You are on page 1of 6

Bipolar Disorder

and Alcoholism

Susan C. Sonne, PharmD, and Kathleen T. Brady, M.D., Ph.D.

Bipolar disorder and alcoholism commonly co-occur. Multiple explanations for the relationship
between these conditions have been proposed, but this relationship remains poorly understood.
Some evidence suggests a genetic link. This comorbidity also has implications for diagnosis and
treatment. Alcohol use may worsen the clinical course of bipolar disorder, making it harder to
treat. There has been little research on the appropriate treatment for comorbid patients. Some
studies have evaluated the effects of valproate, lithium, and naltrexone, as well as psychosocial
interventions, in treating alcoholic bipolar patients, but further research is needed. KEY WORDS:
comorbidity; manic-depressive psychosis; AODD (alcohol and other drug dependence); alcoholic
beverage; prevalence; genetic linkage; disease onset; disease course; diagnosis; drug therapy;
lithium; valproate; naltrexone; patient compliance; psychosocial treatment method; literature review

ipolar disorder and alcoholism between the onset of illness and diag­ least 4 days in a row and is not severe
co-occur at higher than expected nosis and treatment (Lish et al. 1994). enough to require hospitalization.
rates. That is, they co-occur more Bipolar disorder affects approximately Hypomania is interspersed with depres­
often than would be expected by chance 1 to 2 percent of the population and sive episodes that last at least 14 days.
and they co-occur more often than do often starts in early adulthood. People with bipolar II disorder often
alcoholism and unipolar depression. There are a number of disorders in enjoy being hypomanic (due to ele­
This article will explore the relationship the bipolar spectrum, including bipolar vated mood and inflated self-esteem)
between these disorders, focusing on I disorder, bipolar II disorder, and and are more likely to seek treatment
the prevalence of this comorbidity, poten­ cyclothymia. Bipolar I disorder is the during a depressive episode than a
tial theoretical explanations for the high most severe; it is characterized by manic manic episode. Cyclothymia is a disor­
rates of comorbidity, effects of comorbid episodes that last for at least a week der in the bipolar spectrum that is
alcoholism on the course and features and depressive episodes that last for at characterized by frequent low-level
of bipolar disorder, diagnostic issues, least 2 weeks. Patients who are fully mood fluctuations that range from
and treatment of comorbid patients. manic often require hospitalization to hypomania to low-level depression,
Bipolar disorder, often called manic decrease the risk of harming themselves with symptoms existing for at least 2
depression, is a mood disorder that is or others. People can also have symptoms years (American Psychiatric Association
characterized by extreme fluctuations of both depression and mania at the [APA] 1994).
in mood from euphoria to severe depres­ same time. This mixed mania, as it is
sion, interspersed with periods of normal called, appears to be accompanied by a
mood (i.e., euthymia). Bipolar disorder greater risk of suicide and is more diffi­ SUSAN C. SONNE, PHARMD, is a research
represents a significant public health cult to treat. Patients with 4 or more assistant professor of psychiatry and behav­
problem, which often goes undiagnosed mood episodes within the same 12 ioral sciences and clinical assistant profes­
and untreated for lengthy periods. In a months are considered to have rapid sor of pharmacy practice, and KATHLEEN
survey of 500 bipolar patients, 48 per- cycling bipolar disorder, which is a pre­ T. BRADY, M.D., PH.D., is a professor of
cent consulted 5 or more health care dictor of poor response to some medi­ psychiatry and behavioral sciences, both
professionals before finally receiving a cations. Bipolar II disorder is character­ at the Medical University of South
diagnosis of bipolar disorder, and 35 ized by episodes of hypomania, a less Carolina, Center for Drug and Alcohol
percent spent an average of 10 years severe form of mania, which lasts for at Programs, Charleston, South Carolina.

