Mood Disorders Historical Introduction and Conceptual Overview

Technical and detailed article

Clinical and Public Health Scope of Mood Disorders


Depressive disorders afflict one of five women and one of ten men at some time during their lives. Depressive episodes alternating with mania or hypomania represent the domain of bipolar disorders. Increasingly, the conventional figure of 1 percent for bipolar disorders in the general population is being challenged, and there are now convincing data that this group of disorders may account for 5 percent of the population and up to 50 percent of all depressions.

The enlargement of the boundaries for bipolar disorder is largely due to better detection of the bipolar II subtype (depression plus hypomania rather than mania). The current evidence for clinical, therapeutic, and public health implications of such a broadened bipolar concept have been summarized in a World Psychiatric Association monograph.

Despite the availability of effective treatments, many persons with mood disorders are disabled, and rates of suicide (which occurs in approximately 15 percent of depressive patients) are high in young and, particularly, elderly men.

An alarming increase of suicide rates among middle-aged women has been recently reported. Although depressive disorders are more common in women, the traditional view that more men than women die of suicide may therefore need revision in light of these data. High-profile cases of infanticide publicized on television have brought into the public's awareness the role of the reproductive cycle in severe postpartum psychosis and, more generally, the high burden of all forms of depression in women. It is therefore relevant that psychiatric journals targeting clinicians are increasingly devoting larger space to reproductive-related mood disorders in women. Obviously, more than hormones and physiology are involved in the morbidity and mortality of women, and broader developmental and life-cycle–related traumatic, social, and economic factors should be addressed in the research and clinical literature.

The suboptimal outcome of mood disorders documented in recent research reports cannot be ascribed to underdiagnosis and undertreatment alone, for several reasons.

First, Gerald Klerman and colleagues suggested that the incidence of mood disorders may be increasing in younger age groups, especially in cohorts born in the 1960s, and may be associated with rising rates of alcohol and substance abuse.

Second, mood disorders, once believed to be essentially adult disorders, are increasingly being diagnosed in children and adolescents.

Third, clinical studies suggest higher rates of chronicity, recurrence, and refractoriness than previously believed. For instance, chronicity, reported by Emil Kraepelin to occur in no more than 5 percent of this population in the early twentieth century in Germany, is now seen in varying degrees in one of three affectively ill patients. Nonetheless, outcome studies coming from university centers tend to overestimate the proportion of cases with less favorable prognosis, and, undeniably, many patients seen in private practice experience a favorable outcome. In addition, not unexpectedly, current data indicate that depressed patients treated by psychiatrists in private settings receive much better care than those in other settings.

Concepts of Mood Disorders

In the European tradition, the broader rubric of affective disorder (which subsumes mood and anxiety disorders) has been conceptualized along two influential schools. 

Aubrey Lewis and his followers from the Maudsley school have promoted a continuum model—from anxiety disorders to mild neurotic depressions to severe endogenous and psychotic depressions. The Newcastle school, led by Martin Roth, has sharply demarcated those conditions from one another. 

Although vestiges of both approaches are still influential in clinical and basic research, their significance is now overshadowed by European studies in German-speaking countries that subdivide mood disorders on the basis of polarity: Unipolar (depressive episodes only) and bipolar (depressive episodes plus manic, hypomanic, or mixed episodes). That subdivision, in part supported by studies in the United States, has served as the basis for much recent research into the biology, treatment, and classification of mood disorders, and is reflected in the text revision of the fourth edition of the Diagnostic and Statistical Manual of Mental Disorders (DSM-IV-TR) and the tenth revision of the International Statistical Classification of Diseases and Related Health Problems (ICD-10).

Despite such official sanction, many authorities continue to see considerable continuity between recurrent depressive and bipolar disorders. This has led to widespread discussion and debate about the bipolar spectrum, which incorporates classic bipolar disorder, bipolar II, and recurrent depressions.

Emerging data also tend to favor a continuum between juvenile and adult mood disorders. This is based on the pioneering contributions by Elva Poznanski at the University of Michigan, as well as the work of Leon Cytrin and colleagues at the National Institute of Mental Health (NIMH), Gabrielle Carlson in collaboration with Dennis Cantwell at the University of California at Los Angeles, and Joachim Puig-Antich at Columbia University in New York.

Childhood bipolarity, too, is receiving increasing clinical attention, thanks to the seminal work of Elizabeth Weller and colleagues, originally conducted at Ohio State University. In addition, clinical observations at the University of Tennessee on the juvenile offspring of adult patients with bipolar disorder have led to a greater appreciation of the bipolar nature of complex clinical presentations of affective illness in the juvenile offspring and kin of adult bipolar probands.

More recent work by Biederman's group at Harvard suggests intriguing links between pediatric bipolar disorder and attention-deficit/hyperactivity disorder (ADHD). Current concepts of mood disorders in the United States embrace a wide spectrum, including many conditions previously diagnosed as schizophrenia, personality disorder, or neurosis.

The diagnostic shift occurred in part as a result of the U.S.–U.K. Diagnostic Project, which demonstrated that schizophrenia was being diagnosed at the expense of mood disorders. Conceptual boundaries were further broadened by the availability of new and effective treatments and by the unacceptable risk for tardive dyskinesia and suicide in persons with misdiagnosed mood disorders.

More generally, present research interest in mood disorders in the United States emanated from a landmark 1969 NIMH conference on the psychobiology of affective illnesses. The NIMH Collaborative Depression Study—a long-term prospective project deriving directly from recommendations made at the conference—has legitimatized the broader perspective.

Morbidity and Mortality

Unfortunately, findings published by Martin Keller and colleagues in the 1980s documenting gross undertreatment of mood disorders continue to describe the current treatment landscape worldwide. Whatever changes have occurred in diagnostic practice do not appear to have significantly affected the morbidity and mortality of mood disorders. This is all the more scandalous because the 1990s saw new classes of user-friendly antidepressant and mood-stabilizing agents, as well as depression-specific psychotherapies.

This state of affairs results, in part, from the fact that clinical exposure to mood disorders in both specialized (psychiatric) and primary care (general medical) training is suboptimal. Mood disorders, as highly prevalent and lethal disorders, must command a greater share in the clinical curriculum of both psychiatrists and general medical practitioners. Because most mood disorders are chronically relapsing conditions, long-term exposure to patients with these disorders in mood or bipolar clinics should be obligatory training for young doctors.

Unfortunately, few academic centers have such clinics, and those that exist are largely devoted to research. The primary goal of these clinics is the execution of research protocols rather than the acquisition of clinical experience in caring for such patients.

Nearly half of all cases of depression, just like those with adult-onset diabetes, remain undetected for years or inadequately controlled—both of which seem to lag behind hypertension, in which early detection and treatment have significantly reduced complications such as stroke.

Efforts by patient-advocacy organizations—often in concert with national psychiatric organizations and governmental mental health agencies—appear to be increasing public and government awareness of mood disorders. Ultimately, however, the challenge is to provide all primary care physicians with the requisite hands-on experience in this prevalent group of disorders.

User-friendly tools to detect suicidal patients in the general medical sector would further enhance preventive efforts. These would require significant changes in the structure of health care, including, but not limited to, the greater participation of nurses and social workers as liaisons in primary mental health and continuity of care for depressed or suicidal patients.

Because mood disorders underlie 50 to 70 percent of all suicides, effective treatment of these disorders on a national level should, in principle, drastically reduce this major complication of mood disorders. That elderly depressed patients, often with medical comorbidity, constitute the highest-risk group for suicide yet escape clinical detection and treatment is particularly problematic for public health.

A small-scale prospective Swedish study undertaken by Rihmer and Rutz, although not specifically targeted to the elderly, yielded promising results in this regard. In addition, clinical findings in recurrent mood disorders have clearly shown the value of lithium prophylaxis in the prevention of suicide and overall mortality. Emerging data suggest that such benefits may accrue from all efficacious treatments for mood disorders.

Suicide prevention at the national level, attributed to the widespread use of selective serotonin reuptake inhibitors (SSRIs) and primary care physician education, has been achieved in many but not all European countries. Such prevention does not have a tradition in the United States and, as mentioned earlier, suicide rates have actually increased in the United States in the last few years, from 10.5 percent to 11 percent, a net increase of 5 percent.


Mood disorders encompass a large group of psychiatric disorders in which pathological moods and related vegetative and psychomotor disturbances dominate the clinical picture. Known in previous editions of DSM as affective disorders, the term mood disorders is preferred today because it refers to sustained emotional states, not merely to the external (affective) expression of the present emotional state.

Mood disorders are best considered as syndromes (rather than discrete diseases) consisting of a cluster of signs and symptoms, sustained over a period of weeks to months, that represent a marked departure from a person's habitual functioning and tend to recur, often in periodic or cyclical fashion.

Major Depressive Disorder and Bipolar Disorder

Major depressive disorder (unipolar depression) is reported to be the most common mood disorder. It may manifest as a single episode or as recurrent episodes. The course may be somewhat protracted—up to 2 years or longer—in those with the single-episode form. Whereas the prognosis for recovery from an acute episode is good for most patients with major depressive disorder, three of four patients experience recurrences throughout life, with varying degrees of residual symptoms between episodes.

Bipolar disorders (previously called manic-depressive psychosis) consist of at least one hypomanic, manic, or mixed episode. Mixed episodes represent a simultaneous mixture of depressive and manic or hypomanic manifestations. Although a minority of patients experience only manic episodes, most bipolar disorder patients experience episodes of both polarity. Manias predominate in men, depression and mixed states in women. 

The bipolar disorders were classically described as psychotic mood disorders with both manic and major depressive episodes (now termed bipolar I disorder), but recent clinical studies have shown the existence of a spectrum of ambulatory depressive states that alternate with milder, short-lived periods of hypomania rather than full-blown mania (bipolar II disorder).

Bipolar II disorder, which is not always easily discernible from recurrent major depressive disorder, illustrates the need for more research to elucidate the relation between bipolar disorder and major depressive disorder. The border with schizophrenia continues to be fuzzy.

The Kraepelinian dichotomy of dementia praecox and manic-depressive illness has been challenged in favor of a spectrum concept that goes beyond mood-incongruent mood episodes and might spill over into “good-prognosis” schizophrenia.

Dysthymia and Cyclothymia

Clinically, major depressive episodes often arise from a low-grade, intermittent, and protracted depressive substrate known as dysthymic disorder. Likewise, many instances of bipolar disorders, especially ambulatory forms, represent episodes of mood disorder superimposed on a cyclothymic background, which is a biphasic alternating pattern of numerous brief periods of hypomania and numerous brief periods of depression.

Dysthymic and cyclothymic disorders represent the two prevalent subthreshold mood conditions roughly corresponding to the basic temperamental dysregulations described by Emil Kraepelin and Ernst Kretschmer as predisposing to affective illness. It is not always easy to demarcate full-blown syndromal episodes of depression and mania from their subthreshold counterparts commonly observed during the interepisodic periods.

The subthreshold conditions appear to be fertile terrain for interpersonal conflicts and postaffective pathological character developments that may ravage the lives of patients and their families. In North America and some Western European countries many such patients end up being labeled with borderline personality disorder, which, unfortunately, often tends to obscure the affective origin of the presenting psychopathology.

Cyclothymic and dysthymic conditions also exist in the community without progression to full-blown mood episodes. As such, they are best considered, respectively, as trait bipolar and trait depressive conditions.

Understanding the factors that mediate transition from trait to clinical state is important for preventing manic and major depressive episodes. A fascinating development related to these temperaments pertains to the “positive” attributes of the depressive and related temperaments in subserving sensitivity to human suffering, thereby playing a role in altruistic behavior.

The more tempestuous cyclothymic, based on recent data from different centers, appears to be the temperament involved in poetry, the arts, architecture, and other forms of creative human endeavor. In other words, the “dilute” forms of affective illness may have evolved as traits of fundamental importance for human culture.

