These studies conclude that, although familial factors may contribute independently to the risk for psychiatric and alcohol disorders, such factors do not appear to constitute a shared etiologic substrate for both comorbid conditions. This conclusion does not rule out the possibility that as yet unidentified nonfamilial factors do contribute to both comorbid disorders as specified in the common factor model. For example, poor social skills have been separately identified as contributing to depression (Becker 1990) and substance abuse (Millman and Sbriglio 1986). More research is needed to validate this view. Self-Medication and Comorbidity Mechanisms The self-medication hypothesis is widely used to account for inflated comorbidity rates between alcohol and psychiatric disorders. Most consistent with the secondary alcoholism model, this view suggests that people with psychiatric symptoms are motivated to drink alcohol to relieve the symptoms. Despite the commonsense appeal of this view, the few studies that have tested this hypothesis offer equivocal results. Retrospective studies typically find that comorbid patients believe alcohol can provide relief from symptoms of their conditions. This finding is robust across several diagnostic categories, including the anxiety disorders (Kushner et al. 1990), depression (Smith and Pihl 1983), and schizophrenia (Dixon et al. 1990). Experimental evidence suggests that low doses of alcohol initially may decrease anxious tension (Cappell and Greeley 1987; Sher 1987) and dysphoria (Smith and Pihl 1983), but prolonged and highdose alcohol consumption can have the opposite effect (see Kushner et al. 1990; Peyser 1982; Schuckit 1983). These findings lend themselves to various interpretations. Reports of self-medication may represent a post hoc rationalization for drinking that is unrelated to alcohol's effect or the patient's true drinking motivation. Some comorbid people could self-medicate psychiatric symptoms with alcohol, believing it is therapeutic. Studies showing that alcohol-related expectations are strongly associated with drinking behavior offer credibility to this view (Christiansen et al. 1989) (see Chapter 6, Psychological, Social, and Developmental Factors). More consistent with the bidirectional causal model shown in figure 4, alcohol may both de crease and increase psychiatric symptoms. Based on experimental and clinical studies, Kushner et al. (1990) suggest that alcohol may give fast, short-term relief from anxiety symptoms but that prolonged drinking then aggravates the symptoms. This view describes a vicious-cycle, positive-feedback loop in which increased symptoms lead to increased drinking, which leads to increased symptoms, ad infinitum. Although it fits with past findings, this view awaits verification. The self-medication hypothesis... suggests that people with psychiatric symptoms are motivated to drink alcohol to relieve the symptoms. Regarding tests of the self-medication view, several conceptual issues need resolution. It is unclear what psychiatric symptom manifestations (if any) are amenable to self-medication with alcohol. This information is important because a range of qualitatively different symptoms are possible among and within psychiatric disorders. Dixon et al. (1990) observe that schizophrenia patients could use alcohol to decrease positive symptoms such as hallucinations, negative symptoms such as lethargy and social withdrawal, or neuroleptic-induced extrapyramidal side effects such as listlessness and depression. Disorders such as mania, schizophrenia, depression, bulimia, and sexual dysfunction may have differing symptoms that are targets of selfmedication with alcohol. Consistent reports from patients that alcohol is used primarily to relax and relieve dysphoria suggest that the goal of self-medication spans a range of psychiatric disorders. Current understanding of comorbidity mechanisms offers several possible causal connections between alcohol and psychiatric disorders. Multiple pathways to comorbidity may operate across people and diagnoses. Once initiated, comorbid conditions are potentially interactive and mutually maintaining. More rigorous prospective assessment of the developmental-temporal sequence of comorbid manifestations, including drinking motivation, may help future researchers specify factors that underlie the onset and maintenance of comorbid disorders. Implications of An important approach to evaluating the treatment needs of patients with comorbid conditions is to compare the course and outcome of patients with both disorders to those of patients with one disorder. This section reviews the extent to which psychiatric illness is related to the course of alcoholism, and the extent to which alcoholism is associated with the course of psychiatric illness. Psychopathology and the Course of Alcoholism Researchers have examined extensively the relationship between patient psychopathology and the course of alcohol abuse. Major outcome variables include relapse of addictive behaviors, number of days of drinking, hospitalizations, and social functioning. These studies typically are conducted in treatment settings for people with alcoholism and, because of referral patterns, are composed mainly of patients with less socially debilitating comorbid psychiatric illnesses. Some researchers have used continuous measures of psychopathological symptomatology, while others