Vol. 26, No. 2, 2002 103

Alcohol dependence, also known as
alcoholism, is characterized by a craving
for alcohol, possible physical dependence Comorbid Mood Disorders* and Substance Abuse
on alcohol, an inability to control one’s
drinking on any given occasion, and an Any substance abuse Alcohol Alcohol
increasing tolerance to alcohol’s effects or dependence (%) dependence (%) abuse (%)
(APA 1994). Approximately 14 percent Any Mood Disorder 32.0 4.9 6.9
of people experience alcohol dependence Any Bipolar Disorder 56.1 27.6 16.1
at some time during their lives (Kessler Bipolar I 60.7 31.5 14.7
et al. 1997). It often starts in early adult- Bipolar II 48.1 20.8 18.4
hood. Criteria for a diagnosis of alcohol Unipolar Depression 27.2 11.6 5.0
abuse, on the other hand, do not include
the craving and lack of control over NOTES: *Mood disorders include depression and bipolar disorder.
Bipolar disorder, or manic depression, is characterized by extreme mood swings.
drinking that are characteristic of alco­ Bipolar I disorder is the most severe bipolar disorder.
holism. Rather, alcohol abuse is defined Bipolar II disorder is less severe.
as a pattern of drinking that results in Unipolar depression is depression without manic episodes.
SOURCE: Data reported in the table are based on findings of the Epidemiologic Catchment Area study (Regier
the failure to fulfill responsibilities at et al. 1990).
work, school, or home; drinking in
dangerous situations; and having recur-
ring alcohol-related legal problems and the table, alcohol dependence was twice drawal may trigger bipolar symptoms.
relationship problems that are caused as likely to co-occur in people with bipo­ Still other studies have suggested that
or worsened by drinking (APA 1994). lar spectrum disorders than in those people with bipolar disorder may use
The lifetime prevalence of alcohol abuse with unipolar depression (i.e., depression alcohol during manic episodes in an
is approximately 10 percent (Kessler et without mania). It is also noteworthy attempt at self-medication, either to
al. 1997). Alcohol abuse often occurs in that bipolar disorder was more likely prolong their pleasurable state or to
early adulthood and is usually a precur­ to occur with alcohol dependence than sedate the agitation of mania. Finally,
sor to alcohol dependence (APA 1994). with alcohol abuse (see table). As part other researchers have suggested that
of the ECA study, Helzer and Przybeck alcohol use and withdrawal may affect
(1988) found that mania (i.e., bipolar I the same brain chemicals (i.e., neuro­
Prevalence of Comorbidity disorder) and alcohol use disorders are transmitters) involved in bipolar illness,
far more likely to occur together (i.e., thereby allowing one disorder to change
Several studies have reported an associ­ 6.2 times more likely) than would be the clinical course of the other. In other
ation between alcoholism and mood expected by chance. Of all other psy­ words, alcohol use or withdrawal may
disorders. To date, there have been two chiatric diagnoses investigated in this “prompt” bipolar disorder symptoms
large epidemiological studies of psychi­ study, only antisocial personality disorder (Tohen et al. 1998). It remains unclear
atric disorders: the National Institute of was more likely to be related to alco­ which if any of these potential mecha­
Mental Health’s Epidemiologic Catchment holism than mania. The findings of nisms is responsible for the strong asso­
Area (ECA) study (Regier et al. 1990) the NCS with regard to the comorbid­ ciation between alcoholism and bipolar
and the National Comorbidity Survey ity of mood disorders and alcoholism disorder. It is very likely that this relation-
(NCS) (Kessler et al. 1996). The ECA were very similar. ship is not simply a reflection of cause
study (Regier et al. 1990) revealed that and effect but rather that it is complex
60.7 percent of people with bipolar I and bidirectional. Genetic factors may
disorder had a lifetime diagnosis of a Possible Explanations also play a role, as described below.
substance use disorder (i.e., an alcohol for Comorbidity
or other drug use disorder); 46.2 percent
of those with bipolar I disorder had an
Familial Risk of Bipolar
Although researchers have proposed
alcohol use disorder; and 40.7 percent explanations for the strong association
Disorder and Alcoholism
had a drug abuse or dependence diagno­ between alcoholism and bipolar disorder, The role of genetic factors in psychiatric
sis (the percentages of people with alco­ the exact relationship between these disorders has received much attention
hol use disorders and drug abuse disor­ disorders is not well understood. One recently. Some evidence is available to
ders do not add to 100 due to overlap). proposed explanation is that certain support the possibility of familial trans-
Forty-eight percent of people with bipo­ psychiatric disorders (such as bipolar mission of both bipolar disorder and
lar II disorder had a substance use dis­ disorder) may be risk factors for sub- alcoholism (Merikangas and Gelernter
order, 39.2 percent had an alcohol use stance use. Alternatively, symptoms of 1990; Berrettini et al. 1997). Common
disorder, and 21 percent had a drug abuse bipolar disorder may emerge during the genetic factors may play a role in the
or dependence diagnosis (these figures course of chronic alcohol intoxication or development of this comorbidity, but
reflect overlap, as above.) As shown in withdrawal. For example, alcohol with­ this relationship is complex (Tohen et