Other Subthreshold Mood States

Epidemiological studies in both Europe and North America have also revealed other subsyndromal conditions with depressive and hypomanic manifestations with few symptoms (oligosymptomatic mood states) and of short duration (brief episodes). Variously referred to as minor, subsyndromal, brief, or intermittent, these descriptions do not merely represent arbitrary lowering of diagnostic thresholds, but they also herald increasing realization of their importance in early detection of at-risk individuals, as has happened in other medical fields (e.g., diabetes mellitus and essential hypertension).

If disabling mood disorders afflict 5 to 8 percent of the general population (Epidemiologic Catchment Area [ECA] study), milder but still clinically significant mood disorders would raise lifetime rates to 17 percent (National Comorbidity Study [NCS]); if subclinical mood states are added, that figure doubles to involve one third of the general population (as reported, e.g., by Kenneth Kendler and colleagues).

New evidence from both Europe and the United States has shown that bipolar spectrum conditions (bipolar I, bipolar II, and bipolar disorder not otherwise specified [NOS] in formal diagnostic manuals such as the DSM-IV-TR and the ICD-10) may account for at least 50 percent of all mood disorders in the community and in psychiatric practice.

Comorbidity in mood disorders involves considerable overlap with anxiety disorders. As summarized in an NIMH monograph, anxiety disorders can occur during an episode of depression, may be a precursor to the depressive episode, and, less commonly, may occur during the future course of a mood disorder. Those findings suggest that at least some depressive disorders share a common diathesis with certain anxiety disorders.

More recent clinical experience suggests intriguing comorbidity patterns between bipolar II disorder on one hand and panic, obsessive-compulsive, and social phobic states on the other. Furthermore, bipolar I and II disorders are particularly likely to be complicated by the use of alcohol, stimulants, or both. In many cases, the alcohol or substance abuse represents attempts at “self-treatment” of the depression and associated anxiety or insomnia (or both) and, in the case of mania and hypomania, attempts to maintain or enhance the positive moods and energy.

Finally, physical illness—both systemic and cerebral—occurs in association with depressive disorders with a greater frequency than expected by chance alone. Unless properly treated, such depression negatively affects the prognosis of the physical disorder.

More provocatively, there is current reawakening to the contribution of cerebral and cardiovascular factors to the origin of late-onset psychotic depressions (previously classified as involutional melancholia). An integrated framework of pathogenesis is necessary for understanding psychopharmacological, somatic, and psychotherapeutic approaches in the clinical management of patients with mood disorders.

A historical perspective on current developments is also a valuable lesson in the study of mood disorders.

Ancient Greek and Roman Descriptions

Much of what is known today about mood disorders was described by the ancient Greeks and Romans, who coined the terms melancholia and mania and noted their relation. The ancients also hypothesized a temperamental origin for those disorders. Much of modern thinking about mood disorders (e.g., the work of French and German schools in the middle and latter parts of the nineteenth century, which influenced current British and U.S. concepts) can be traced back to these ancient concepts.


Hippocrates (460 to 357 Before the Common Era [BCE]) described melancholia (“black bile”) as a state of “aversion to food, despondency, sleeplessness, irritability, and restlessness.” Thus, in choosing the name of the condition, Greek physicians (who may have borrowed the concept from ancient Egyptians) postulated the earliest biochemical formulation of any mental disorder. They believed that the illness often arose from the substrate of the somber melancholic temperament, which, under the influence of the planet Saturn, made the spleen secrete black bile, ultimately leading to mood darkening through its influence on the brain.

Greek descriptions of the clinical manifestations of depression and of the temperament prone to melancholia are reflected in the DSM-IV-TR in the subdepressive lethargy, self-denigration, and habitual gloom of the person with dysthymic disorder.

One Hippocratic aphorism recognized the close link between anxiety and depressive states: “Patients with fear of long-standing are subject to melancholia.” Hippocrates, who described the first historical case of melancholia, may have also been the first to describe a depressive mixed state, an activated form of depression: A woman of Thasos became morose because of a justifiable grief, and although she did not take to her bed, she suffered from insomnia, loss of appetite…she complained of fears and talked much; she showed despondency and…talked at random and used foul language…many intense and continuous pains…she leapt up and could not be restrained. [Emphasis added]

According to Galen (131 to 201), melancholia manifested in “fear and depression, discontent with life, and hatred of all persons.” A few hundred years later, another Roman, Aurelianus, citing the now-lost works of Soranus of Ephesus, amplified the role of aggression in melancholia (and its link to suicide) and described how the illness assumed delusional coloring: “Animosity toward members of the household, sometimes a desire to live and at other times a longing for death, suspicion on the part of the patient that a plot is being hatched against him.”

In addition to natural melancholia, which, presumably, arose from an innate predisposition to overproduce the dark humor and led to a more severe form of the malady, ancient medicine recognized such environmental contributions to melancholia as immoderate consumption of wine, perturbations of the soul due to the passions (e.g., love), and disturbed sleep cycles. Autumn was considered the season most disposing to melancholy.


A state of raving madness with exalted mood was noted by the ancient Greeks, although it referred to a somewhat broader group of excited psychoses than that in modern nosology.

Its relation to melancholia was probably noted as early as the first century BCE, but, according to Aurelianus, Soranus discounted it. Nonetheless, Soranus had observed the coexistence of manic and melancholic features during the same episode, consisting of continual wakefulness and fluctuating states of anger and merriment and, sometimes, of sadness and futility. Thus, Soranus seemed to have described what today are called mixed episodes in the DSM-IV-TR and the ICD-10.

Natural melancholy was generally considered a chronic disorder, but Soranus noted the tendency for attacks to alternate with periods of remission. Although others before him hinted at it, Aretaeus of Cappadocia (ca. 150) is generally credited with making the connection between the two major mood states: “It appears to me that melancholy is the commencement and a part of mania.” He described the cardinal manifestations of mania as it is known today: There are infinite forms of mania, but the disease is one of them. If mania is associated with joy, the patient may laugh, play, dance night and day, and go to the market crowned as if a victor in some contest of skill. The ideas the patients have are infinite. They believe they are experts in astronomy, philosophy, or poetry.

Aretaeus described the extreme psychotic excitement that could complicate the foregoing clinical picture of mania: The patient may become excitable, suspicious, and irritable; hearing may become sharp…[and they might] get noises and buzzing in the ears; or may have visual hallucinations; bad dreams and his sexual desires may get uncontrollable; aroused to anger, he may become wholly mad and run unrestrainedly, roar aloud; kill his keepers, and lay violent hands upon himself.

Noting the fluctuating nature of symptoms in the affectively ill, Aretaeus commented: “They are prone to change their mind readily; to become base, mean-spirited, illiberal, and in a little time extravagant, munificent, not from any virtue of the soul, but from the changeableness of the disease.” Aretaeus was thus keenly aware of the characterological distortions so commonly manifested during the different phases of cyclical mood disorders.

Finally, consolidating the knowledge of several centuries, Aretaeus described mania as a disease of adolescent and young men given intermittently to “active habits, drunkenness, lechery” and an immoderate lifestyle (what today might be called cyclothymic disorder). Exacerbations were most likely to occur in the spring.

Affective Temperaments

The concept of health and disease in ancient medicine was based on harmony and balance of the four humors, of which sanguine humor was deemed the healthiest. Even a desirable humor, however, such as blood, which made persons habitually active, amiable, and prone to jest, could, in excess, lead to the pathological state of mania.

The melancholic temperament, dominated by black bile and predisposed to pathological melancholia, was described as lethargic, sullen, and given to brooding or contemplation; its modern counterparts are depressive personality disorder (now in a DSM-IV-TR appendix) and its clinical expression as dysthymic disorder (included in both the ICD-10 and the DSM-IV-TR).

A long tradition dating back to Aristotle (384 to 322 BCE) attributed creative qualities to the otherwise tortured melancholic temperament in such fields as philosophy, the arts, poetry, and politics. The remaining two temperaments—choleric and phlegmatic—were less desirable, in that yellow bile made persons choleric (irritable, hostile, and given to rage) and phlegm made them phlegmatic (indolent, irresolute, and timid).

The choleric and phlegmatic temperaments would probably be recognized today as borderline personality disorder and avoidant or schizoid personality disorder, respectively.

Many of the original Greek texts on melancholia were transmitted to posterity through medieval Arabic texts such as those of Ishaq Ibn Imran and Avicenna (and their Latin renditions by Constantinus Africanus). In describing different affective states, Avicenna developed the theory of the temperaments to its fullest. He speculated that a special form of melancholia supervened “if black bile be mixed with phlegm,” when the illness was “coupled with inertia, lack of movement, and quiet.”

Furthermore, mania was not necessarily linked to the sanguine (what today is termed hyperthymic) temperament, in that many forms of excited madness were believed to represent a mixture of black and yellow bile.

Avicenna further observed that the mixture of anger and restlessness in melancholia indicated that the disease was manic in nature and that the appearance of such signs and symptoms along with violence heralded the transition from melancholia to mania. Avicenna was prescient in this respect because these activated, irritable depressions with racing thoughts have yet to receive the DSM-IV-TR “blessing” for being classified as bipolar mixed states.

Those elaborations on Galen's temperamental types might be considered the forerunners of current personality dimensions, deriving mood states from various mixtures of neuroticism and introversion-extroversion. (What both the ICD-10 and the DSM-IV-TR describe as cyclothymic disorder represents the intense mood lability of high neuroticism coupled with cyclic alternation between extroversion and introversion.)

Speculation on how diverse depressive phenomena could be understood as a mix of humors anticipated modern multiple-transmitter hypotheses of depression. Ishaq Ibn Imran summarized the existing knowledge of melancholia by considering the interaction of genetic factors (“injured prenatally as the result of the father's sperm having been damaged”) with a special temperament given to “mental overexertion”—although not necessarily physical overactivity—that, in turn, was associated with “disruption of the correct rhythms of sleeping and waking.” Those views, too, have a very modern ring to them. 

Modern ERA

The first English text entirely devoted to affective illness was Robert Burton's Anatomy of Melancholy, published in 1621. A scholarly review of medical and philosophical wisdom accumulated in past centuries, it also anticipated many modern developments. The concept of affective disorder endorsed by Burton was rather broad (as it always has been in the United Kingdom), embracing mood disorders and many disorders that are today considered somatoform disorders, including hypochondriasis. Although he described “causeless” melancholias, Burton also categorized the various forms of love melancholy and grief. Particularly impressive was his catalogue of causes, culminating in a grand conceptualization:

Such as have Saturn misaffected in their genitures such as are born of melancholy parents as offend in those six non-natural things, are of a high sanguine complexion, are solitary by nature, great students, given to much contemplation, lead a life out of action, are most subject to melancholy. Of sexes both, but men more often. Of seasons of the year, autumn is most melancholy. Jobertus excepts neither young nor old.

Burton's six nonnatural things referred to such environmental factors as diet, alcohol, biological rhythms, and perturbations induced by passions such as intense love. Burton did not definitively indicate age prevalences. Like nearly all of his predecessors, he favored male (rather than the currently reported female) preponderance. Finally, Burton considered both the melancholic (contemplative) and the sanguine (hot-blooded) temperaments to be substrates of melancholia. Burton's work thus linked certain forms of depression with the softer expressions of the manic disposition, or bipolar II disorder, from which he appears to have suffered.

The eighteenth and nineteenth centuries introduced humane hospital care of the mentally ill, thereby permitting systematic clinical observation of the psychopathology and outcome of mood disorders. Concept of Affective Disorder Although Celsus (ca. 30) had described “forms of madness that go no further than sadness,” the French alienist Jean-Philippe Esquirol (1840) may have been the first psychiatrist in modern times to suggest that a primary disturbance of mood might underlie many forms of depression and related paranoid psychoses.

Until Esquirol's work, melancholia had been categorized as a form of insanity (i.e., ascribed to deranged reasoning or thought disturbance). Esquirol's observations on melancholic patients led him to postulate that their insanity was partial (dominated by one delusion, a monomania) and that “the symptoms were the expression of the disorder of the affections. The source of the evil is in the passions.” He coined the term lypemania (from the Greek, “sorrowful insanity”) to give nosological status to a subgroup of melancholic disorders that were affectively based. Esquirol cited Benjamin Rush (1745 to 1813), the father of American psychiatry, who had earlier described tristimania, a form of melancholia in which sadness predominated.

