benefits that cannot always be realized in an outpatient setting.

Twelve-step groups and 12-step facilitation.

The 12-step approach has been less influential in the treatment of patients with serious mental illness than in general substance use disorder treatment, but it is an important component of treatment and social support for many individuals with co-occurring disorders (44). Patients with schizophrenia are less likely to become involved with 12-step programs than are other people with addictions. Many patients with co-occurring disorders have been found to avoid traditional self-help groups because of fears of stigmatization, paranoia, and negative perceptions among other group members of taking psychiatric medications. However, those who do attend appear to benefit from 12-step participation about as much as those without serious mental illness. Because of this, several specialized groups such as Double Trouble in Recovery (DTR) and Dual Recovery Anonymous (DRA) have been created. These groups focus on the 12-step model approach to recovery from both substance use disorders and psychiatric illness (45). As such, participants are encouraged to discuss the problems they face in relation to both problems and to take an active role in their recovery.

Effects of Psychiatric Medications on Substance Use

The possible effects of antipsychotic medications on substance-use-related reward and outcomes are discussed in the preceding text. In addition, because adherence to medications may be more difficult in patients with co-occurring disorders than in those with schizophrenia alone, consideration should be given to using long-acting injectable formulations of antipsychotic medications. Lithium and most of the anticonvulsants used as mood stabilizers (including carbamazepine and valproate) have been studied as potential treatments for alcohol and cocaine dependence and have not been shown to be effective, although carbamazepine, valproate, and gabapentin can be effective off-label treatments for alcohol withdrawal (46). Benzodiazepines are frequently used as adjunctive medications in schizophrenia and appear to be useful in decreasing anxiety, although they have no effect on psychosis and have been associated with possible increases in dementia risk (47) and mortality (48). The abuse potential of benzodiazepines is fairly low in the general population, but it is somewhat higher among patients with substance use disorders; therefore, caution is always advised when prescribing benzodiazepines to patients with substance use disorders. Benzodiazepines should rarely be used with patients who are actively misusing alcohol or opioids because of the potential for synergistic effects leading to respiratory depression and increased mortality risk. Although anti-Parkinsonian medications are not usually thought of as addictive, they can be misused, and, rarely, they can be a primary substance of abuse. Therefore, clinicians should be vigilant for signs of abuse among all patients prescribed these medications, particularly patients with substance use disorders. Most antidepressants have not been shown to have significant effects on substance use disorders. Although selective serotonin reuptake inhibitors (SSRIs) may have effects on alcohol use that differ within subgroups of alcoholics (49), it is unclear how these findings would extrapolate to populations with schizophrenia.

The Relationships Between Schizophrenia and Use of Specific Substances

In the following text, we review what is known about the relationships between substance use and schizophrenia for each of the main classes of substances. We also discuss the role of specific antiaddictive medications that have been studied or approved (or both) by the U.S. Food and Drug Administration (FDA) for certain substances. Unfortunately, there are no FDA-approved medications for the treatment of cocaine, stimulant, or cannabis use disorders. A review of FDA-approved antiaddictive medications is summarized in Table 1.

Cannabis

Cannabis is the most commonly used illicit substance globally, and its main psychoactive component, Δ⁹- tetrahydrocannabinol (THC), has been widely recognized for its ability to cause acute psychotic symptoms and cognitive impairment similar to schizophrenia. Cannabis is also the most commonly used illicit substance among people with schizophrenia, and there is substantial evidence that continued use after a schizophrenia diagnosis is associated with worsening psychotic symptoms, relapse, and decreased functioning over time (21). Furthermore, epidemiologic evidence has accumulated that implicates cannabis use as an important environmental risk factor for developing schizophrenia.

An analysis of results from 35 studies revealed a pooled odds ratio of 1.4 for risk of developing psychosis in individuals with any history of cannabis exposure and an odds ratio of 2.1 among individuals with histories of more frequent use (50). In addition, a relationship between the age of initiation of cannabis use and age of onset of psychosis has been well documented, with most studies finding that cannabis use precedes onset of psychosis and that the amount of cannabis exposure in adolescence may be the strongest determinant of this association (51). However, several issues have complicated the interpretation of these findings, including the role of other substances in comorbid substance use, male gender (which is associated with both higher rates of cannabis use and earlier age of schizophrenia onset), lower socioeconomic status, and the potentially poor reliability of self-reported histories of past cannabis use.

