Unhooked
Science Readings Unhooked
Science Readings
The unhooked.com science section contains selected educational readings from the scientific and popular literature about alcoholism, addiction, and recovery. The views expressed in the articles are those of their authors and not necessarily those of the science pagemaster or the webmaster or of the person who suggested the article to the list. This material is made available solely for the nonprofit educational use of unhooked.com readers as an aid in their personal recovery, and no other use is authorized or intended. Click here for the current Science Section reading list.
The Link Between Parental Alcoholism and Childhood Mood Disorders:
A Familial/Genetic Perspective
Author: Richard D. Todd, MD, PhD, Washington University School of Medicine
Abstract: A variety of recent family studies suggest that adult relatives of depressed children have increased rates of both alcoholism and mood disorders (major depression and bipolar disorder), and that the child and adolescent offspring of adult alcoholics have increased rates of mood disorders. However, there has been marked confusion as to whether the mode of transmission of alcoholism and mood disorders in these families can be characterized as independent transmission or cotransmission. It is proposed that depressed children identify families with a specific form of bipolar disorder that shares some genetic components with alcoholism. [Medscape Mental Health 2(4), 1997. © 1997 Medscape, Inc.]
Introduction
A large and complex literature has emerged on the interaction between mood disorders and alcoholism. These studies have included assessment of individuals, adult relatives, and child and adolescent offspring. Although there is general agreement in the psychiatric community that depressed or bipolar individuals are at increased risk for alcoholism and alcoholic individuals are at increased risk for depression, the nature and magnitude of this relationship between alcoholism and depression are unclear. The purpose of this article is to review several key findings regarding risk for depression in children of depressed or alcoholic parents and the risk for alcoholism in parents of depressed children, and to suggest a unifying hypothesis for the observed pattern of illness in these families. This review highlights a number of relevant studies but is in no way exhaustive in its scope or content.
Children of Depressed or Alcoholic Parents
Since the recognition that children suffer depressive episodes phenomenologically similar to those of adulthood,[1] a variety of studies have observed factors predisposing to the development of childhood-onset mood disorders. These have frequently been high-risk family studies of children of parents with depression[2-6] or manic-depressive illness.[3,7-11] They demonstrated that the child and adolescent offspring of parents with depression or bipolar disorder have an increased risk for developing mood disorders. Most of these studies were restricted to nuclear families and therefore could not separate the effects of genes and environment on the psychopathology of these subjects.[12]
Children of patients with bipolar disorder. Two extended-family studies reported on the risk for developing bipolar disorder among child and adolescent offspring of extended pedigrees.[13,14] Such studies measured the risk for illness over several degrees of genetic relatedness. In both of these studies, the subjects were personally interviewed. There was a significant positive linear relationship between the offspring's risk for developing an early-onset mood disorder and the degree of genetic relatedness to an affected adult in the family (Fig. 1). There was no pattern of increased risk for offspring developing other psychiatric disorders, such as disruptive behavior disorders (attention deficit/hyperactivity disorder, oppositional defiant disorder, or conduct disorder), substance abuse problems, or anxiety disorders. In this series of consecutive families, identified through adult probands with bipolar disorder, the distribution of illness in offspring was compatible with the presence of important additive genetic factors contributing to early-onset bipolar disorder.[14]
Figure 1 (not reproduced here). Risk of childhood mood disorder is proportional to genetic distance to mood disordered biological relative. Percentage of child and adolescent offspring who were found to have mood disorder is plotted versus percentage of genomic DNA shared with the nearest mood disordered relative. First-degree relatives share 50%, second-degree relatives share 25%, and third-degree relatives share 12.5% of genome identical by descent. Error bars indicate estimated standard errors based on number of individuals studied. Data from Todd et al.[14]
Children of alcoholic parents. A number of studies assessed the psychopathology of the children of alcoholic parents, and the findings were similar to those seen in families having a parent with a mood disorder. These nuclear-family studies assessed categorical diagnoses[15,16] or symptoms of psychiatric disorders[17,18] in children. All of the studies reported elevated rates of "internalizing disorders" (ie, symptoms or diagnoses of depression or anxiety disorders). In these studies, moreover, the general risk for psychopathology in offspring appeared to be correlated with the number of alcoholic parents[15,16] (Wendy Reich, PhD, personal communication). Reich and coworkers[15] found elevated rates of overanxious disorders and disruptive behavior disorders, but not of depression, in children of alcoholics. Hill and Muka[16] studied the children of very high-risk families (ie, children with many alcoholic relatives) versus low-risk families (ie, children with no alcoholic relatives) and reported elevated rates of both internalizing and disruptive-behavior disorders in the former group. However, due to the small group sizes studied, the only significant finding was the presence of any internalizing disorder or any psychiatric disorder in offspring.