Esquirol's influence led other European psychiatrists to propose milder states of melancholia without delusions, which were eventually categorized as simple melancholias and, ultimately, as primary depressions. Such descriptions culminated in the Anglo-Saxon psychiatric term affective disorder, coined by Henry Maudsley (1835 to 1918), the renowned British psychiatrist after whom a London hospital is named.

Manic-Depressive Illness and the Question of Psychogenic Depressions

Although the connection between mania and depression had been rediscovered sporadically since it was first described 2,000 years ago, the clinical work that finally established circular insanity (Jean-Pierre Falret's term) and folie à double forme (Jules Baillarger's term) as discrete nosological entities with both depressive and manic poles was undertaken by these two Esquirol disciples in the 1850s. That accomplishment built on Philippe Pinel's reforms, which championed humane treatment of the mentally ill in Paris around the turn of the eighteenth century and emphasized systematic clinical observations of patients, which were detailed in case records.

French alienists made longitudinal observations on the same patient from one psychotic attack into another. Furthermore, Esquirol had introduced the practice of chronicling events in statistical tables. Thus, the Hippocratic approach to defining a particular case by its onset, circumstances, course, and outcome was applied by French alienists in studying the affectively ill. The humanitarian reforms introduced in the nineteenth century ensured that standards of general health and nutrition would improve the outlook for the mentally ill—especially those with potentially reversible disorders such as affective disorders, who could now be discharged from the asylums. The French school, by segregating the nondeteriorating mood disorders from other types of insanity, then paved the way for the Kraepelinian system.

Kraepelin's (1856 to 1926) unique contribution was not so much his grouping together of all the forms of melancholia and mania, but his methodology and painstaking longitudinal observations, which established manic-depressive illness as a nosological entity and (he hoped) a disease entity. His rationale was based on the following:

  1. the various forms had a common heredity measured as a function of familial aggregation of manic and depressive cases,
  2. frequent transitions from one to the other occurred during longitudinal follow-up of patients,
  3. a recurrent course with illness-free intervals characterized most cases,
  4. the superimposed episodes were commonly opposite to the patient's habitual temperament—that is, mania could be superimposed on a depressive temperament and depression on a hypomanic temperament—and
  5. both depressive and manic features could occur during the same episode (mixed states).

Kraepelin's synthesis was developed as early as the sixth (1899) edition of his Lehrbuch der Psychiatrie and most explicitly stated in the opening passages of the section on manic-depressive psychosis in the eighth edition (published in four volumes, 1909 to 1915): Manic-depressive insanity includes on the one hand the whole domain of so-called periodic and circular insanity, on the other hand simple mania, the greater part of the morbid states termed melancholia and also a not inconsiderable number of cases of confusional insanity.

Lastly, we include here certain slight and slightest colorings of mood, some of them periodic, some of them continuously morbid, which on the one hand are to be regarded as the rudiment of more severe disorders, on the other hand, pass over without boundary into the domain of personal predisposition.

For Kraepelin, the core pathology of clinical depression consisted of lowered mood and slowed (retarded) physical and mental processes. In mania, by contrast, the mood was elated, and physical and mental activity were accelerated. His earlier observations on what he termed involutional melancholia (referring to 40-to 65-year-old patients with extreme anxiety, irritability, agitation, and delusions) had led him to separate that entity from the broader manic-depressive rubric. In the eighth edition of Lehrbuch der Psychiatrie, however, he united melancholia with the manic-depressive group, with the justification that it was a special form of mixed state and that follow-up conducted by his pupil Dreyfus had demonstrated unmistakable excited phases. 

The classification of mood disorders is still evolving. Karl Leonhard in 1957, Jules Angst in 1966, Carlo Perris in 1966, and George Winokur, Paula Clayton, and Theodore Reich in 1969, working independently in four different countries, proposed that depressive disorders without manic or hypomanic episodes (unipolar depressive disorder) that appear in middle age and later are distinct from depressive episodes that begin at earlier ages and alternate with manic or hypomanic episodes (bipolar disorder). The main difference between the two affective subtypes is the greater familial loading for mood disorder—especially for bipolar disorder—among bipolar disorder probands. 

Kraepelin had conceded the occurrence of psychogenic states of depression occasioned by situational misfortune. However, he believed that manic-depressive illness was hereditary, although he could not document postmortem anatomopathological findings in the brains of manic-depressive patients. Therefore, manic-depression had to be considered a functional mental disorder in which brain disturbances were presumed to lie in altered physiological functions. Such biological factors were deemed absent in the psychogenic depressions. Thus, Kraepelin's classification of mood disorders is both dualistic and unitary. It is dualistic to the extent that he designated them as either psychologically occasioned or somatically caused. It is unitary with respect to disorders in the latter group, which have been termed endogenous affective disorders (i.e., due to internal biological causes). In other words, Kraepelin restricted the concept of clinical depression to what the DSM-TR-IV terms major depressive disorder with melancholic features. Moreover, he postulated a continuum between that condition and what the DSM-IV-TR and the ICD-10 now term bipolar disorders.

Table 13.1-1. Overlapping Dichotomies of Affective Disorders That Are Not Necessarily Synonymous Manic-depressive Psychogenic S (somatic) type J (justified) type Autonomous Reactive Endogenous Exogenous Psychotic Neurotic Acute Chronic Major Minor Melancholic Neurasthenic Typical Atypical Primary Secondary Biological Characterological 

As summarized in Table 13.1-1, until recently, endogenous depressions were contrasted with those of exogenous cause (i.e., external and, presumably, psychogenic causes). Transitions between the two groups are so frequent, however, that the two-type thesis of depression has been largely abandoned in official classifications in North American psychiatry and most parts of the world. In one study conducted by Akiskal's mood clinic team in Memphis, Tennessee, during the 1970s, 100 patients with neurotic depression (the prototype of exogenous depression), prospectively followed over 3 to 4 years, developed episodes with endogenous, psychotic, and even bipolar features (Table 13.1-2). Nonetheless, the endogenous–exogenous dichotomous grouping still has a few adherents in England and Australia who continue to research its potential for clinical predictions. Such research generally attempts to validate the various subtypes on the basis of their clinical characteristics rather than presumed cause. Today, most mood disorder experts would probably agree that depressive illness has endogenous and exogenous components in most patients presenting clinically. This does not necessarily imply that a continuum exists between all forms of depressive disorders, but suggests that neurotic and endogenous clinical features are not the best way to capture the heterogeneity of these disorders. Consensus would be less likely reached on how to delimit clinical depressive disorder from comorbid disorders such as the various anxiety disorders, substance use disorders, and personality disorders. Clarifying the boundaries between those disorders has emerged as a principal challenge in the classification of mood disorders.

Table 13.1-2. Three- to Four-Year Prospective Follow-up in Neurotic Depressions (N = 100) Diagnosis and Outcome Numbera Manic episode 4 Hypomanic episode 14 Psychotic depression 21 Endogenous depression 36 Episodic course 42 Unstable characterological features 24 Social invalidism 35 Suicide 3

aThe total exceeds 100 because more than one outcome was possible in each patient. Summarized with permission from Akiskal H, Bitar A, Puzantian V, Rosenthal TL, Walker PW: The nosological status of neurotic depression: A prospective 3- to 4-year examination in light of the primary–secondary and unipolar–bipolar dichotomies. Arch Gen Psychiatry. 1978;35:756. Printed with permission.

Cartesian thinking in seventeenth-century France conceptually separated mind from body, thereby providing physicians autonomy over the somatic sphere, free from interference by the Church. The dichotomous paradigm ensured that the study of the two aspects of the human organism would not be confounded by the complexities of mind–body interactions. That is one reason Kraepelin's descriptive observations have proved valuable to subsequent generations of clinicians. Furthermore, his approach exemplifies the best tradition of scientific humanism in medicine: Description and diagnostic categorization of an individual patient are necessary for the physician to apply the knowledge gained from past observation of similarly described and diagnosed patients. One limitation to the Kraepelinian approach is that because of its biological reductionism, it is not sufficiently articulate to account for mind–body interactions in the genesis of mental disorders.

Depressions as Psychobiological Affective Reaction Types

It was the ambition of Swiss-born Adolf Meyer (1866 to 1950) to bridge the divide between psyche and soma. Meyer, who dominated psychiatry from his chair at Johns Hopkins University during the first half of the twentieth century, coined the term psychobiology to emphasize that both psychological and biological factors could enter into the causation of depressive and other mental disorders. Because of the nascent state of brain science during Meyer's time, he was more adept at biography than biology and therefore paid greater attention to psychosocial causation. He preferred the term depression (pressed down) to melancholia because of its lack of biological connotation. He conceived of depressive states in terms of unspecified constitutional or biological factors interacting with a series of life situations beginning at birth or even at conception. From that viewpoint arose the unique importance accorded personal history in depressive reactions to life events.

Meyer's terminological revision left a somewhat confusing legacy, in that the term depression is now applied to a broad range of affective phenomena ranging from sadness and adjustment disorders to clinical depression and bipolar disorders. Repercussions can be seen in the low threshold for diagnosing major depressive disorder in the DSM-IV-TR, which makes it difficult to differentiate major depressive disorder from transient life stresses that produce adjustment disorder with depressed mood. Nosological nuances to which Meyerians paid little attention, such as the difference between melancholic depression and more mundane depressions, are not just a matter of semantics. To the extent that those two forms of depression are seen in different clinical settings, hypotheses based on one population may not apply to the other. For instance, uncontrollable traumatic events may have taught study participants to feel helpless or to view the world in a negative light, but that does not equate with clinical depression; nor does the process appear to be specific to depression. Failure to make such nosological distinctions further clouds interpretations of the results of trials comparing psychotherapy and pharmacotherapy for depressive disorders.

On the other hand, the Meyerian emphasis on biographical factors for the patient represented a more practical approach to depth psychology. Recent sociological interpretations of depression can also be traced to Meyer's work. In final analysis, however, the Meyerian concern for the uniqueness of the individual has proved heuristically sterile. It de-emphasizes what is diagnostically common to different individuals, thereby obscuring the relevance of accrued clinical wisdom for the index patient. For that reason, the Meyerian approach, after enjoying clinical popularity for several decades in North America, has given way to neo-Kraepelinian rigor. However, the psychobiological vision of bridging biology and psychology, one of the major preoccupations of psychiatric thought and research today, owes much to Meyer's legacy.

Contemporary Models of Depression

From classical times through the early part of the twentieth century, advances in the understanding of mood disorders involved conceptual shifts from supernatural to naturalistic explanations; from reductionistic, unitary theories of causation to pluralistic theories; and from dualism to psychobiology. Knowledge of those conceptual developments provides a useful base from which to scrutinize models and concepts of mood disorder developed later in the twentieth century. The new approaches, derived from competing theoretical positions, have generated models for understanding various aspects of mood disorders, particularly depressive disorders.

The formative influence of early experience as it is dynamically shaped by emerging mental structures during development is the common denominator for the psychoanalytic concepts of psychopathological phenomena. By contrast, behavioral approaches in their more traditional formulations focus on the pathogenetic impact of proximate contexts. The cognitive approaches, which are akin to the behavioral–pathogenetic tradition, nonetheless concede that negative styles of thinking might mediate between proximate stressors and more remote experiences. In explaining the origin of mood disorders, all three schools—psychoanalytic, behavioral, and cognitive—emphasize psychological constructs. The biological models, however, are concerned with defining the somatic mechanisms that underlie or predispose individuals to morbid affective experiences. The schism between psychological and biological conceptualizations is an instance of the mind–body dichotomy that has characterized the Western intellectual tradition since Descartes. After all, psychological and somatic approaches represent merely convenient investigational strategies that attempt to bypass the methodological gulf between mental and neural structures. The ultimate aim is to understand how mood disorders develop within the psychoneural framework of a given person.