104 Alcohol Research & Health

Bipolar Disorder and Alcoholism

al. 1998). Preisig and colleagues (2001) essarily indicate that alcoholism worsens patients with bipolar disorder can have
conducted a family study of mood disor­ bipolar symptoms, it does point out the remission of their alcoholism.
ders and alcoholism by evaluating 226 relationship between them. A compari­
people with alcoholism with and with- son of patients with bipolar disorder
out a mood disorder as well as family and a coexisting substance use disorder
Order of Onset
members of those people. The researchers with others who had bipolar disorder An important factor in studying the
found that there was a greater familial alone found that those with comorbid influence of one comorbid disorder on
association between alcoholism and substance use disorders had an earlier another is the order of onset of the two
bipolar disorder (odds ratio of 14.5) age of onset for their mood disorder, disorders. A mood disorder that occurs
than between alcoholism and unipolar were more likely to be male, had more prior to the onset of another psychiatric
depression (odds ratio of 1.7). These comorbid psychiatric disorders in addi­ disorder is called a primary affective
findings have implications for prevention tion to bipolar disorder, and were sig­ disorder. Secondary affective disorders
and treatment. A positive family his- nificantly more likely to have mixed occur after the onset of other psychi­
tory of bipolar disorder or alcoholism is mania at the time of interview (Sonne atric disorders. Feinman and Dunner
an important risk factor for offspring. and Brady 1999b). (1996) conducted a retrospective chart
Although research suggests that alco­ review of three groups of patients:
hol and other drug abuse may worsen 1. Those with primary bipolar disor­
Issues Surrounding the course of bipolar disorder, some der with no history of substance abuse
the Treatment of data indicate that patients with bipolar (primary group), with 103 patients
Comorbid Bipolar disorder and alcoholism do better in 2. Those with primary bipolar dis­
Disorder and Alcoholism substance abuse treatment than alcoholic order complicated by substance abuse,
patients with other mood disorders. which began after the onset of bipolar
This section examines some of the O’Sullivan and colleagues (1988) found disorder (complicated group), with 35
issues to consider in treating comorbid that alcoholics with bipolar disorder patients
patients, and a subsequent section functioned better during a 2-year fol­ 3. Those with bipolar disorder that
reviews pharmacologic and psychother­ lowup period than did primary alcoholics came after the onset of substance abuse
apeutic treatment approaches. (i.e., those without comorbid mood (secondary group), with 50 patients.
disorders) or alcoholics with unipolar The researchers found that patients
depression. This suggests that bipolar in the complicated group had a signifi­
Alcoholism’s Effect on Comorbid patients may use alcohol primarily as a cantly earlier age of onset of bipolar
Bipolar Disorder means to medicate their affective symp­ disorder than the other groups. They
A growing number of studies have shown toms, and if their bipolar symptoms are also found that the complicated and
that substance abuse, including alco­ adequately treated, they are able to stop secondary groups had higher rates of
holism, may worsen the clinical course abusing alcohol. Hasin and colleagues suicide attempts than did the primary
of bipolar disorder. Sonne and colleagues (1989) found that patients with bipolar group. Preisig and colleagues (2001)
(1994) evaluated the course and features II disorder were likely to have an earlier also reported that the onset of bipolar
of bipolar disorder in patients with and remission from alcoholism compared disorder tended to precede that of
without a lifetime substance use disor­ with patients with schizoaffective disor­ alcoholism. They concluded that this
der. They found that compared to non- der or bipolar I disorder. Researchers finding is in accordance with results of
substance abusers, substance-abusing have also proposed that the presence clinical studies that suggest alcoholism
bipolar patients were more likely to of mania may precipitate or exacerbate is often a complication of bipolar disor­
have frequent hospitalizations for affec­ alcoholism (Hasin et al. 1985). der rather than a risk factor for it.
tive symptoms, earlier onset of bipolar In conclusion, it appears that alco­ In a 5-year followup study, Winokur
disorder, more rapid cycling, and more holism may adversely affect the course and colleagues (1995) evaluated a group
mixed mania (the latter two considered and prognosis of bipolar disorder, lead­ of bipolar patients with and without
to be the most severe, treatment-resistant ing to more frequent hospitalizations. alcoholism. In the alcoholic patients,
forms of bipolar disorder). Keller and In addition, patients with more treatment- bipolar illness and alcoholism were
colleagues (1986) compared patients resistant symptoms (i.e., rapid cycling, categorized as being either primary or
who had pure depression or pure mania mixed mania) are more likely to have secondary. The patients with primary
with patients who had mixed or rapid comorbid alcoholism than patients with alcoholism had significantly fewer
cycling bipolar disorder and found that less severe bipolar symptoms. If left episodes of mood disorder at followup,
a higher percentage of patients with untreated, alcohol dependence and which may suggest that these patients
mixed or rapid cycling bipolar disorder withdrawal are likely to worsen mood had a less severe form of bipolar illness.
had concurrent alcoholism (13 percent) symptoms, thereby forming a vicious Thus, there is growing evidence that
and that these patients had a slower cycle of alcohol use and mood instability. the presence of a concomitant alcohol
recovery from the bipolar disorder. However, some data indicate that with use disorder may adversely affect the
Although this association does not nec­ effective treatment of mood symptoms, course of bipolar disorder, and the order