In a more recent study involving a large number of alcoholic families, Reich and colleagues (W. Reich, PhD, personal communication) found a significant correlation between the frequency of internalizing and disruptive-behavior disorders in offspring and the number of alcoholic parents. In particular, the offspring of 2 alcoholic parents had a 2-fold increased risk of child- and adolescent-onset depression when compared with the offspring of nonalcoholic parents. Similarly, there were 2-fold increases in the frequencies of attention deficit/hyperactivity disorder and conduct disorder in the offspring of 2 alcoholic parents compared to the offspring of nonalcoholic parents.
In most of these studies, however, there has been inadequate control for diagnoses other than alcoholism in parents. Hence, many parents may also suffer from affective or other disorders. Although the offspring of these families are undoubtedly at greater risk for the development of substance-abuse problems, these generally have not been found. Presumably, an increased prevalence of substance-abuse problems in offspring will develop as youth enter the age of risk for the onset of substance abuse.
Relatives of Depressed and Bipolar Children
Determination of family risk factors for early-onset mood disorders has also been investigated by identifying families through their depressed children. Three family-history studies reported on the distribution of psychiatric illness in the adult relatives of childhood depression probands.[19-21] In the study by Puig-Antich and colleagues,[19] the lifetime age-corrected risk for major depressive disorder and for alcoholism increased 2-fold in first-degree relatives of depressed children compared with that in the adult relatives of control-group children. Alcoholism was more common among adult males, and depression was more common among adult females.
Similarly, Todd and coworkers[20,22,23] found increased prevalences of mood disorders and alcoholism among the adult relatives of 76 prepubertal children with major depressive disorder. However, the rate of bipolar disorder was markedly elevated in the parents of depressed children (15.4%) compared with that in the parents of normal children (0%).[20] Alcoholism not associated with a mood disorder was more common among male relatives, as was bipolar disorder.[22] Depression was more common among female relatives.
All of the children who served as probands for these families have been followed by Barbara Geller longitudinally.[24] On a 2- to 5-year follow-up, 8 depressed children progressed from major depressive disorder to bipolar I disorder (10.5%), and 14 depressed children progressed to bipolar II disorder (18.4%). No children in the control group became ill, and no children in either group developed substance-abuse problems. The prevalences of bipolar disorder and major depressive disorder were similar among the adult relatives of depressed children, regardless of whether the children progressed or did not progress from major depressive disorder to bipolar disorder. In contrast, the prevalence of alcoholism (20.3%) was higher in adult relatives of bipolar children than in relatives of depressed (12.3%) or unaffected (6.4%) children.[22]
Using a hospital-based, case-control chart review of depressed children versus children with other inpatient psychiatric disorders, Weller and colleagues[21] found an increased prevalence of mood disorders among mothers and of antisocial personality disorders among fathers. The children with depression and a comorbid behavior disorder had a higher prevalence of family histories of alcoholism or antisocial personality disorder.
In summary, identifying a family through a child with major depressive disorder predicts increased prevalence of several disorders among family members, including depression, bipolar disorder, and alcoholism. If the depressed child also has a comorbid behavior disorder, there is an increased prevalence of antisocial personality disorder in adult relatives. Also, there is a higher prevalence of alcoholism among the adult relatives of depressed children who progress to bipolar disorder.
Relatives of Alcoholic or Mood-Disorder Adults
It has long been recognized that the prevalence of alcoholism is elevated among depressed adults, and the prevalence of depression is elevated among alcoholic adults.[25,26] These relationships appear to hold true for adolescents as well. Fergusson and coworkers[27] reported on the association between alcohol abuse and depression in a population-based cohort of 965 children who had been studied from birth to 15 years of age. In this sample, 4.9% of the youths met the diagnostic criteria for alcohol abuse at 15 years of age. These adolescents had a nearly 5-fold increased likelihood of having experienced a major depressive episode during the preceding 12 months compared to youths who did not drink. There was also a significant elevation in the risk for depression among adolescents who reported problems associated with alcohol use but did not meet alcohol abuse criteria.