Aggression-Turned-Inward Model

Sigmund Freud was initially interested in a psychoneural project for all mental phenomena. Limitations of the brain sciences of the day led him to adopt instead a model that relied on a concept of mental function borrowed from physics. The notion that depressed affect is derived from retroflexion of aggressive impulses directed against an ambivalently loved internalized object was actually formulated by his Berlin disciple Karl Abraham and later elaborated by Freud. Abraham and Freud hypothesized that turned-inward anger was intended as punishment for the love object that had thwarted the depressed patient's need for dependency and love. Because, in an attempt to prevent the traumatic loss, the object had already been internalized, the patient then became the target of his or her own thanatotic impulses. A central element in those psychic operations was the depressed patient's ambivalence toward the object, which was perceived as a frustrating parent. Aggression directed at a loved object (parent) was therefore attended by considerable guilt. In the extreme, such ambivalence, guilt, and retroflexed anger could lead to suicidal behavior.

According to this model, depression is an epiphenomenon of the transduction of thanatotic energy, a reaction that takes place in the closed hydraulic space of the mind. Freud's earlier writings had similarly portrayed anxiety as being derived from the transformation of dammed-up sexual libido. Although Freud envisioned that psychoanalytic constructs would one day be localized neuroanatomically, the hydraulic mind is a metaphor that does not refer to actual physiochemical space in the brain.

Table 13.1-3. Major Models of Depression Proponents (Year)a Model Mechanism Scientific and Clinical Implications Karl Abraham (1911) Aggression turned inward Transduction of aggressive instinct into depressive affect Hydraulic mind closed to external influences; nontestable Sigmund Freud (1917) John Bowlby (1960) Object loss Disruption of an attachment bond Ego-psychological; open system; testable Edward Bibring (1953) Self-esteem Helplessness in attaining goals of ego ideal Ego-psychological; open system; social and cultural ramifications Aaron Beck (1967) Cognitive Negative cognitive schemata as intermediary between remote and proximate causes Ego-psychological; open system; testable; predicts phenomenology; suggests treatment Martin Seligman (1975) Learned helplessness Belief that one's responses will not bring relief from undesirable events Testable; predicts phenomenology; predicts treatment Peter Lewinsohn (1974) Reinforcement Low rate of reinforcement, or reinforcement presented noncontingently; social deficits might preclude responding to potentially rewarding events Testable; predicts phenomenology; predicts treatment Joseph Schildkraut (1965) William Bunney and John Davis (1965) Alec Coppen (1968) I. P. Lapin and G. F. Oxenkrug (1969) David Janowsky et al. (1972) Arthur Prange et al. (1974) Larry Siever and Kenneth Davis (1985) Biogenic amine (neurochemical) Impairment or dysregulation of aminergic transmission Testable; reductionistic; explains phenomenology and opposite episodes; suggests treatment Bernard Carroll et al. (1981) Impaired glucocorticoid and mineralocorticoid receptors (neuroendocrine) Testable; reductionistic; explains phenomenology; explains anxious comorbidity; suggests treatment Alec Coppen and D. M. Shaw (1963) Peter Whybrow and Joseph Mendels (1968) Robert Post (1990) Neurophysiological Electrophysiological disturbances leading to neuronal hyperexcitability and/or kindling Testable; reductionistic; explains phenomenology and recurrence; suggests treatment Hagop Akiskal and William McKinney (1973) Frederick Goodwin and Kay Jamison (1990) Final common pathway Stress-diathesis interaction converging on midbrain mechanisms of reward and biological rhythms Testable; integrative, psychobiological; pluralistic; explains phenomenology; suggests treatment

aDates provided for the models refer to the original paper or work in which they first appeared. In some instances, the bibliography at the end of the section provides references reflecting more updated thinking by those authors. Updated from Akiskal H, McKinney W: Overview of recent research in depression: Integration of 10 conceptual models into a comprehensive clinical frame. Arch Gen Psychiatry. 1975;32:285.

The concept of emotional behavior as an arena of incompatible forces confined to a psyche that is relatively impervious to current influences outside of the organism is the major liability of the aggression-turned-inward model and, perhaps, of orthodox psychoanalysis itself. Although the sexual-energy-transduction hypothesis of anxiety has been discarded in modern psychoanalytic thought, in modified versions, the aggression-turned-inward model continues to be used in clinical conceptualization. The lingering popularity of the model may be due, in part, to its compatibility with the clinical observation that many depressed patients suffer from lack of assertion and outwardly directed aggressiveness. However, a substantial number of hostile, depressed patients are also encountered in clinical practice (indeed, “depression with anger attacks” has been recently described), and clinical improvement in most patients typically leads to decreased, not increased, hostility. Such observations shed doubt on the aggression-turned-inward mechanism as a universal explanation for depressive behavior. Finally, little evidence exists to support the contention that outward expression of anger has therapeutic value in clinical depression.

Outwardly directed hostility in depression is not a new clinical observation; in fact, ancient Greek and Roman physicians had noted it. Hostility is best considered a manifestation rather than a cause of depressive disorder, especially when the disorder is attended by mixed bipolar features. The hostility of the depressed patient can also be understood as an exaggerated reaction to frustrating love objects, as secondary to self-referential attribution, or simply as nonspecific irritability of an ego in affective turmoil; this could, in part, be a function of a concurrent personality disorder from the erratic cluster. Such commonsense explanations that do not invoke unobservable hydraulic transmutations have greater appeal from heuristic and clinical perspectives.

Object Loss and Depression

Object loss refers to traumatic separation from significant objects of attachment. Ego-psychological reformulations of the Abraham–Freud conceptualization of depression have paid greater attention to the effect of such losses on the ego, de-emphasizing the id-libidinal and related hydraulic aspects. The depressant effect of separation events often resides in their symbolic meaning for a person rather than in any arbitrary objective weight that the event may have for clinical raters. However, love loss, bereavement, and other exits from the social scene, as defined by the London psychiatrist Eugene Paykel, are presently the concepts most commonly used in practice and research.  

Although love melancholy had been described since antiquity, the two affective states were systematically compared for the first time in Freud's 1917 paper on mourning and melancholia. According to current data, the transition from grief to pathological depression occurs in no more than 10 percent of adults and 20 percent of children. These figures suggest that such transition occurs largely in persons predisposed to mood disorders.

John Bowlby of the Tavistock Clinic, London, did a comprehensive clinical investigation of the attachment that the child establishes with the mother or mother substitutes during development, a bond considered the prototype for all subsequent bonds with other objects. Like many psychoanalytic explanations of adult symptom formation, the object loss model is formulated as a two-step hypothesis, consisting of early breaks in affectional bonds, which provide the behavioral predisposition to depression, and adult losses, which are said to revive the traumatic childhood loss and so precipitate depressive episodes. However, the role of proximate separations in provoking depressive reactions rests on more solid clinical evidence than the hypothesized sensitization resulting from developmental object loss. That realization led Bowlby to regard childhood sensitization resulting from early deprivation as a generic characterological vulnerability to a host of adult psychopathological conditions.

Compared with aggression turned inward, object loss is more directly relevant to clinical depression, yet it is still pertinent to question whether it is an etiological factor. Studies at the Wisconsin Primate Center indicated that optimal homeostasis with the environment is most readily achieved when the individual is securely attached to significant others, and the dissolution of such ties appears to be relevant to the emergence of a broad range of psychopathological disturbances rather than depression per se. A related methodological question is whether object loss operates independently of other etiological factors. For instance, a history of early breaks in attachment may reflect the fact that one or both of the patient's parents had mood disorder, with resultant separation, divorce, suicide, and so forth.

On balance, the ego-psychological object loss model is conceptually superior to its id-psychological counterpart. In postulating an open system of exchange between a person and the environment, the model permits consideration of etiological factors other than separation, such as heredity, character structure, and adequacy of social support—all of which might modulate the depressant effect of adult separation events. Conceptualizing the origin of depression along those lines is in the mainstream of current ideas of adaptation, homeostasis, and disease. An important treatment implication is the value of social support in preventing relapse and mitigating chronicity of depression. That is, indeed, an ingredient in the interpersonal psychotherapy of depression, which can be conceptualized as a form of brief, focused, and practical psychodynamic therapy.

Loss of Self-Esteem and Depression

Reformulation of the dynamics of depression in terms of the ego suffering a collapse of self-esteem represents a further conceptual break with the original id-psychological formulation; depression is said to originate from the ego's inability to give up unattainable goals and ideals. The model further posits that the narcissistic injury that crushes the depressed patient's self-esteem is imposed by the internalized values of the ego rather than the hydraulic pressure of retroflected thanatotic energy deriving from the id. Because the construct of the ego is rooted in social and cultural reality, loss of self-esteem may result from symbolic losses involving power, status, roles, identity, values, and purpose for existence. Thus, the existential and sociocultural implications of depression conceived as a derivative ego state provide the clinician with a far more flexible and pragmatic tool for understanding depressed persons than the archaic hydraulic metaphors related to libidinal vicissitudes. That model represents one of the first attempts to formulate depression in terms that subsequent psychological theory and research could operationalize in more testable form.

Self-esteem is part of the habitual core of the individual and, hence, is integral to the personality structure. Indeed, low self-esteem conceived as a trait is a major defining attribute of the depressive (melancholic) personality. Although it is understandable that such individuals can easily sink into melancholia in the face of environmental adversity, it is not obvious why persons with apparently high self-esteem (e.g., those with hypomanic and narcissistic personalities) also succumb to melancholy with relative ease. To explain such cases, one must invoke an underlying instability in the system of self-esteem that renders it vulnerable to depression. The opposite is also known to occur; that is, manic episodes may develop from a baseline of low self-esteem, as in the case of bipolar disorder patients with antecedent traits of shyness, insecurity, and dysthymia.

The foregoing considerations suggest that the vicissitudes of self-esteem deemed central to the model of depression as loss of self-esteem are manifestations of a more fundamental mood dysregulation. In classic psychoanalysis, such dysregulation is considered to be of constitutional origin. In general, attempts by psychoanalytic writers to account for bipolar oscillations have not progressed beyond metapsychological jargon, with the notable exception of denial of painful affects as a mechanism in the phenomenology of mania.

Cognitive Model

The cognitive model, developed by Aaron Beck at the University of Pennsylvania, hypothesizes that thinking along negative lines (e.g., thinking that one is helpless, unworthy, or useless) is the hallmark of clinical depression. In effect, depression is redefined in terms of a cognitive triad, according to which patients think of themselves as helpless, interpret most events unfavorably vis-à-vis the self, and believe the future to be hopeless. In more recent formulations in academic psychology, these cognitions are said to be characterized by a negative attributional style that is global, internal, and stable and that exists in the form of latent mental schemata that generate biased interpretations of life events.

Because the cognitive model is based on retrospective observations of already depressed persons, it is virtually impossible to prove that causal attributions such as negative mental schemata precede and, therefore, predispose to clinical depression; they can just as readily be regarded as subclinical manifestations of depression. The theoretical importance of the cognitive model lies in the conceptual bridge it provides between ego-psychological and behavioral models of depression. It has also led to a new and widely accepted system of psychotherapy that attempts to alter the negative attributional style, to alleviate the depressive state, and, ultimately, to fortify the patient against future lapses into negative thinking, despair, and depression.

The cognitive model, therefore, has the cardinal virtue of focusing on key reversible clinical dimensions of depressive illness, such as helplessness, hopelessness, and suicidal ideation, while providing a testable and practical psychotherapeutic approach. That approach, however, is less likely to succeed in patients with the full-blown melancholic manifestations of a depressive disorder. It is doubtful that negative cognitions alone could account for the profound disturbances in sleep, appetite, and autonomic and psychomotor functions encountered in melancholic depressions. Furthermore, conceptualizing a multifaceted malady such as depression largely or solely as a function of distorted cognitive processes is reminiscent of pre-Esquirolian notions that emphasized impaired reasoning in the development of depression. Finally, recent extensions or modifications, or both, of cognitive therapy in association with behavioral therapy (cognitive–behavioral therapy) for all emotional disorders (and even for schizophrenia) are reminiscent of earlier global claims of the psychodynamic perspective.