Vol. 26, No. 2, 2002 105

of onset of the two disorders has prog­ mood problems leading to substance substance abusers, alcoholics appear to
nostic implications. Specifically, bipolar use, which leads to a worsening of be at greater risk for developing mixed
patients with secondary alcoholism mood symptoms, which in turn may mania and rapid cycling. Researchers
may be better able to stop drinking if worsen the substance abuse, leading to have found that patients with mixed
their bipolar illness is adequately treated; even worse mood symptoms. mania respond less well to lithium than
and, conversely, bipolar patients with As a general rule, it seems appropriate patients with the nonmixed form of
primary alcoholism (alcoholism occurs to diagnose bipolar disorder if the symp­ the disorder (Prien et al. 1988). This
first) may be better able to control their toms clearly occur before the onset of suggests that lithium may not be the
mood symptoms if they are able to the alcoholism or if they persist during best choice for a substance-abusing
stop drinking. periods of sustained abstinence. The bipolar patient. However, in a 6-week
adequate amount of abstinence for diag­ trial of lithium versus placebo in 25
nostic purposes has not been clearly adolescents with bipolar disorder and
Comorbidity and Diagnostic Issues defined. Family history and severity of secondary substance dependence,
Almost every alcoholic will report having symptoms should also factor into diag­ Geller and colleagues (1998) found a
mood swings. It is very important to nostic considerations. Given that bipo­ significant reduction in positive urine
distinguish these alcohol-induced lar disorder and substance abuse co­ tests for substances of abuse and signifi­
symptoms from actual bipolar disorder. occur so frequently, it also makes sense cant improvement in psychiatric symp­
However, diagnosing bipolar disorder to screen for substance abuse in people toms. This suggests that lithium may
in the face of alcohol abuse can be dif­ seeking treatment for bipolar disorder. be a good choice for adolescent sub-
ficult because alcohol use and with­ stance abusers. The presence of bipolar
drawal, particularly with chronic use, subtypes was not addressed in this
can mimic nearly any psychiatric disor­ Treatment of Comorbid study, so it is not clear if these adoles­
der. Alcohol intoxication can produce a Bipolar Disorder and cents had the subtypes of bipolar illness
syndrome indistinguishable from mania Alcoholism that are more difficult to treat.
or hypomania, characterized by eupho­
ria, increased energy, decreased appetite, In spite of the significant prevalence of
grandiosity, and sometimes paranoia. comorbid alcoholism and bipolar disor­
However, these alcohol-induced manic der, there is little published data on In 1998, the anticonvulsant Depakote®
symptoms generally occur only during specific pharmacologic and psychother­ (also called divalproex sodium, or val­
active alcohol intoxication, which makes apeutic treatments for bipolar disorder proate) was approved by the Food and