have compared the course of alcoholism among different categorical psychiatric disorders. Severity of concomitant psychopathology (or the presence of comorbid psychiatric disorders) may be associated with a poorer outcome of alcohol and other drug use (Herz et al. 1990; Kosten and Rounsaville 1988; Rounsaville et al. 1982, 1987). An important approach to evaluating the treatment needs of patients with comorbid conditions is to compare the course and outcome of patients with both disorders to those of patients with one disorder. To illustrate the relationship between psychopathology and outcome of patients being treated for alcoholism (McLellan 1986; McLellan et al. 1981, 1983; Woody et al. 1990), researchers. classified patients being treated for alcoholism into three groups based on the severity of their psychopathology (low, medium, and high severity). As expected, higher levels of psychopathol ogy were associated with a poorer response to traditional, abstinence-oriented treatment programs for alcohol abuse. Similar results were reported by Rounsaville et al. (1987) and Hesselbrock et al. (1985), who found that comorbid psychiatric diagnoses predicted a poorer outcome for alcoholics in treatment. While there is a general relationship between comorbid psychopathology and the course of alcoholism, several studies have evaluated this issue separately for different psychiatric conditions. ASPD is most consistently associated with earlier onset of drinking, greater impairment in social functioning, and a worse response to traditional alcoholism treatment programs (Hesselbrock 1991; Hesselbrock et al. 1985; Rounsaville et al. 1987). The poor response to treatment for alcoholics with ASPD, also reflected in the response of nonalcohol drug abusers with ASPD, is consistent with the treatment-refractory nature of this disorder to all psychotherapeutic interventions (Lewis 1989). Nace et al. (1986) compared alcoholics with and without BPD 12 months after treatment for alcoholism. Both groups improved, and after controlling for pretreatment alcohol use, no significant posttreatment group differences were noted. The authors also reported that BPD alcoholics used other drugs more often, even after controlling for pretreatment levels of other drug abuse. Kosten et al. (1989) found that opiate addicts with BPD had more severe depression, alcoholism, and other symptoms 21⁄2 years after treatment. These findings suggest that, in addition to ASPD, BPD may mark those alcoholics (at least those who also are abusing other drugs) for whom a poorer clinical course can be expected. The functional relation between depression and alcohol disorders has been extensively studied (see Jaffe and Ciraulo 1986; Merikangas and Gelernter 1990; Mirin and Weiss 1986; Schuckit 1986; Schuckit and Monteiro 1988). Although findings across a range of studies are mixed, most results suggest that depression is associated with a more severe course of alcoholism. Depression has been linked to relapses in drinking behavior (Hatsukami and Pickens 1982), increased alcohol intake (Hasin et al. 1989; Roy et al. 1991), severity of alcohol-related problems such as blackouts and arrests (Yates et al. 1988), and a pattern of earlier treatment for alcoholism, more hospitalizations, and longer lengths of stay for inpatient treatment (Hasin et al. 1988; O'Sullivan et al. 1988). Depression in alcoholism also increases the risk of suicide, which already is elevated in either disorder alone (Black et al. 1986; Hesselbrock et al. 1988; Petronis et al. 1990; Roy et al. 1991). Although most evidence supports the prognostic significance of depression for the course of alcoholism, some studies do not confirm this connection. Several studies reported no relationship between depression and relapse rates in alcoholism (Bowen et al. 1984; Schuckit 1983). Others show only small clinical differences between alcoholics with or without depression (Jaffe and Ciraulo 1986; Merikangas and Gelernter 1990). These negative findings seem to cast doubt on a simple cause-and-effect relationship between the presence of depression and alcohol disorder outcome. Gender appears to be an important factor that mediates the prognostic significance of depression in alcoholism. Consistent with gender differences in the prevalence of major depression in the general population, females with alcoholism have higher rates of depression than males (Hatsukami and Pickens 1982; Roy et al. 1991), but females with these comorbid conditions have a better prognosis than males. Hesselbrock (1991) (see Rounsaville et al. 1987) found that depression predicted a favorable response to alcoholism treatment for women, but a poorer outcome for men. Robins et al. (1989) reported that depression relates to increased relapse rates for male, but not female, alcoholics. Alcoholism can also lead to transient depression, anxiety, and other psychiatric symptoms through withdrawal, high-dose consumption, adjustment reaction to the treatment setting, or alcohol-related problems. Psychiatric symptoms in this context often are transitory and decrease within the first several weeks of treatment (Brown and Schuckit 1988; Brown et al. 1991; Powell et al. 1987; Schuckit et al. 1990). These results are consistent with research indicating that abstinent alcoholics report low levels of depression (Schuckit 1985) and that the longer alcoholics are abstinent, the lower their depression (Dorus et al. 1987). These findings do not mean that alcohol abuse leads only to transient depressive states. Recent evidence suggests that a history of chronic heavy alcohol abuse in younger persons is predictive of depression later in life that cannot be explained by drinking habits (Saunders et al. 1991). Some researchers have emphasized the prognostic importance of subtyping comorbid patients. based on several potentially relevant parameters. Separating patients on the basis of the comorbid disorder that began first, Schuckit (1985; see also Schuckit 1986) longitudinally assessed alcoholrelated course differences for patients in alcoholism treatment for whom 1. alcoholism began first (primary alcoholics); 2. nonalcohol drug abuse began first (secondary alcoholics); 3. ASPD began first (secondary alcoholics); and 4. affective disorder began first (secondary alcoholics). Depression has been linked to relapses in drinking behavior, increased alcohol intake, severity of alcohol-related problems such as blackouts and arrests, and a pattern of earlier treatment for alcoholism, more hospitalizations, and longer lengths of stay for inpatient treatment. He reported that primary alcoholics (group 1) showed a less severe course compared with groups 2 and 3, and that group 4 (primary affective disorder) showed more suicide attempts, psychiatric treatment contacts, and history of familial affective disorders. Other attempts to subtype comorbid alcoholics have produced mixed results. Hesselbrock et al. (1985) found that comorbidity in general, and ASPD in particular, identified alcoholics with a more severe disorder course. However, these investigators also found that the order-of-onset distinction is insignificant when applied to major depression; neither subgroup differed significantly from nondepressed alcoholics. Liskow et al. (1991) reported that ASPD alcoholics who met diagnostic criteria for depression or other drug dependence showed more pathological global alcohol-related severity measures than those with ASPD alone. They also reported that ASPD subgroups differed little on specific alcoholism symptoms and consequences. Effects of Alcohol Abuse on Most research on the course of psychiatric disorders in patients with comorbid alcoholism has been conducted in psychiatric inpatient settings or outpatient veterans or community mental health centers that serve the chronically mentally ill. Patients with chronic psychiatric illnesses— primarily schizophrenia or bipolar disorderwho also abuse alcohol tend to have a worse course in their psychiatric disorder. Much research in this area has been conducted with mixed groups of psychiatric patients who abuse alcohol and other drugs (Chen et al. 1992). A convergence of findings across studies of patients with chronic psychiatric disorders indicates that alcohol and other drug abuse is associated with a higher rate of relapse and rehospitalization, more admissions to emergency rooms, noncompliance with psychosocial and pharmacological interventions, and a history of arrests (Lyons and McGovern 1989; Safer 1987; Solomon 1986). Patients with chronic psychiatric illnesses— primarily schizophrenia or bipolar disorder-who also abuse alcohol tend to have a worse course in their psychiatric disorder. Although the relationship between alcohol abuse and the course of psychiatric illness has been studied extensively for schizophrenia (see special issue of Schizophrenia Bulletin, 16 [1] 1990), the research focus varies. Some studies have examined the impact of alcohol abuse on the onset of schizophrenia; others have compared chronicity of illness (number of relapses or rehospitalizations) and symptom severity among patients with and without alcohol abuse or dependence. Alcohol has been hypothesized to hasten the onset of schizophrenia and to mask its presence (Freed 1975), but studies comparing the age of first symptoms or the age of first hospitalization of alcoholic and nonalcoholic schizophrenics generally have failed to find differences, thus suggesting that alcohol abuse does not precipitate an earlier onset of illness (Hays and Aidroos 1986; Barbee et al. 1989; Gorelick et al. 1990). Studies examining the clinical course of schizophrenia in alcohol abusers compared to nonabusers have also produced contradictory results, although such studies tend to show that alcohol abusers have a more debilitating course of psychiatric illness. Findings from several studies suggest that a history of alcohol abuse in individuals diagnosed with schizophrenia is asso ciated with a higher relapse rate and more rehospitalizations (Cleghorn et al. 1991; Soni and Brownlee 1991). Some studies also report that a history of alcohol abuse in schizophrenia is related to more severe symptomatology and poorer self-care skills (Soni and Brownlee 1991). A large prospective study (Drake et al. 1989, 1990) found that alcohol abuse among schizophrenic outpatients living in a rural area was associated with a range of problems and a poor clinical outcome as measured by rehospitalizations, depression, poor treatment compliance, and housing instability and homelessness. Some studies report a seemingly contradictory finding that schizophrenics using alcohol or other drugs demonstrate better global functioning than nonabusers (Dixon et al. 1991; Gorelick et al. 1990; Mueser et al. in press). A key to understanding these results may be the social setting in