Learned Helplessness Model

The learned helplessness model is, in some ways, an experimental analogue of the cognitive model. The model proposes that the depressive posture is learned from past situations in which the person was unable to terminate undesirable contingencies. The model is based on experiments in dogs that were prevented from taking adaptive action to avoid unpleasant electrical shock and subsequently showed no motivation to escape such aversive stimuli, even when escape avenues were readily available. Armed with evidence from many such experiments, University of Pennsylvania psychologist Martin Seligman postulated a trait of learned helplessness (a belief that it is futile to initiate personal action to reverse aversive circumstances) formed from the accumulation of past episodes of uncontrollable helplessness.

The learned helplessness paradigm is a general one and refers to a broader mental disposition than depression. Thus, it is potentially useful in understanding such diverse conditions as social powerlessness, defeat in sporting events, and posttraumatic stress disorder (PTSD). In addition, past events might shape a characterological cluster—consisting of passivity, lack of hostility, and self-blame—relevant to certain depressive phenomena. The low hostility observed in some patients during clinical depression could, for instance, be ascribed to the operation of such factors. Learned helplessness could thereby provide plausible links between aspects of personal biography and clinical phenomenology in depressive disorders. Therapeutic predictions for alleviating depression and related psychopathological states capitalize on new cognitive strategies geared to modifying expectations of uncontrollability and the negative attributional style. This illustrates how insights gained from experimental paradigms can be combined fruitfully to address clinical disorders.

Nonetheless, the clinician should be wary of unwarranted clinical extrapolations. For example, some therapists have argued that the depressed patient's passivity is “manipulative,” serving to obtain interpersonal rewards. It has also been suggested that such factors have a formative influence on the development of the depressive character. That interpretation appears more relevant to selected aspects of depression than to the totality of the disorder. Depressive behavior and verbalizations clearly have a powerful interpersonal impact, but casting depression as merely a masochistic lifestyle developed to secure interpersonal advantages represents a mechanistic circular argument that could be viewed as disrespectful of the clinical agony of patients with mood disorders. Finally, although most formulations focusing on helplessness have emphasized acquisition through learning, recent experimental research in animals tends to implicate genetic factors in the vulnerability of learning to behave helplessly. The value of the helplessness paradigm may reside in its utility in predicting a variety of subthreshold affective disturbances generic to civilian reactions to adversity and trauma.

Depression and Reinforcement

Other behavioral investigators, including, notably, psychologist Peter Lewinsohn, have developed clinical formulations of depression that hinge on certain deficits in reinforcement mechanisms. According to the reinforcement model, depressive behavior is associated with lack of appropriate rewards and, more specifically, with receipt of noncontingent rewards. The model identifies several contributory mechanisms. Some environments may consistently deprive persons of rewarding opportunities, thereby placing them in a chronic state of boredom, pleasurelessness, and, ultimately, despair. That reasoning, however, may offer more insight into social misery than clinical depression. A more plausible postulated mechanism is the provision of rewards that are not in response to the recipient's actions; in other words, the gratis provision of what a person considers undeserved rewards may lead to lowering of self-esteem. Predisposition to depression is formulated in terms of inadequate social skills, which are hypothesized to decrease a person's chances of responding to potentially rewarding contingencies in the environment. Indeed, recent research on the relationship between personality and depression suggests that such deficits might underlie certain depressive states. Therefore, psychotherapeutic approaches designed to enlarge a patient's repertoire of social skills may prove valuable in preventing some types of depression.

The concepts of depression that have been derived from behavioral methodology and developed in the last several decades are scientifically articulate and, therefore, testable approaches to clinical depression. However, the important distinction between depression on self-report inventories and clinical depression tends to be overlooked in investigations testing the reinforcement paradigm. Furthermore, the behavioral model does not address the distinct possibility that reinforcement deficits may, in part, represent the psychomotor deficits of depressive illness. Nevertheless, by focusing on reward mechanisms, the behavioral model provides a conceptual bridge between purely psychological and emerging biological conceptualizations of depression.

Biogenic Amine Imbalance

The formulation of sophisticated biological explanations of mood disorders had to await the development of neurobiological techniques that could probe the parts of the brain involved in emotions. Although the complex physiology of the limbic-diencephalic centers of emotional behavior generally cannot be directly observed in humans, much has been learned from animal work. The limbic cortex is linked with both the neocortex, which subserves higher symbolic functions, and the midbrain and lower brain centers, which are involved in autonomic control, hormonal production, and sleep and wakefulness. Norepinephrine-containing neurons are involved in many functions that are profoundly disturbed in melancholia, including mood, arousal, appetite, reward, and drives. Other biogenic amine neurotransmitters that mediate such functions are the catecholamine dopamine—especially important for drive, pleasure, sex, and psychomotor activity—and the indoleamine serotonin, which is involved in the regulatory control of affects, aggression, sleep, and appetite. Cholinergic neurons, secreting acetylcholine at their dendritic terminals, are generally antagonistic in function to catecholaminergic neurons.

Although the opioid system might, on experimental and theoretical grounds, also serve as one of the neurochemical substrates for mood regulation, no cogent model of mood disorders involving that system has appeared. Likewise, biochemical formulations of mood disorders have paid relatively little attention to the major excitatory brain neurotransmitter glutamate and the inhibitory neurotransmitter γ-aminobutyric acid (GABA).

Biogenic Amine Hypotheses

Joseph Schildkraut at Harvard University and William Bunney and John Davis at NIMH published the first formal hypothesis connecting depletion or imbalance of biogenic amines (specifically norepinephrine) and clinical depression. The serotonin counterpart of the model was emphasized in the models proposed by Alec Coppen in England and I. P. Lapin and G. F. Oxenkrug in Russia. Both catecholamine and indoleamine hypotheses were essentially based on two sets of pharmacological observations. First, reserpine (Serpasil), which decreases blood pressure P.1639 by depleting biogenic amine stores, precipitates clinical depression in some patients. Second, antidepressant medications, which alleviate clinical depression, raise the functional capacity of the biogenic amines in the brain. This style of thinking is known as the pharmacological bridge, extrapolating from evidence on the mechanism of drug action to the neurotransmitter pathologies presumed to underlie a given psychiatric disorder. Such pharmacological strategies have been of heuristic value in developing research methods for the investigation of mood disorders and schizophrenia. Indeed, the research methodology developed by the relatively few investigators working in this area during the last half century is among the most elegant in the history of psychiatry.

Variations of the biogenic amine model assign somewhat different relative weights to the biogenic amines norepinephrine and serotonin in the development of pathological mood states. Arthur Prange and colleagues at the University of North Carolina formulated a permissive biogenic amine hypothesis in which serotonin deficits permit the expression of catecholamine-mediated depressive or manic states. That hypothesis was supported by subsequent animal research showing that an intact serotonin system is necessary for optimal functioning of noradrenergic neurons. Omission of tryptophan from the diet of antidepressant-responsive depressed patients may annul the efficacy of the antidepressant; among healthy volunteers, that special diet also induces sleep electroencephalographic characteristics of clinical depression. Although such findings are provocative, the precursor-loading strategy to increase the brain stores of serotonin (e.g., with L-tryptophan) has not been unequivocally successful in reversing clinical depression. Dietary loading with catecholamine precursors has fared even worse than serotonin-precursor loading in the treatment of depression.

The cholinergic–noradrenergic imbalance hypothesis proposed by David Janowsky and colleagues represents yet another attempt to elucidate the roles of biogenic amines. This hypothesis, along with the related cholinergic supersensitivity hypothesis developed by J. Christian Gillin, has been tested extensively at the University of California at San Diego. Subsequent formulations by Larry Siever and Kenneth Davis at the Mount Sinai Hospital in New York have refocused on noradrenergic dysregulation. The model assumes oscillation from one output mode to the other at different phases of depressive illness. In a provocative extrapolation from that model, bipolar depression would have low noradrenergic output, but many instances of major depressive disorder, as with some anxiety disorders, could be biochemically conceptualized as high-output conditions.

Despite more than four decades of extensive research and indirect evidence, however, no deficiency or excess of biogenic amines in specific brain structures has been shown to be necessary or sufficient for the occurrence of mood disorders. It has not been possible either to confirm the putative role of central norepinephrine in depression or to discard it altogether. The role of dopamine as formulated, among others, by the Italian pharmacologist Gian Luigi Cessa, although studied less extensively than that of norepinephrine, deserves greater recognition, in that it might have relevance to atypical and bipolar depression, as well as to mania.

Preliminary data from a small brain imaging study has shown blunted serotonin responsivity in prefrontal and temporoparietal areas in unmedicated patients with major depressive disorder. Such data, considered in the context of the overall serotonin literature in depression, is provocative but not conclusive and serves to illustrate the fact that the case for serotonergic disturbance in depression continues to be based on indirect evidence. Moreover, the putative permissive role of serotonin is better documented for aggressive suicide attempts. Serotonergic dysfunction might subserve other conditions characterized by lack of inhibitory control, among them obsessive-compulsive disorder (OCD), panic disorders, bulimia nervosa, certain forms of insomnia, alcoholism (alcohol abuse or dependence), and a host of impulse-ridden personality disorders. Such considerations have led Dutch psychiatrist Herman van Praag and colleagues to postulate a dimensional neurochemical disturbance generic to a large group of disorders within the traditional nosology. This hypothesis might be variously regarded as a challenge to psychiatric nosology or as a statement of the need to supplement clinical classification with biochemical parameters. Both interpretations are in line with clinical observations during the last two decades testifying to the high prevalence of comorbidity in depressive, other emotional disorders, and certain impulse-control disorders.

It is implied that the foregoing postulated biochemical faults are genetically determined. Although biogenic amine models of mood disorders were developed retrospectively from the pharmacological action of antidepressant and thymoleptic agents, they have stimulated the development of new classes of antidepressants with more selective action on specific neurotransmitter receptors. Their introduction has virtually revolutionized the treatment of depression. Yet the fundamental biochemistry of mood disorders is still far from being understood. Curiously, although selective in action, the new compounds working on the serotonin system have broad effectiveness in a variety of mood-related conditions, such as dysthymic disorder, PTSD, OCD, panic disorder, social phobia, bulimia nervosa, and borderline personality disorder. Such data indirectly favor the hypothesis of an underlying biological commonality to several of these disorders. The foregoing considerations have, in turn, led to a provocative formulation of an increasingly prevalent “social syndrome” in populations experiencing social disruption, immigration, and abuse and characterized by anxiety, depression, violence-proneness, impulsivity, and suicidality—reflecting a perturbed serotonin system, the oldest, most basic brain structure involved in human socialization and territoriality, coping with stress, danger, and survival.

New antidepressants with dual action on both serotonergic and noradrenergic receptors and emerging data on their possible greater efficacy in melancholic depressions suggest that the biochemistry of mood disorders involves more complex dysregulation than is implied in single-neurotransmitter hypotheses. The work of George Henninger and colleagues at Yale University further suggests that monoamines better explain how antidepressants facilitate recovery from depression than their being the fundamental causes of depression. Moreover, emerging biochemical paradigms are moving away from distal biochemical lesions to focus on molecular perturbations closest to the putative genetic underpinnings of mood disorders. Originally tied to the mechanism of action of mood stabilizers in bipolar disorder, such work is exploring second messenger systems, phosphorylation G proteins, signal transduction, DNA transcription, and messenger RNA translation. Again, such search for molecular mechanisms represents “backward logic” from the putative mechanism of action of selected thymoleptic agents. The same can be said about Frederick Petty's GABAergic and Shih-Jen Tsai's brain-derived neurotrophic factor (BDNF) hypothesis on the origin of bipolar disorder.