In addition, rather than a causal association, it is possible that cannabis use represents self-medication of early symptoms of a psychotic diathesis (e.g., anxiety) or that both the predilection for early use of cannabis and the risk for psychosis are related to a third factor (52). Consistent with this model, schizophrenia risk alleles have been linked both to risk for cannabis use and the quantity of use among the general population (53). Cannabis exposure during adolescence is most strongly associated with the onset of psychosis among those who are particularly vulnerable, such as those who have been exposed to child abuse and those with family histories of schizophrenia. Schizophrenia that develops after cannabis use may have a unique clinical phenotype, with a shorter prodromal period, more positive symptoms, and fewer negative and cognitive symptoms, and several genetic polymorphisms may modulate the relationship between cannabis use and psychosis (21).

Regardless of how important cannabis use is in causing the onset of schizophrenia, it is fairly clear that discontinuing cannabis use improves mood and anxiety symptoms (54), psychotic symptoms (55), and psychosocial functioning. Unfortunately, however, essentially no psychosocial or pharmacological interventions have been proven to help people with schizophrenia stop or reduce cannabis use (56).

Stimulants

Stimulant use is very likely to exacerbate psychotic symptoms of people with schizophrenia (57). Stimulant intoxication can cause psychotic symptoms even among individuals with no predisposition to psychosis. Although these symptoms most often persist for days or, at most, weeks after stimulant use has ceased, it is thought that these symptoms can, at times, persist indefinitely, causing a chronic persistent psychosis through a process of sensitization or kindling (58). These psychoses are described as typically involving paranoid delusions, with or without auditory hallucinations, but can also cause the full range of schizophrenia symptoms (59). However, it remains unclear whether the disorder represents true schizophrenia (precipitated by stimulants) or a different disorder (chronic stimulant-induced psychosis) with similar symptomatology. There are no FDA-approved medications for the treatment of stimulant use disorders, and no medication has consistently demonstrated efficacy in randomized trials, although a number of promising medications are in development (60). A recent study also showed promise for extended-release mixed amphetamine salts in reducing cocaine use among people with attention-deficit hyperactivity disorder (61), although it is not likely that amphetamines would be helpful for patients with comorbid schizophrenia, because of the risk of exacerbating positive symptoms.

Hallucinogens

Both the serotonergic hallucinogens (e.g., lysergic acid diethylamide [LSD], psilocybin, mescaline, dimethyltryptamine) and the NMDA receptor antagonist hallucinogens (e.g., phencyclidine [PCP], ketamine, dextromethorphan) have been used as experimental models of schizophrenia. Hallucinogens can both precipitate brief psychotic episodes and exacerbate chronic psychotic illness. Whereas all hallucinogens can cause transient psychotic symptoms among people without prior psychosis and exacerbate psychotic symptoms of patients with schizophrenia, NMDA antagonists tend to cause more severe symptoms that are more consistent with the full schizophrenia syndrome, likely because of the fact that glutamate is involved in the pathophysiology of schizophrenia (62). Acutely, these drugs produce not only altered perceptions of self and reality but also cognitive impairment and formal thought disorder. Psychotic episodes persisting for at least days are relatively common among people who use NMDA antagonist hallucinogens (62).

Opioids

Cases of psychosis have been reported because of both opioid intoxication and withdrawal. However, there is evidence from older studies that opioids have a moderate but significant antipsychotic effect among patients with schizophrenia (63), and there is no evidence that opioid use disorders can cause a persistent psychosis or exacerbate psychotic symptoms of patients with schizophrenia.

Methadone, buprenorphine, and naltrexone have been approved for the treatment of opioid use disorders. Methadone and buprenorphine are considered safe and effective for patients with schizophrenia, and they are commonly used in this population, although there is a paucity of data in this area. There is minimal evidence or clinical experience concerning the use of naltrexone to treat opioid use disorder of patients with schizophrenia.