which the abuse occurs because alcohol and other drug abuse among schizophrenics usually occurs in a social context (Dixon et al. 1990). Because poor social functioning is a core characteristic of schizophrenia, these studies suggest that schizophrenics prone to alcohol or drug abuse may be less socially and symptomatically impaired than nonabusers to begin with, but may suffer a course of illness exacerbated by their abuse. More than 20 years ago, Cohen and Klein (1970) observed that the most psychotic and impaired psychiatric patients lacked the social skills to sustain heavy drug use. Although alcohol may worsen psychiatric clinical course, those with a higher level of baseline functioning may be most able to participate in social encounters that involve alcohol use. Regarding bipolar disorder, studies have noted a tendency for patients to increase alcohol intake during manic phases of their illness (Liskow et al. 1982), but the effect on symptoms is not established. Evidence suggests that alcohol can mask some symptoms of mania (Goodwin and Jamison 1990). Bipolar patients who abuse alcohol may have longer manic periods before hospitalization than their nonabusing counterparts (Morrison 1974). Evidence also suggests that alcoholism is associated with a poor course of the psychiatric illness and a lack of response to conventional pharmacological treatments in bipolar patients with mixed affective states, such as simultaneous manic and depressive symptoms (Himmelhoch 1979). It is unknown whether alcoholism makes these patients less responsive to treatment, or whether more severely ill patients are independently more prone to alcohol abuse. Treatment Issues The need to improve treatment technologies for patients with psychiatric disorders and alcoholism is apparent from the preceding review of the course and outcome of comorbid disorders. The presence of concomitant psychiatric diagnoses and the severity of psychopathology in alcoholics are important predictors of a poor response to standard treatments for addictive disorders. Likewise, alcohol abuse appears to mark those for whom treatment of psychiatric disorders, such as schizophrenia and affective disorder, may be less successful. Treatment implications of these and related observations are the focus of this section. Issues A commonsense strategy for treating comorbidity seems to be to provide validated interventions for specific psychiatric disorders in addition to alcohol abuse treatment, but there are questions about this approach: 1. Which disorder should be treated first, or should they be treated concomitantly? 2. Do interventions validated in non-comorbid patients work in a similar way for comorbid patients? 3. Do comorbid patients respond better to hybrid treatments aimed at comorbidity as a specific disorder or to traditional treatments aimed separately at each disorder? 4. Do traditional treatment delivery systems suit the needs of comorbid patients? Treatment Priority Those who treat psychiatric disorders in alcohol-disordered patients must proceed with caution because of the many possible etiologies of these conditions. Evidence indicates that depression and anxiety are common sequelae to alcohol abuse and withdrawal and may parallel the course of alcoholism. Because of the transient nature of anxiety and depressive symptoms in alcoholism, it is recommended that treatment for these disorders be initiated 2 to 4 weeks after withdrawal from alcohol, when the disorder's stable presence can be established in the absence of alcohol effects (Jaffe and Ciraulo 1986; Schuckit and Monteiro 1988; Weiss and Mirin 1989). Although these considerations suggest that alcohol disorders should typically be treated first, this sequence of treatment may not always be possible, especially when failure to address the psychiatric disorder could endanger the patient or compromise the success of alcohol treatment. Postponing treatment of acutely symptomatic patients with schizophrenia or bipolar disorder may compromise the efficacy of alcoholism treatment because psychotic patients often have trouble acquiring new information and may be behaviorally unsuitable for an alcohol treatment setting. Stabilized outpatients already receiving treatment for psychiatric illnesses could receive treatment for alcoholism. Although these recommendations take into account good clinical judgment, few empirical data address the relative efficacy of treatment priority strategies with comorbid patients. The presence of concomitant psychiatric diagnoses and the severity of psychopathology in alcoholics are important predictors of a poor response to standard treatments for addictive disorders. Psychiatric Treatment Approaches for Comorbid Patients Pharmacotherapies Several empirical and clinical observations in the literature suggest that the pharmacologic treatment of psychiatric disorders may differ for patients with co-occurring alcohol disorders. As outlined by Zweben and Smith (1989), important issues include the following: 1. Some medications may present a high abuse potential and may increase the risk for relapse to the drug of choice. 2. Medications that are dangerous when mixed with alcohol may be inappropriate for comorbid patients. 3. Past alcohol abuse may alter the pharmacokinetics of some drugs. 4. Comorbid patients may not comply with medication regimens, thus making such treatments less effective or more dangerous. |