Neuroendocrine Links

Functionally inadequate mobilization of neurotransmitters in the face of continued or repeated stress, as indirectly reflected in pathological modification of noradrenergic and serotonergic receptor function, could represent neurochemical final common pathways of homeostatic failure. Such mechanisms could also provide links with psychoendocrine dysfunction; the hypothesized neurotransmitter deficits may underlie the disinhibition of the hypothalamic–pituitary–adrenal axis, characterized by steroidal overproduction, the most widely studied endocrine disturbance in depressive illness.

When challenged with dexamethasone (Decadron), the altered axis resists suppression, which offered Bernard Carroll's team (then at the University of Michigan) the possibility of developing the dexamethasone-suppression “test” (DST) for melancholia. This procedure is of uncertain specificity for depressive illness and, thus, is unsuitable to serve as a diagnostic test. However, that line of research has been useful in pathogenetic understanding. For instance, it led to the demonstration by Emory University's Charles Nemeroff of increased concentrations of corticotropin-releasing factor (CRF) in the cerebrospinal fluid (CSF) of patients with major depressive disorder. CRF also appears relevant to the pathophysiology of anxiety disorders, such as panic disorder, and PTSD.

The research of Florian Holsboer's group at Munich's Max Planck Institute has shown impaired glucocorticoid and mineralocorticoid receptor function in these disorders, with relevant pathophysiological and therapeutic implications.

Another neuroendocrine index of noradrenergic dysregulation, blunted growth hormone response to the α2-adrenergic receptor agonist clonidine (Catapres), likewise points to limbic-diencephalic disturbance. However, studies performed in the United States suggest that it is positive in both endogenous depression and severe anxiety disorder (panic disorder). Thyroid-stimulating hormone (TSH) blunting on thyrotropin stimulation, another common neuroendocrine disturbance in depression, also shows limited specificity.

What is remarkable, however, is that the DST, clonidine, and thyrotropin challenge data, in aggregate, identify most persons with clinical depression. Such evidence of midbrain disturbance argues for considering clinical depression to be a legitimate disease. The disease concept of depression is further buttressed by computed tomography (CT) scans showing enlarged pituitary and adrenal glands, a state marker of depressive illness.

Stress and Depression

The concept of a pharmacological bridge implies two-way traffic. The hypothesized chemical aberrations may be primary or genetically based. Provision should also be made, however, for the likelihood that psychological events that precipitate clinical depression might initiate or exacerbate neurochemical imbalance in vulnerable subjects. That suggestion is supported by studies in animals in which early separation from peers and inescapable frustration effect profound alterations in the turnover of biogenic amines and in postsynaptic receptor sensitivity. Thus, in genetically predisposed persons, environmental stressors might more easily lead to perturbations of limbic-diencephalic neurotransmitter balance. Finally, in vulnerable individuals, especially during the formative years of childhood, psychological mechanisms might more easily perturb midbrain neurochemistry. Traumatic experiences appear particularly potent in this regard.

The hippocampus has been the subject of intense recent research as the possible neuro-anatomical substrate linking such loss and trauma to adult depression. Ongoing ingenious experimental paradigms in primates and rodents continue to explore the role of early experience and stress in subsequent depressive-like behaviors in these animals. In humans, a new provocative finding indicates that a polymorphism of the serotonin transporter gene would identify who among traumatized children would develop adult depression. Likewise, a polymorphism of the monoamine oxidase (MAO), A gene plays a significant role in determining who among battered children will grow into an adult sociopath. Animal models of mania are sparse and problematic.

Neurophysiological Approaches

Neuronal Hyperexcitability

Lithium is known to replace intracellular sodium and hyperpolarize the neuronal membrane, thereby decreasing neuronal excitability. Abnormalities in neuronal electrolyte balance (an excess of residual sodium, defined by radioisotope techniques) and hypothesized secondary neurophysiological disturbances were the focus of British investigations by Alec Coppen and colleagues in the early 1960s. The data appear compatible with the hypothesized movement of excess sodium into the neuron during an episode of mood disorder and redistribution toward the pre-illness electrolyte balance across the neuronal membrane during recovery.

Intraneuronal sodium leakage is postulated in both depressive and manic disorders but is deemed more extreme in the latter. Because the harmonious activity of the neuronal cell and, by implication, that of a group of neurons depends on the electrical gradient maintained across its membrane by differential distribution of sodium, abnormalities in sodium concentrations and transport are hypothetically relevant to the production of an unstable state of neurophysiological hyperexcitability. In formulating their thesis of neurophysiological arousal in melancholic states, Joseph Mendels and Peter Whybrow (both of whom have worked at the University of Pennsylvania) have capitalized on the foregoing electrolyte disturbances.

The view that mania represents a more extreme electrophysiological dysfunction in the same direction as depression violates the commonsense notion of symptomatological “opposition” between the two kinds of disorder, yet it may, in part, account for the existence of mixed states in which symptoms of depression and mania coexist. The NIMH team led by Frederick Goodwin first showed that a substantial minority of depressed patients with a bipolar substrate respond to lithium salts, which further supports the concept of a neurophysiological common denominator to mania and depression.

Perturbations of calcium metabolism also appear relevant to bipolar patients. Therapeutic implications of this observation (e.g., the use of calcium channel inhibitors in bipolar I disorder) have not yielded consistent results. Finally, rubidium, another alkali metal, has been explored in the depressive phase of bipolar disorders, again with inconclusive results.


European studies have shown that depressed patients are phase advanced in many biological rhythms, including the latency to the first rapid eye movement (REM) in sleep. Shortened REM latency, which has been extensively studied by David Kupfer and colleagues at the University of Pittsburgh, has been proposed as another laboratory “test” for depressive disorder. Shortened REM latency may serve as a trait marker for depression because it has been found in dysthymia and so-called borderline personality, as well as among the clinically “well” offspring of adults with major depression.

Formulations of circadian rhythms by Thomas Wehr and Norman Rosenthal, working at NIMH, have focused on abnormalities on brain regulation of temperature, activity, and sleep cycles. Others have investigated the role of the pineal hormone melatonin in mood disorders, without achieving consistent results. The application of circadian rhythm research concepts to women with mood disorders has also led to imaginative methods, but, again, without definitive characterization of the neurophysiologic faults. 

At a basic level, instrumental REM sleep deprivation in neonates has been recently shown to lead to adult “depression-like” behaviors in rats. In human studies, sleep deprivation and exposure to bright white light has been shown to correct phase disturbances and thereby terminate depressive episodes, especially in subjects with periodic and seasonal depressions.

It has even been shown that the average person is light deprived, and that phototherapy can benefit even those without clear-cut seasonal patterns. Unfortunately, except for their use in mild seasonal depressions and suppression of sleep deprivation to prevent hypomania in bipolar disorder, the foregoing circadian studies have not had a palpable impact on practice.

Their application in the large segment of mood-disordered patients remains cumbersome, if not elusive. Their effect can be better assessed at the level of theory. Although the specificity and efficacy of these neurophysiological indices and manipulations for clinical depression and bipolar disorder require more extensive research, cumulatively, they point to midbrain dysregulation as the likely common neurophysiological substrate of affective disorders.

The foregoing considerations further suggest that the ancient Greeks, who ascribed melancholia to malignant geophysical influences, did not indulge in mere poetic metaphor. The ancients had observed the disturbed circadian patterns and advocated their readjustment to restore euthymia.

Affective Dysregulation

A major challenge for research in mood disorders is to characterize the basic molecular mechanisms that underlie the neurophysiological rhythmopathies, which, in turn, might account for the recurrent nature of the affective pathology as envisioned by Kraepelin. This means that in the most typical recurrent forms of the disorders, the constitutional foundations (manifested as cyclothymic and dysthymic traits and/or a broad range of emotional disequilibrium covered by the rubric of “neuroticism”) are so unstable that the illness may run its entire course more or less autonomously, with the environment largely serving to turn on and off the more florid phases (episodes).

Parisian psychiatrist Jean Delay, a pioneer in psychopharmacology in the 1950s, also emphasized affective dysregulation as the fundamental pathology in the spectrum of mood disorders.

Robert Post, at NIMH, hypothesized that the electrophysiological substrates could be so kindled that an oligoepisodic disorder initially triggered by environmental stressors could assume an autonomous and polyepisodic course. He hypothesized that this phenomenon might occur because neuronal perturbations brought about by stressors in the early course of mood disorders get incorporated into the DNA. This fascinating kindling hypothesis, however, does not seem to pertain to garden-variety mood disorders but to those with extreme cyclicity. The monograph on manic-depressive illness by Fred Goodwin and Kay Jamison presents in-depth arguments for this cyclical paradigm of thymopathy.

The amygdala, through its connection with the serotonin transporter, appears to play a crucial role in the dysregulation of the emotional life of affectively ill patients. Data indicate that the dysregulation is both in function and anatomic structure.

Theoretical Synthesis

Pathophysiological Understanding

Modern psychobiology attempts to link experience and behavior to the central nervous system (CNS). Building conceptual bridges between the psychological and biological approaches to mood disorders requires sophisticated strategies that go beyond the Cartesian notion of limited mind–body interactions through the pineal gland and the generalizations of the Meyerian school.

William McKinney and Hagop Akiskal in 1973 developed a conceptual framework that considers the affective syndromes as the final common pathway of various psychological and biological processes. The overarching hypothesis is that psychological and biological etiological factors converge in reversible deficits in the diencephalic substrates of pleasure and reward. Those areas of the brain subserve the functions that are disturbed in melancholia and mania.

The integrative model links the central chemistry and physiology of reward mechanisms with the object loss and behavioral models of depression, both of which give singular importance to the depressant role of loss of rewarding interpersonal bonds. A key element of the model is the circadian disturbances observed since ancient times in both depressive and manic syndromes. Both syndromes are conceptualized as clinical manifestations of a disordered limbic system with its subcortical and prefrontal extensions.

Brain imaging studies in melancholic patients by Wayne Drevets, originally at Washington University, have visualized limbic disturbances extending into subcortical structures and occurring primarily in patients with a familial diathesis for depression; the amygdala appears to be the focal limbic structure in the latter studies. Clinical experience and research data suggest that multiple factors described in the following subsections converge to produce or exacerbate dysregulation in these brain regions, leading to the final common pathway of clinical depression. The data on mania are more tentative and are mentioned when relevant.


Evidence indicates a significant genetic role in the causation of bipolar and recurrent major depressive disorders. Although it is not known exactly what is inherited and biological endophenotypes have not been delineated, temperamental dysregulation might be hypothesized to fulfill the role of a behavioral endophenotype. The depressive inheritance might translate into impaired coping under stress (neuroticism), and bipolar inheritance might translate into affective dysregulation (cyclothymia), involving over- and underreactivity to life situations, circadian events, and biological stressors.

Whatever the precise nature of the inherited fault or excess, current research suggests that heritability involves a broad spectrum of disorders, including milder affective states, as well as temperamental inclinations. Recent findings, from both clinical and genetic investigations, have emphasized the importance of broad affective phenotypes that incorporate panic and anxiety reactivity within both traditional unipolar and bipolar disorders. For instance, the affective dysregulation underlying bipolar disorder can manifest in euphoria, irritability, depression, panic attacks, and social anxiety. Such tendencies are observed among patients and their first-degree relatives.

Genetic heterogeneity is likely and may involve inheritance of a single dominant gene with variable penetrance in some families or specific subtypes, or oligogenic inheritance in the majority of cases. Different genetic mechanisms will, in all likelihood, involve more than one disorder (e.g., depression and generalized anxiety; bipolar I disorder, psychosis, stimulant abuse, and dipsomania; bipolar II disorder, panic disorder, and bulimia nervosa).

Another distinct possibility is that some forms of schizophrenia, bipolar I and II disorders, and recurrent depressions lie on an oligogenic bipolar spectrum. A polymorphism involving the short alleles of the serotonin transporter gene appears relevant to depressive dispositions. The genetic mechanism involving mood reactivity and panic appears to be subserved by chromosome 18q. The partial overlap of manic and schizophrenic phenomenology might be related to a polymorphism related to a variety of genes.