them fairly easy to differentiate from in the presence of alcoholism. The Drug Administration (FDA) for the
mania associated with bipolar I disorder. medications most frequently used for initial treatment of manic episodes asso­
Still, alcoholic patients going through treating bipolar disorder are the mood ciated with bipolar disorder. Numerous
alcohol withdrawal may appear to have stabilizers lithium and valproate. As studies have concluded that patients
depression. Depression is a key symptom stated previously, preliminary evidence with mixed or rapid cycling bipolar
of withdrawal from several substances suggests that alcoholic bipolar patients disorder are more likely to respond to
of abuse, and studies have demonstrated may have more rapid cycling and more anticonvulsant medications than to
that symptoms of withdrawal-related mixed mania than other bipolar patients. lithium (Bowden 1995). Because, as
depression may persist for 2 to 4 weeks There is also evidence to suggest that stated previously, bipolar patients with
(Brown and Schuckit 1988). Because these subtypes of bipolar disorder have concomitant alcoholism appear to have
of this phenomenon, it is likely that different responses to medications more mixed or rapid cycling bipolar
observation during lengthier periods of (Prien et al. 1988), which would help disorder than do bipolar patients who
abstinence (i.e., continued observation provide a rationale for the choice of are not alcoholic, alcoholic bipolar
following the withdrawal stage) is impor­ agents in the alcoholic bipolar patient. patients may also respond better to
tant for the diagnosis of depression as Available research on the use of lithium, anticonvulsant medications (e.g., val­
compared with mania. valproate, and naltrexone for comorbid proate) than to lithium therapy. In fact,
Bipolar II disorder and cyclothymia patients is reviewed below. in an open-label study (i.e., a study in
are even more difficult to reliably diag­ which all participants receive the exper­
nose because of the more subtle nature imental treatment), Brady and colleagues
of the psychiatric symptoms. Because
Lithium (1995) found valproate to be safe and
of the diagnostic difficulties, it may be Lithium has been the standard treatment effective in nine mixed-manic bipolar
that this diagnostic group is often over- for bipolar disorder for several decades. patients with concurrent substance
looked. Although these less severe forms Unfortunately, several studies have dependence (primarily alcohol depen­
of bipolar disorder may not be as dis­ reported that substance abuse is a pre­ dence) who previously had either not
ruptive as bipolar I disorder, it is still dictor of poor response of bipolar dis­ tolerated lithium or not responded to
important to recognize and treat them order to lithium. More specifically, as it. Similarly, Albanese and coworkers
in order to break the potential cycle of stated previously, compared to non- (2000) reported on 20 patients treated