Similarly, there was an increased prevalence of alcoholism among patients with bipolar disorder. For example, in the family study by Todd and colleagues,[22] the frequency of comorbid alcoholism among bipolar relatives was nearly 40%. Similarly, in a series of consecutively admitted inpatients and outpatients, Maier and coworkers[28] found that 35% of bipolar individuals had comorbid alcoholism. Despite the recognition that alcoholism and mood disorders occur together in adults, there has been significant disagreement over whether this comorbidity represents the development of a secondary environmental illness (ie, alcoholism leads to depression or, conversely, depressed individuals drink to self-medicate), or whether both illnesses are expressions of the same underlying process.
This issue has been examined by a number of investigators using the family-study method (reviewed by Maier and Merikangas[29]). These studies demonstrated that depression, bipolar disorder, and alcoholism aggregate in families. Again, whether this aggregation of disorders within families represents the sharing of familial etiologies between alcoholism and mood disorders or represents their joint occurrence through other mechanisms has been controversial.
Recently, Maier and Merikangas[29] investigated the association between alcoholism, bipolar disorder, and depression using family-study data to determine the degree of cosegregation in families. This is the only family study that determined the risk to relatives of developing affective disorder and alcoholism using families identified through all possible proband types (bipolar, bipolar with alcoholism, depressed, depressed with alcoholism, alcoholic without affective disorder). In accord with the findings of previous family studies, Maier and Merikangas found increased comorbidity between affective disorder and alcoholism among the relatives of all family types. However, analysis revealed no evidence for cosegregation of depression and alcoholism, but did provide significant evidence for correlation of the nonfamilial components of depression and alcoholism. These investigators argued that excessive comorbidity between alcoholism and unipolar depression is secondary to nonfamilial factors (eg, primary alcoholism causing nonfamilial depression as a complication of demoralization, or primary depression causing nonfamilial alcoholism as a result of social decline). In contrast to the lack of cosegregation of depression and alcoholism, a modest correlation between the familial components of alcoholism and bipolar disorder was observed. In particular, the investigators found that the relatives of comorbid probands (bipolar disorder plus alcoholism) were at elevated risk for bipolar disorder (with or without alcoholism) compared with the relatives of probands in whom bipolar disorder was the only illness or was absent. There was no correlation between nonfamilial components for either disorder. The investigators suggested that the observed comorbidity and cofamiliality of bipolar disorder and alcoholism may be secondary to the presence of a distinct form of familial disorder that may be expressed as bipolarity or alcoholism, or both, and is enriched in families identified through comorbid probands.
Synthesis and Hypothesis
The studies reviewed previously demonstrate a familial aggregation of depression, bipolar disorder, and alcoholism that is independent of the age of the proband. The age and type of the proband, however, do affect the relative prevalence of these disorders in family members. When correcting for age of onset and cohort effects, there is a 2-fold to 5-fold increased prevalence of mood disorders among the adult relatives of childhood-onset depression probands.[23] For those children who progress to bipolar disorder, there is a higher prevalence of alcoholism not associated with mood disorders in adult relatives.[22] When comparing the distribution of illness among relatives of child or adult probands, the relatives of bipolar children most closely resemble the adult relatives of comorbid bipolar/alcoholism probands. As discussed above by Maier and Merikangas,[29] the probands with comorbid bipolar disorder and alcoholism may represent a distinct etiologic entity in which the same familial components result in bipolar disorder or alcoholism, or both.
My colleagues and I have previously argued that early-onset mood disorders are more familial and may represent more genetically homogeneous types of disorders.[20] I hypothesize that these families in which the proband is a child with early-onset, bipolar disorder represent a special etiologic form of bipolar disorder such as that described by Maier and Merikangas.[29] If correct, this finding would have important clinical implications for the treatment, education, and surveillance of these children and families. From a research perspective, this hypothesis suggests that, for genetic studies, families identified through adult bipolar probands should be treated separately, based on the presence or absence of alcoholism and the age of onset of bipolar disorder. Moreover, this hypothesis suggests that such comorbid adult proband families could be combined with childhood-onset, bipolar-disorder families for genetic analysis. The obvious testable prediction of this hypothesis is that when these childhood-onset individuals become adolescents or young adults, the vast majority will develop serious alcohol abuse, or other types of substance-abuse, problems.
About the Author
Dr. Todd is the Blanche F. Ittleson Professor of Psychiatry and Professor of Genetics in the Departments of Psychiatry and Genetics at the Washington University School of Medicine, St. Louis, Mo.