Methadone is a long-acting opioid agonist used for both detoxification and maintenance treatment of people with an opioid use disorder. Methadone has been used for decades and has been found to be efficacious and safe (64). Side effects are moderate and may include constipation and sedation when used at high doses. Methadone is metabolized by the cytochrome P450 3A4 isoenzyme, and blood levels may increase when combined with medications that compete for this isoenzyme, such as some SSRIs. Carbamazepine, on the other hand, is an inducer of this enzyme and may decrease blood levels of methadone. Maintenance treatment consistently improves functioning and health and decreases mortality of patients who have failed attempts at abstinence (64). Patients must have documented evidence of physiological dependence before being started on methadone. If they are under 18 years old, they must have two previous failed attempts at detoxification and abstinence. Methadone for maintenance must be dispensed at programs that are licensed and accredited for this purpose.

Buprenorphine also has good evidence of efficacy for opioid use disorder (65). It is a high-affinity, partial agonist at mu opioid receptors, as well as a weak kappa receptor antagonist. Buprenorphine is available in several formulations. Sublingual tablets exist, which are either buprenorphine alone or buprenorphine with naloxone. Naloxone, which is an opioid receptor antagonist, is added to the sublingual form to decrease diversion and injection. Naloxone is poorly absorbed through the oral mucosa, so it is not active unless it is injected. Buprenorphine may be prescribed in a physician’s office, and treatment induction must be monitored. A special Drug Enforcement Administration number is required to prescribe buprenorphine.

Naltrexone is a mu opioid receptor antagonist. When administered in therapeutic doses, it blocks or greatly attenuates the effects of exogenous opioids. Oral naltrexone has limited efficacy because of poor adherence with daily dosing, but naltrexone is now available as a long-acting injection, which may enhance adherence. In addition to being an effective treatment for alcohol use disorder (discussed later), naltrexone has shown promise in the treatment of gambling disorder (66), and preliminary results also suggest that it may attenuate weight gain and fat deposition among patients treated with antipsychotic medications (67, 68).

Alcohol

Other than tobacco, alcohol is the most commonly misused substance among people with schizophrenia. Psychosis can occur as a consequence of heavy alcohol use, most commonly in patients with severe, chronic alcohol dependence. This psychosis can resemble schizophrenia but with an older average age of onset and often involving other alcohol-induced cognitive deficits. There are four medications that have been FDA approved for the treatment of alcohol dependence: disulfiram, oral naltrexone, long-acting injectable naltrexone, and acamprosate. The use of these medications has not been extensively studied in schizophrenia. However, there is preliminary evidence that naltrexone and disulfiram may be beneficial in this population (11).

Disulfiram is an irreversible inhibitor of aldehyde dehydrogenase, which is one of the enzymes responsible for metabolizing ethyl alcohol. When a person taking a therapeutic dose of disulfiram drinks alcohol, acetaldehyde accumulates and causes the characteristic alcohol–disulfiram reaction: nausea, vomiting, facial flushing, tachycardia, hypotension, and physical discomfort. Disulfiram has not been found to be consistently effective in placebo-controlled studies, primarily because of issues related to adherence and a large placebo effect.

Disulfiram should be used carefully and with close monitoring of people with schizophrenia, because it can cause psychosis resulting from inhibition of dopamine-beta-hydrolylase, thereby preventing the conversion of dopamine to norepinephrine and increasing dopamine levels in the mesolimbic system (69). In practice, this does not appear to be a major concern for patients who are taking antipsychotic medication. Disulfiram requires extensive dietary and hygiene product restrictions, because patients can have severe adverse reactions when alcohol is ingested or absorbed through the skin. In addition, disulfiram has multiple drug interactions, including, but not limited to, many antiretroviral medications, antibiotics (e.g., metronidazole, isoniazid), certain benzodiazepines, antiepileptic drugs (e.g., phenytoin, fosphenytoin), and warfarin. Therefore, patients must be provided with extensive information about the alcohol–disulfiram reaction, products containing alcohol, and potential drug interactions. Because patients must be reliable and able to follow instructions, disulfiram should not be given to people who are highly impulsive.

The proposed mechanism of naltrexone in the treatment of alcohol use disorder is attenuation of the rewarding effects of alcohol, thereby decreasing the quantity of alcohol consumed. Naltrexone has also decreased craving for alcohol in some studies. Naltrexone has been shown to have a consistently moderate effect on drinking outcomes across multiple studies (70). There may be genetic variations in response to naltrexone.

Acamprosate is an NMDA glutamate receptor antagonist. Its exact mechanism of action is unknown, but it is thought to balance excitatory (glutamate) and inhibitory (GABA) neurotransmission, thereby decreasing chronic alcohol withdrawal. It may be more effective for relapse prevention than for the initiation of abstinence (71).