106 Alcohol Research & Health

Bipolar Disorder and Alcoholism

with divalproex sodium and found that Compliance as a result of being assigned to it by the
even at fairly low doses divalproex researchers. Potential study participants
Medication compliance is an important
effectively treated the mood symptoms, were told that the investigators were
issue to consider when assessing the
and based on self-report, all patients interested in better understanding the
remained abstinent during the trial. effectiveness of medications. One study relationship between bipolar disorder
Both valproate and alcohol con­ of the lifetime medication compliance and substance abuse and therefore wished
sumption are known to cause tempo­ of lithium and valproate in 44 alcohol to see them monthly for 6 months.
rary elevations in liver function tests, and other drug-abusing bipolar patients The investigators found that psychother­
and in rare cases, fatal liver failure found that patients were significantly apy and Alcoholics Anonymous (AA)
(Sussman and McLain 1979; Lieber more likely to take valproate (50 percent attendance decreased over time and
and Leo 1992). Therefore, the safety compliant) compared with lithium (21 that substance use tended to increase
of valproate in the alcoholic population percent compliant). Side effects, includ­ from month 1 to month 6. The focus
has been questioned because of the ing lethargy, weight gain, and tremors, of the study participants’ psychotherapy
potential for hepatotoxicity in patients were listed as the main reason for non- also changed, with less emphasis on
who are already at risk for this compli­ compliance with lithium (Weiss et al. their specific disorders and more empha­
cation. However, recent preliminary 1998). However, it is also important sis on family, school, work, and other
evidence suggests that liver enzymes do to note that prescription bottles for personal issues. Although differences in
not dramatically increase in alcoholic lithium usually have a warning label on mood or substance use between months
patients who are receiving valproate, them not to drink alcohol while taking 1 and 6 were not statistically significant,
even if they are actively drinking (Sonne the medication. Thus, if an alcoholic there was a trend for increased substance
and Brady 1999a). Thus, valproate has the choice between taking lithium use. If the study participants had con­
appears to be a safe and effective medi­ or drinking alcohol, it is very likely the tinued with AA and if psychotherapy
cation for alcoholic bipolar patients. alcoholic will not be compliant with had continued to focus on bipolar dis­
lithium. Increased medication compli­ order and alcoholism, the patients’ sub-
ance with valproate may be an impor­ stance use might have improved. Given
Naltrexone tant factor in selecting a mood stabi­ the generally poor prognosis associated
Because evidence suggests that active lizer for alcoholic bipolar patients. with bipolar disorder and alcoholism,
drinking may worsen bipolar symptoms, it is particularly important to continu­
it makes sense that medications designed Psychosocial Interventions ously educate patients concerning the
to decrease alcohol consumption may relationship between these two disorders.
be useful in bipolar alcoholics. Naltrexone Psychosocial interventions have often The authors concluded that the devel­
(ReVia™) is an FDA-approved medication been considered the mainstays of treat­ opment of dually focused psychosocial
designed to decrease cravings for alco­ ment for alcoholism and other sub- treatments for this population may
hol. Maxwell and Shinderman (2000) stance use disorders. Several studies help improve substance use and affec­
reviewed the use of naltrexone in the have demonstrated success with cognitive tive outcomes.
treatment of alcoholism in 72 patients behavioral therapy in treating alcoholism
with major mental disorders, including (Project MATCH Research Group
bipolar disorder and major depression. 1998). Many of the principles of cog­ Conclusion
Eighty-two percent of patients stayed on nitive behavioral therapy are commonly
naltrexone for at least 8 weeks, 11 per- applied in the treatment of both mood Bipolar disorder and alcoholism com­
cent discontinued the medication disorders and alcoholism. Weiss and monly co-occur. In two epidemiologic
because of side effects, and the remaining colleagues (1999) have developed a survey studies, alcohol dependence was
7 percent discontinued for other rea­ relapse prevention group therapy using more likely to occur with bipolar disor­
sons. The authors concluded that nal­ cognitive behavioral therapy techniques der than with all other psychiatric dis­
trexone was useful in treating patients for treating patients with comorbid orders except antisocial personality
with comorbid psychiatric and alcohol bipolar disorder and substance use dis­ disorder. The nature of the relationship
problems. However, Sonne and Brady order. This therapy uses an integrated between alcoholism and bipolar disorder
(2000) reported on two cases of bipolar approach; participants discuss topics is complex and not well understood. It
women (both actively hypomanic) who that are relevant to both disorders, such appears that alcohol use may worsen
received naltrexone for alcohol cravings, as insomnia, emphasizing common the clinical course of bipolar disorder,
and both had significant side effects aspects of recovery and relapse. making it harder to treat. There is also
similar to those of opiate withdrawal. Interestingly, the same investigators evidence for a genetic link between the
Given that there is only preliminary (Weiss et al. 2000) evaluated the progress two conditions. Bipolar disorder com­
data on the use of naltrexone in bipolar of a group of substance abusers with plicated by alcoholism is associated
alcoholics to date, naltrexone should be comorbid bipolar spectrum disorders with an increased number of hospital­
used with caution in patients who have who were pursuing psychosocial treat­ izations, more mixed mania, earlier age
been actively hypomanic. ment independently, rather than of onset of bipolar disorder, and more