It is relevant to point out at this juncture that at least five genetic loci have been identified with shared liability for both bipolar and schizophrenic disorders (see Section 13.3 by Kelsoe on the genetics of mood disorders): These are the loci for brain-derived neurotrophic factor (BDNF), catechol-O-methyl transferase (COMT), neuroregulin (NRGl), D-amino acid oxidate activator (DAOA; also known as G72), and DISC1 (Disrupted in Schizophrenia 1).

It is unlikely that these findings coming from laboratories worldwide represent systematic diagnostic error. One heuristic favored theoretical interpretation of these data is that bipolar and schizophrenic disorders represent oligogenic inheritance, requiring multiple additive genes. A competing hypothesis is that there exist distinct mood disorder phenotypic spectra, each with their own genotypes, only some of which cross the border into psychosis; others cross the border with anxiety and panic; still others may involve “comorbidity” with migraine.

Developmental Predisposition

Parents with mood disorders are often in conflict, which may lead to separation, divorce, and suicide. It can be said that heredity often determines the type of environment into which the child predisposed to mood disorder is born. Developmental object loss, although not specifically involved in causing mood disorder, might modify the expression of the illness, possibly by leading to earlier onset, more severe episodes, and an increased likelihood of personality disorder and suicide attempts. The serotonin transporter polymorphism mentioned earlier appears to mediate the relationship between early trauma and depression. Likewise, this polymorphism appears relevant to neuroticism and suicide attempts.


Since ancient times, persons prone to mania and melancholia have been described as possessing certain temperamental attributes, representing variations on the theme of what today is subsumed under cyclothymic, dysthymic, and anxious-inhibited temperaments, as well as the traits of high neuroticism describing emotionality.

Many monozygotic twins discordant for full-blown mood disorders, studied by Aksel Bertelsen's Danish research team, exhibited affective instability with temperamental moodiness, which strongly suggests that such attributes are genetically determined.

Research conducted by Kendler's team at the Medical College of Virginia further suggests that several of the temperamental attributes might be transmitted as part of the genetic liability to mood disorders. Research has also identified such temperaments in the prepubertal offspring of parents with bipolar I disorders, suggesting that they precede by years to decades the overt onset of major mood disorder episodes.

The atmosphere of high expressed emotion and the negative critical remarks by relatives and affectively unstable patients documented in the recent psychological literature on mood disorders often reflect the interpersonal clashes between patients and their temperamentally intense relatives. Thus, temperaments appear to be intimately involved in generating much interpersonal friction, emotional arousal, and sleep loss (just to cite common perturbations), thereby eliciting many of the life stressors that precipitate affective episodes.

The use of stimulant drugs either to self-treat lethargy or to enhance hypomanic traits could further contribute to episode precipitation. As for the depressive disposition, the work of Maria Kovacs at the University of Pittsburgh has shown that dysthymia in children evolves into major depressive episodes postpubertally, of which a proportion switch to bipolar states. These data cohere with the work conducted at the University of Tennessee, Memphis, showing shortened REM latency in early-onset dysthymic subjects. The familial bipolar diathesis revealed in the Tennessee work, along with its tendency to switch to hypomania, suggests commonalities between depressive and bipolar II disorders.

Life Events

Most individuals do not develop clinical depression when exposed to environmental adversity. Such adversity seems to play a pathogenic role primarily in those with an affective diathesis. In fact, the work of Kendler at the Medical College of Virginia indicates that genetic factors might underlie the depressive disorder patients' susceptibility to life events. Furthermore, current data suggest that social stressors in the onset of depression are more relevant to the first few episodes of the illness.

The evidence linking such events to mania is less convincing. At any rate, stressful events often appear to be triggered by the temperamental instability that precedes clinical episodes. Interpersonal losses are common events in the lives of individuals with intense temperaments. The arousal and sleep loss associated with such events can precipitate both depressive and manic states.

A recent study by Peter McGuffin's team at the Institute of Psychiatry, London, raised the possibility that one mechanism by which heredity produces depression is the creation of environmental adversities in the lives of individuals predisposed to this illness. This work has been replicated by independent groups of investigators. Whatever the origin of environmental adversity, it is common clinical experience that loss represents an important, perhaps even central, theme in clinical depression.

Variables that seem to modulate the effect of adult losses include concurrent life events, resultant changes in lifestyle, lack of interpersonal support, deficient social skills, and the symbolic meaning of the putative loss.

The research program of George Brown and his followers in London capitalizes on the foregoing considerations, particularly the importance of early and proximate losses in socioeconomically disadvantaged women who lack supportive relationships. However, that conceptualization downplays the degree to which the social context of the depression reflects the dysthymic temperamental liabilities of those depressed women.

Recent research indicates that even social support is determined to a considerable degree by the genetic mechanisms that underlie mood disorders. Indeed, the short alleles of the serotonin transporter gene are now implicated in mediating between adverse events and clinical depression.

Biological Stressors

Many physical diseases and pharmacological agents are known to precede the onset of both depressive and manic episodes. Like psychosocial stressors, however, they do not generally seem to cause de novo episodes, but mobilize them in persons with a personal and family history of mood disorders. Thyroid disturbances have a role in practice because they are associated with rapid cycling in bipolar patients, especially women; lithium is often contributory to such disturbances occurring in the depressions in which bipolar women are not uncommonly “stuck.”


Clinical and epidemiological studies concur in suggesting that women are at higher risk for mood disorders, with the risk highest for depression. This now appears to be, in part, a function of anxious-depressive traits represented by neuroticism. These traits have strong genetic determinants. Women have higher concentrations of monoamine oxidase (the enzyme that breaks down monoamine transmitters) in the brain and more precarious thyroid status. In addition, low estrogen and high progesterone concentrations have been postulated as possible mediating factors in postpartum depressions, premenstrual accentuation of affective instability, and women's vulnerability to the depressant effect of steroidal contraceptives. Finally, recent data point to the role of estradiol in depressions occurring during the transition to menopause.

Personality factors might also be relevant to the sex differences in depression. In recent collaborative work with University of Pisa psychiatrist Giulio Perugi, Akiskal proposed the hypothesis that female sex might favor greater expression of dysthymic attributes, whereas hyperthymic traits appear to be favored by male sex.

Those considerations tend to parallel, respectively, the ruminative and active cognitive response styles reported by Susan Nolen-Hoeksema, originally at Stanford University, to distinguish the sexes. What specific sex-related biographical factors might interact with sex-related biological factors to produce such trait differences is largely unknown.

An intriguing possibility is that women, because of temperamental inclination to depressive cognitions, might react more intensely to childhood adversities, as well as be more specifically vulnerable to adult stressors related to bonding with men and child rearing.

Research by Mark George and colleagues has raised the provocative possibility that women overrespond to sad circumstances over a lifetime, thereby permanently altering anterior limbic and prefrontal brain function in a “depressive” direction.

The high prevalence of anxious-depressive conditions in women—most pronounced at the clinical level—might be linked, hypothetically, to an evolutionary adaptive advantage conferred by traits of fear, inhibition, and avoidance to women who bear the responsibility of pregnancy and child rearing.

The higher prevalence of minor depressions in women, rediscovered in a recent Danish study, is in line with the foregoing hypothesis.

Figure 13.1-3. An integrative pathogenetic model of mood disorders. The integrative model presented here (Fig. 13.1-3) goes beyond the general provisions of the unified approach developed three decades ago. It is submitted that, at least in the highly recurrent forms of the malady, affective temperaments represent the intermediary stage between remote (hereditary) and proximate (stressful) factors and that limbic-diencephalic dysfunction is best characterized as the biological concomitant of the clinical manifestations of the affective syndromes. Like the temperamental dysregulations, these biological disturbances represent a putative stage in the pathogenetic chain. They emerge as temperamental instabilities that react to, provoke, or invite life events, substance use, and alterations in circadian rhythms—which, in turn, appear to usher in the behavioral, emotional, and cognitive manifestations of the illness. It is finally relevant to point out that biological stressors such as hormonal disturbances and traumatic brain injury appear to compromise limbic-diencephalic function as their depressant mechanism.

Therapeutic Perspectives

The foregoing integrative model mandates the joint use of somatic-pharmacological and psychosocial interventions. Although the milder forms of mood disorders can be managed with psychotherapy, somatic treatments are usually required to reverse the biological disturbances in melancholia before the patient can respond to interpersonal feedback. Depressive disorders with psychotic features often necessitate more definitive somatic interventions, such as electroconvulsive therapy (ECT). Continued psychopharmacological treatment is also effective in decreasing rates of relapse and future recurrence in most. Bipolar disorder and most forms of depression are considered lifelong illnesses needing indefinite maintenance pharmacotherapy.

Psychosocial therapy by skilled clinicians can provide support, combat demoralization, change maladaptive self-attributions, and improve conjugal and vocational functioning. Recent fascinating data support the notion that psychosocial interventions such as cognitive–behavioral therapy modulate cortical-limbic function described earlier as the final common pathway of depressive illness. Whether such therapy can also modify personality traits to fortify the patient against new episodes is a future research challenge. It might prove more profitable to attempt to help patients to explore professional and object choices that match their temperamental proclivities and assets, which, in turn, might provide them greater harmony and adaptation in life.

Although much needs to be learned about the indications for medication and psychotherapy in different subtypes of mood disorders, research not only does not support a negative interaction between the two forms of treatment, but, on selected parameters, suggests additive and even synergistic interaction. There is a great need for patients, their families, and clinicians to understand how a biologically driven illness like depression should be approached from a pragmatic psychotherapeutic perspective.

The challenge for psychiatric research in the decade ahead is to elucidate the basic mechanisms by which the predisposing, precipitating, and mediating variables reviewed here and others yet to be identified interact to produce the final common path of decompensation in melancholia. 

Because of the heterogeneity of depressive conditions presenting as a psychobiological final common clinical syndrome and because antidepressant agents, irrespective of specificity to one or another biogenic amine, are approximately equally effective in two thirds of those with depressive disorders, the antidepressant agents may be acting not on the primary lesions of these disorders but on a neurochemical substrate distal to the underlying biological faults.

Choice of antidepressants is still highly determined by the side effect profile least objectionable to a given patient's physical status, temperament, and lifestyle. That so many different classes of antidepressants—with different mechanisms of action—have been marketed since the 1990s represents indirect evidence for heterogeneity of putative biochemical lesions.

The investigation of central neurotransmitter receptor function continues to occupy much current effort to delineate the mechanism of antidepressant action and side effects of classic agents, as well as the new compounds that have made the treatment of depression “clinician and patient friendly.” Whether study of specific receptors will unravel the molecular mystery of depression remains to be seen. Since the 1990s, studies have begun on antidepressant and mood-stabilizing effects on molecular mechanisms believed to be closer to the “genetic underpinnings” of mood disorders.

Herein is the promise of the future—a new generation of psychiatrists conversant with both clinical phenomenology and molecular biology. Data suggest that the biological specificity of genetic factors in mood disorders might be translated into distinct temperamental dysregulations, which, in turn, might predispose to different affective subtypes.

Returning to the therapeutic arena, mounting clinical evidence indicates that, in a special subgroup of depressed patients with bipolar disorder, antidepressants might provoke mixed episodes, hypomanic episodes, or both, and possibly increase later cycling.

The kindling-sensitization model suggests the utility of anticonvulsant medication on episode escalation and might represent yet another example of pathophysiological intervention. Whatever the merit of this model, the 1990s witnessed intense clinical and research interest and U.S. Food and Drug Administration (FDA) approval of the clinical introduction of divalproex (Depakote) and lamotrigine (Lamictal) for bipolar disorder, and many other promising anticonvulsants are being developed for that disorder.

Anticonvulsant mood stabilizers appear to possess a broad spectrum of activity on bipolar disorders, including mixed dysphoric and rapid-cycling forms. Lithium, by contrast, seems more specific to euphoric or “classic” mania. On the other hand, the introduction of atypical antipsychotics—olanzapine (Zyprexa), risperidone (Risperdal), quetiapine (Seroquel), and aripiprazole (Abilify)—for mania raises intriguing questions about a common neural substrate for schizophrenic and manic disorders.