I would like to acknowledge the contributions of my collaborators to our earlier studies, including Barbara Geller, MD, Wendy Reich, PhD, Ted Reich, MD, and Rosalind Neuman, PhD.
Editorial Comment
Richard Todd summarizes important recent work on the lineage and linkage of bipolar disorder, alcoholism, and depression in his article, "The Link Between Parental Alcoholism and Childhood Mood Disorders: A Familial/Genetic Perspective." Much of the work covered in this review has been done by Todd and colleagues at Washington University; a novel study by Maier and Merikangas[1] is also discussed. The studies warrant discussion at several levels: results, methods, and heuristic implications.
The results are clear but somewhat complex. Bipolar disorder in parents increases the risk of early mood disorders across several generations. More importantly, the relationship is specific to affective disease (ie, there is no increased risk of another disorder, such as attention-deficit/hyperactivity disorder or conduct disorder). A more complex relationship appears to exist between alcoholism in parents and early-onset psychiatric disorders in children, with increased frequency of both depression and various anxiety-related conditions.
Perhaps most notably, Todd and associates used a different paradigm to examine the transmission of mood disorders in families, one that is inverse to that used in the previously mentioned studies. These researchers started with the child who had early-onset depression or unequivocal bipolar disorder, and then looked backward at the prevalence of illness in parents and grandparents. The strength of this approach is starting with a powerfully expressed index situation (ie, early-onset disease), one that does not wait until the third or fourth decade of life to make an appearance. There is strong evidence that such early-onset illness is relatively severe. The prevalence of alcoholism and mood disorders in the parents of such children is quite elevated. However, the most striking difference between the patient group and the healthy control group was the prevalence of bipolar affective disorder in parents of children with depression, which was 15%, versus 0% in parents of matched healthy control children. There was also a striking difference in the prevalence of alcoholism between the parents of the 2 groups. The rates were 20.3%, 12.3%, and 6.4% in parents of children with bipolar disorder, depression, and good health, respectively.[2] Most recently, Maier and Merikangas[1] demonstrated an association between bipolar disorder and the familial component of alcoholism using a clever, scientifically sound design that helps to tease apart biologically linked relationships and relationships that may be consequent to secondary, nonbiologic factors.
Methodologically, researchers studying these disorders should consider the strengths of the designs that start with early-onset expression of illness. It is likely that, at least for bipolar disorder, earlier expression is more reflective of biologically expressed, rather than environmentally influenced, factors. Indirect evidence suggests that a host of environmental factors may precipitate or worsen the expression of bipolar disorder.[3] These factors include prescription and illicit drugs, artificial light and other circadian rhythm disturbances, and adult-onset neurologic and other general medical disorders. Although important in understanding the spectrum of bipolar-type syndromes, these factors merely add noise to efforts at identifying possible genetically influenced relationships, and should be considered as less prominent in early-onset illness.
Bipolar disorder offers a unique opportunity to study genetic relationships. In its full manic form, there is almost no false-positive identification of cases, since the key features of the illness--reduced need for sleep, excessive energy and zeal, speeded associations and speech, and even at times improved productivity--are not characteristic of more than a small number of drug-induced states. There are caveats in that bipolar disorder among youth most often presents with depressive or mixed episodes, often quite short in duration. However, this occurrence may be less of a disadvantage than it is in adult-assessed depression, since consistent but inconclusive evidence suggests that most early-onset mood disorder will ultimately appear clinically as bipolar disorder.[4]
This situation is less optimal than one would wish for. Limiting illness designation to a syndromal, cross-sectional diagnostic scheme is a relatively weak strategy. However, in the absence of biological markers, we may not be able to do better.
The same strategies employed by Todd and colleagues and by Maier and Merikangas may be usefully applied in standard genetic linkage studies.[5] They also lend themselves to study of candidate neurotransmitter and second-messenger systems (eg, myristoylated alanine-rich C-kinase substrate [MARCKS] protein, calcium signaling, protein kinase C, cortisol function, and norepinephrine and dopamine function). Of course, one would wish to have equally rigorous clinical and biologic assessments in such studies. A potential benefit of including biologic measures is earlier and more confident identification of cases, or identification of persons with a high likelihood of developing the fundamental illness. This inclusion could, in turn, improve efficiency of diagnosis, facilitate early treatments, and provide educational information about bipolar disorders and their complications, thereby substantially reducing morbidity over the lifetime expression of these disorders. The potential advances that can be made in the alcoholism component alone warrant serious studies in this area.