Vol. 26, No. 2, 2002 107

suicidal ideation. Given the prevalence HASIN, D.; ENDICOTT, J.; AND LEWIS, C. Alcohol Project MATCH Research Group. Matching alco­
and drug abuse in patients with affective syndromes. holism treatments to client heterogeneity: Treatment
and morbidity of these two disorders, Comprehensive Psychiatry 26:283–295, 1985. main effects and matching effects on drinking dur­
it is important to screen for substance ing treatment. Journal of Studies on Alcohol 59(6):
abuse in all bipolar patients and to treat HASIN, D.S.; ENDICOTT, J.; AND KELLER, M.B.
RDC alcoholism in patients with major affective 631–639, 1998.
aggressively. Unfortunately, there has syndromes: Two-year course. American Journal of REGIER, D.A.; FARMER, M.E.; RAE, D.S.; ET AL.
been little study of the appropriate Psychiatry 146:218–323, 1989. Comorbidity of mental disorders with alcohol and
treatment of this comorbidity. Several other drug abuse: Results from the Epidemiologic
HELZER, J.E., AND PRZYBECK, T.R. The co-occur­
studies suggest that mood stabilizers rence of alcoholism with other psychiatric disorders in Catchment Area (ECA) study. JAMA: Journal of the
(particularly valproate) may work bet­ the general population and its impact on treatment. American Medical Association 264:2511–2518, 1990.
ter than lithium in treating alcoholic Journal of Studies on Alcohol 49:219–224, 1988.
SONNE, S.C., AND BRADY, K.T. Safety of Depakote
bipolar patients, but head-to-head KELLER, M.B.; LAVORI, P.W.; CORYELL, W.; ET AL. in bipolar patients with comorbid alcohol abuse/
comparison of lithium and valproate Differential outcome of pure manic, mixed/cycling, dependence. American Journal of Psychiatry 156:
has not been carried out. Further study and pure depressive episodes in patients with bipolar 1122, 1999a.
illness. JAMA: Journal of the American Medical
of this important comorbidity is needed Association 255:3138–3142, 1986. SONNE, S.C., AND BRADY, K.T. Substance abuse
to better understand its course and and bipolar comorbidity. Psychiatric Clinics of North
treatment. ■ KESSLER, R.C.; NELSON, C.B.; MCGONAGLE, K.A.; America. 22:609–627, 1999b.
ET AL. The epidemiology of co-occurring addictive
and mental disorders: Implications for prevention SONNE, S.C., AND BRADY, K.T. Naltrexone for
and service utilization. American Journal of individuals with comorbid bipolar disorder and
References Orthopsychiatry 66:17–31, 1996. alcohol dependence. Journal of Clinical Psycho-
pharmacology 20:114–115, 2000.
E.J. Divalproex sodium in substance abusers with Lifetime co-occurrence of DSM–III–R alcohol SONNE, S.C.; BRADY, K.T.; AND MORTON, W.A.
mood disorder. Journal of Clinical Psychiatry 61: abuse and dependence with other psychiatric disor­ Substance abuse and bipolar affective disorder.
916–921, 2000. ders in the National Comorbidity Survey. Archives Journal of Nervous and Mental Disease 182:349–
of General Psychiatry 43:313–321, 1997. 352, 1994.
American Psychiatric Association (APA). Diagnostic
and Statistical Manual of Mental Disorders, Fourth LIEBER, C.S., AND LEO, M.A. Alcohol and the liver. SUSSMAN, N.B., AND MCLAIN, L.W. The direct hep­