Topiramate is not FDA approved for treatment of alcohol dependence. However, it has been shown to produce a moderate but robust response in controlled studies for decreasing the quantity of alcohol consumed (72), and it may have some benefit in treating cocaine dependence as well (73). It is an antiepileptic medication with multiple mechanisms of action.

Inhalants

Inhalants include a wide range of volatile solvents, as well as nitrous oxide and vasodilators such as amyl nitrate. Many of the volatile solvents (e.g., gasoline, toluene, and difluoroethane) are neurotoxic and are known to cause brain damage, particularly white matter changes, leading to cognitive impairment. This brain damage can also cause symptoms of psychosis (74). Because inhalants are most commonly used by adolescents, it can be difficult to tell whether the onset of psychosis in a teenager who uses inhalants represents a first episode of a primary psychotic disorder or a consequence of inhalant use. Neuroimaging abnormalities can suggest, but do not prove, that inhalant use is the cause of the psychosis.

Tobacco

More than half of patients with schizophrenia are addicted to nicotine-containing tobacco products. Acutely, nicotine appears to improve cognitive function, normalize sensory gating in schizophrenia, and possibly offset some of the side effects of antipsychotic medications. Therefore, it has been proposed that people with schizophrenia use nicotine-containing products in an effort to self-medicate their negative symptoms and possibly even ameliorate positive symptoms (75). It is possible that this type of self-medication may have an underlying physiological mechanism related to the dysregulation of nicotinic acetylcholine transmission in schizophrenia (76). However, it is unclear whether long-term use of nicotine has positive or negative effects on cognitive or other symptoms of schizophrenia. Any possible benefits are more than offset by the severe health consequences of smoking, which are among the principal reasons that life expectancy of people with schizophrenia is at least 20 years less than that of the general population (77). Chemicals in tobacco smoke also decrease blood levels of many antipsychotic medications, including haloperidol, chlorpromazine, olanzapine, and clozapine, by inducing hepatic enzymes such as CYP1A2 and UDP-glucuronosyltransferase and thereby increasing the rate of their metabolism and clearance.

Several medications have been FDA approved in tobacco use disorder, including nicotine replacement therapy (i.e., a combination of patches, gum, inhalers, and/or lozenges), bupropion, and varenicline. Bupropion, which was initially FDA approved for the treatment of depression, has subsequently been approved for use in tobacco use disorder. The mechanism of action is largely unknown, although it is a weak inhibitor of the neuronal re-uptake of norepinephrine and dopamine. Varenicline is a partial nicotinic acetylcholine receptor agonist, which works by decreasing cravings and urges to use tobacco. There has been some concern that it may cause an increase in neuropsychiatric symptoms in some patients, including increased agitation and aggressive behavior, as well as increased suicidal ideation and behaviors. Therefore, the use of this medication should be closely monitored in patients with schizophrenia for worsening psychiatric symptoms. However, recent research has shown varenicline to be relatively safe for patients with serious mental illnesses, with minimal significant increases in psychiatric adverse events (78). While some research has suggested that varenicline or intensive nicotine replacement therapy may have superior smoking cessation efficacy in the general population, more recent research questions efficacy differences between these approaches (79). There is little guidance as to which treatments to use, the order in which to use them, and in what combinations to aid smoking cessation specifically for patients with schizophrenia. However, a recent comprehensive review found strong evidence for both bupropion and varenicline for patients with serious mental illness in general (80).

Conclusions

In summary, substance use disorders are prevalent co-occurring problems among people with schizophrenia, and they correlate with significant adverse effects in this population, including increased morbidity and mortality. The etiology of this relationship is multifactorial, involving neurobiological, genetic, and environmental factors. Multiple substances of abuse are common, especially tobacco, alcohol, cannabis, and cocaine. Screening, diagnosis, and treatment of substance use disorders are important and can have profound effects on clinical outcomes. Treatments include psychopharmacological, psychotherapeutic, and multidisciplinary interventions. Several medications have been FDA approved for the treatment of substance use disorders, which also appear to be helpful in treating patients with comorbid schizophrenia, although few controlled trials have been conducted specifically in this population. Psychosocial and psychological interventions have been adapted for use with patients with schizophrenia as well. Treatment of both psychotic illness and substance use disorders in an integrated way improves functional and clinical outcomes.