Edition. Washington, DC: APA, 1994. In: Lieber, C.S., ed. Medical and Nutritional atotoxic effect of valproic acid. JAMA: Journal of the
Complications of Alcoholism. New York: Plenum
American Medical Association 242:1173–1174, 1979.
BERRETTINI,W.H.; FERRARO, T.N.; GOLDIN, L.R.; Medical Book Company, 1992. pp. 185–239.
ET AL.A linkage study of bipolar illness. Archives of TOHEN, M.; GREENFIELD, S.F.; WEISS, R.D.; ET AL.
General Psychiatry 54:27–35, 1997. LISH, J.D.; DIME-MEENAN, S.; WHYBROW, P.C.; ET
The effect of comorbid substance use disorders on
AL. The National Depressive and Manic-Depressive
Association (DMDA) survey of bipolar members. the course of bipolar disorder: A review. Harvard
BOWDEN, C.L. Predictors of response to divalproex
Journal of Affective Disorders 31:281–294, 1994. Review of Psychiatry 6:133–141, 1998.
and lithium. Journal of Clinical Psychiatry 56(Suppl.
3):25–30, 1995. WEISS, R.D.; GREENFIELD, S.F.; NAJAVITS, L.M.;
trexone in the treatment of alcohol use disorders in ET AL. Medication compliance among patients with
J.C. Valproate in the treatment of acute bipolar patients with concomitant major mental illness. bipolar disorder and substance use disorder. Journal
affective episodes complicated by substance abuse: Journal of Addictive Disease 19:61–69, 2000. of Clinical Psychiatry 59:172–174, 1998.
A pilot study. Journal of Clinical Psychiatry 56: MERIKANGAS, K.R., AND GELERNTER, C.S. Comorbidity WEISS, R.D.; NAJAVITS, L.M.; AND GREENFIELD,
118–121, 1995. for alcoholism and depression. Psychiatric Clinics of S.F. A relapse prevention group for patients with
BROWN, S.E., AND SCHUCKIT, M. Changes in North America 13(4):613–632, 1990. bipolar and substance use disorders. Journal of
depression among abstinent alcoholics. Journal of Substance Abuse Treatment 16:47–54, 1999.
Studies on Alcohol 49:412–417, 1988. follow-up study on alcoholics with and without co­ WEISS, R.D.; KOLODZIEJ, M.E.; NAJAVITS, L.M.;
FEINMAN, J.A., AND DUNNER, D.L. The effect of existing affective disorder. British Journal of Psychiatry ET AL. Utilization of psychosocial treatments by
152:813–819, 1988. patients diagnosed with bipolar disorder and sub-
alcohol and substance abuse on the course of bipo­
lar affective disorder. Journal of Affective Disorders PREISIG, M.; FENTON, B.T.; STEVENS, D.E.; AND stance abuse. American Journal on Addictions
37:43–49, 1996. MERIKANGAS, K.R. Familial relationship between 9:314–320, 2000.
mood disorders and alcoholism. Comprehensive WINOKUR, G.; CORYELL, W.; AKISKAL, H.S.; ET AL.
GELLER, B.; COOPER, T.B.; SUN, K.; ET AL. Double-
Psychiatry 42(2):87–95, 2001.
blind and placebo-controlled study of lithium for Alcoholism in manic-depressive (bipolar) illness:
adolescent bipolar disorders with secondary substance PRIEN, R.F.; HIMMELHOCH, J.M.; AND KUPFER, Familial illness, course of illness, and the primary-
dependency. Journal of the American Academy of Child D.J. Treatment of mixed mania. Journal of Affective secondary distinction. American Journal of Psychiatry
& Adolescent Psychiatry 37:171–178, 1998. Disorders 182:9–15, 1988. 152:365–372, 1995.

108 Alcohol Research & Health