Psychoeducational interventions geared to disturbed rhythms of the disorder represent another example of rational therapeutics. Mood clinics should help patients and their significant others to dampen stimulation so that it is kept at an optimal level for depressed patients with cyclothymic traits. All offending drugs (e.g., cocaine, caffeine, and sedative-hypnotic agents) should be gradually eliminated and circadian disruptions and sleep loss minimized.

The greater challenge is to learn how to curb the ill-advised actions of patients with cyclical depressions. Unfortunately, current data indicate neurocognitive deficits, particularly in executive function, that underlie and/or complicate these disorders. Psychoeducation and psychotherapy have the task of ameliorating the resulting social problems, and despite much progress in this realm, these deficits remain stable over time. Compliance with mood-stabilizer regimens that, for many, would attenuate episodes and prevent such sequelae is difficult to achieve. Further research on treatment or medication-adherence techniques is needed for promoting more efficient use of mood stabilizers.

It is tempting to suggest that biogenic amines, the “humors” of modern psychobiology, play the same heuristic role as the ancient humors did for many centuries. The black humor, appropriately evoked in the construct of melancholia in the DSM-IV-TR, may not have the same claim for etiological relevance to depressive disorders as norepinephrine and serotonin, but at least it has a classic heritage. Dopamine, by contrast, may represent the sanguine humor that drives hypomanic temperaments and manic behavior.

When genetic factors contributing to clinical depression and mania are discovered, in all likelihood, they will be more linked to temperamental dispositions than to full-blown affective disease phenotypes. The clinician will still need to interpret the myriad influences that effect such inclinations to produce disease in an individual patient—that is, fundamental scientific advances in mood disorders, rather than diminishing the role of practitioners, will actually increase it. It is to be regretted that, despite destigmatization efforts and a rich armamentarium of therapeutic developments, the clinical care of affectively ill patients at the severe end of the spectrum continues to be grossly inadequate. More research will not improve this dismal situation unless the human and social dimensions of severe mood disorders are addressed with the requisite clinical and public health policies.

Caring for patients with mental illness is a dimension distinct from evidence-based treatments along psychopharmacological and psychotherapeutic lines: It requires the allocation of human resources and integrated mental health structures geared to the total patient.

In any discipline, scientific truth is a function of its technology, but understanding the phenomena under consideration is a matter of philosophical temperament that seeks integration and the hope for a unified vision. Research into the causes and treatment of mood disorders has generated abundant recent data suitable for integration into theory and practice, and conceptualizing the origin and treatment of mood disorders can no longer be justified on the grounds of ideological preference alone.

Suggested Cross-References Sections 13.2, 13.3, 13.4, 13.5, 13.6, 13.7, 13.8, 13.9 cover the various aspects of mood disorders in great detail.


Abraham K: Notes on the psychoanalytical investigation and treatment of manic-depressive insanity and allied conditions. In: Selected Papers of Karl Abraham. London: Hogarth Press; 1948.

Adams F, ed: Aretaeus of Cappadocia: The Extant Works of Aretaeus, the Cappadocian. London: Sydenham Society; 1856.

Ahrens B, Muller-Oerlinghausen M, Schou M, Wolf T, Alda E: Excess cardiovascular and suicide mortality of affective disorders may be reduced by lithium prophylaxis. J Affect Disord. 1995;33:67.

Akiskal HS: Dysthymia and cyclothymia in psychiatric practice a century after Kraepelin. J Affect Disord. 2001;62:17.

Akiskal HS, Akiskal KK: In search of Aristotle: Temperament, human nature, melancholia, creativity and eminence. J Affect Disord 2008;100:1. 

Akiskal HS, McKinney WT: Depressive disorders: Toward a unified hypothesis. Science. 1973;182:20.

Akiskal HS, Yerevanian BI, Davis GC, King D, Lemmi H: The nosologic status of borderline personality: Clinical and polysomnographic study. Am J Psychiatry. 1985;142:192.

Berger M, Vollmann J, Hohagen F, Konig A, Lohner H: Sleep deprivation combined with consecutive sleep phase advance as a fast-acting therapy in depression: An open pilot trial in medicated and unmedicated patients. Am J Psychiatry. 1998;155:1134.

Berretta S, Pantazopoulos H, Lange N: Neuron numbers and volume of the amygdala in subjects diagnosed with bipolar disorder or schizophrenia. Biol Psychiatry 2007;62:884. 

Bertelsen A, Harvald B, Hauge M: A Danish twin study of manic-depressive disorders. Br J Psychiatry 1997;130:330.

Biederman J, Mick E, Faraone SV, Spencer T, Wilens TE: Pediatric mania: A developmental subtype of bipolar disorder? Biol Psychiatry. 2000;48:458.

Brandon AR, Shivakumar G, Freeman MP: Depression: Covering mood and vasomotor symptoms. Curr Psychiatry. 2008;7:39.

Carlson GA, Cantwell DP: Unmasking masked depression in children and adolescents. Am J Psychiatry. 1980;137:445.

Carroll BJ, Feinberg M, Greden IF, Tarika J, Albala AA: A specific laboratory test for the diagnosis of melancholia: Standardization, validation, and clinical utility. Arch Gen Psychiatry. 1981;38:15. 

Caspi A, Sugden K, Moffit TE, Taylor A, Craig IW: Influence of life stress on depression: Moderation by a polymorphism in the 5-HTT gene. Science. 2003;301:386.

Coppen A: The biochemistry of affective disorders. Br J Psychiatry. 1967;113:1237.

Cytryn L, McKnew D: Growing Up Sad: Childhood Depression and Its Treatment. New York: WW Norton; 1996.

Davidson RJ, Scherer KR, Goldsmith HH, eds: Handbook of Affective Sciences. Oxford: Oxford University Press; 2003.

Delay J: Les Dérèglements de l'Humeur. Paris: Presses Universitaires de France; 1946.

Depression Guideline Panel: Depression in Primary Care (Clinical Practice Guideline, Number 5). Rockville, MD: U.S. Department of Health and Human Services, Public Health Service, Agency for Health Care Policy and Research; 1993.

Frasure-Smith N, Lesperance F: Depression—A cardiac risk factor in search of a treatment. JAMA. 2003;289:3171.

Freud S: Mourning and melancholia. In: Standard Edition of the Complete Psychological Works of Sigmund Freud. Vol. 4. London: Hogarth Press; 1975.

Gershon ES, Hamovit J, Guroff JJ, Dribble E, Leckman JF: A family study of schizoaffective, bipolar I, bipolar II, unipolar, and normal control probands. Arch Gen Psychiatry. 1982;39:1157.

Gessa GL: Dysthymia and depressive disorders: Dopamine hypothesis. Eur Psychiatry. 1996;11:123.

Gillin JC, Sitaram N, Duncan WC: Muscarinic supersensitivity: A possible model for the sleep disturbance of primary depression? Psychiatry Res. 1979;1:17. 

Goodwin FK, Jamison KR: Manic-Depressive Illness. 2nd ed. New York: Oxford University Press; 2007.

Hariri AR, Mattay VS, Tessitore A, Kolachana B, Fera F: Serotonin transporter genetic variation and the response of the human amygdale. Science. 2002;19:400.

Heninger GR, Delgado PL, Chamey DS: The revised monoamine theory of depression: A modulatory role for monoamines, based on new findings from monoamine depletion experiments in humans. Pharmacopsychiatry. 1996;29:2.

Hippocrates: Epidemics III. In: Jones WHS, tr. Hippocrates. Vol. 1. Cambridge, MA: Harvard University Press; 1923.

Holsboer F: The role of peptides in treatment of psychiatric disorders. J Neural Transm Suppl. 2003;64:17.

Jackson SW: Melancholia and Depression: From Hippocratic Times to Modem Times. New Haven, CT: Yale University Press; 1986.

Janowsky DS, El-Yousef MK, Davis JM, Sekerke HJ: A cholinergic-adrenergic hypothesis of mania and depression. Lancet. 1972;1:632.

Jorge RE, Robinson RG, Moser D, Tateno A, Crespo-Facorro B: Major depression following traumatic brain injury. Arch Gen Psychiatry. 2004;61:42.

Keller MB, Klerman GL, Lavori PW, Fawcett JA, Coryell W: Treatment received by depressed patients. JAMA. 1982;248:1848. P.1645

Kendler KS: Social support: A genetic-epidemiological analysis. Am J Psychiatry. 1997;154:1398.

Kendler KS, Karkowski-Shuman L: Stressful life events and generic liability to major depression: Genetic control of exposure to the environment? Psychol Med. 1997;27:359.

Kendler KS, Neale MC, Kessler RC, Heath AC, Eaves LJ: A longitudinal twin study of personality and major depression in women. Arch Gen Psychiatry. 1993;50:853.

Klerman GL, Lavori PW, Rice J, Reich T, Endicott J: Birth-cohort trends in rates of major depressive disorder among relatives of patients with affective disorder. Arch Gen Psychiatry. 1985;42:689.

Kovacs M, Akiskal HS, Gatsonis C, Parrone PL: Childhood-onset dysthymic disorder: Clinical features and prospective naturalistic outcome. Arch Gen Psychiatry. 1994;51:365. 

Kraepelin E: Manic-Depressive Insanity and Paranoia. Edinburgh: Livingstone; 1921. Krishman KR: Pituitary size in depression. J Clin Endocrinol Metab. 1991;72:256.

Kupfer DJ: REM latency: A psychobiologic marker for primary depressive disease. Biol Psychiatry. 1976;11:159.

Lake CR, Berrettini WH, Carpenter WT, Fawcett JA, Zubieta J: The psychoses dissected: The Kraepelinian dichotomy, a continuum or one disorder? Program No. S4. Annual Meeting of the American College of Neuropsychopharmacology, Boca Raton, FL; 2007.

Lapin IP, Oxenkrug GF: Intensification of the central serotoninergic processes as a possible determinant of the thymoleptic effect. Lancet. 1969;1:132.

Leonhard K: The Classification of Endogenous Psychoses. 5th ed. New York: Irvington; 1979.

Lesch KP, Bengel D, Heils A, Sabol SZ, Greenberg BD: Association of anxiety-related traits with a polymorphism in the serotonin transporter gene regulatory region. Science. 1996;274:1483.

Lewinsohn PM, Youngren MA, Grosscup SJ: Reinforcement and depression. In: Depue RE, ed. The Psychobiology of Depressive Disorders; Implications for the Effects of Stress. New York: Academic Press; 1979.

Lewis A: States of depression: Their clinical and aetiological differentiation. Br Med. 1938;J2:875.

Lozano AM, Mayberg HS, Giacobbe P, Hamani C, Craddock RC: Subcallosal cingulated gyrus deep brain stimulation for treatment-resistant depression. Biol Psychiatry. 2008;64:461.

Manji HK, McNamara R, Chen G, Lenox RH: Signalling pathways in the brain: Cellular transduction of mood stabilization in the treatment of manic-depressive illness. Aust N Z Psychiatry. 1999;33(Suppl):S65.

Mann JJ, Malone KM, Diehl DJ, Perel J, Cooper TB: Demonstration in vivo of reduced serotonin responsivity in the brain of untreated depressed patients. Am J Psychiatry. 1996;153:174.

MacQueen GM, Hajek T, Alda M: The phenotypes of bipolar disorder: Relevance for genetic investigations. Mol Psychiatry. 2005;10:811.

McMahon FJ, Simpson SG, McInnis MG, Badner JA, MacKinnon DF: Linkage of bipolar disorder to chromosome 18q and the validity of bipolar II disorder. Arch Gen Psychiatry. 2001;58:1025.

Merikangas KR, Akiskal HS, Angst J, Greenberg PE, Hirschfeld RMA: Lifetime and 12-month prevalence of bipolar spectrum disorder in the National Comorbidity Survey Replication. Arch Gen Psychiatry. 2007:64:543.

Mur M, Portella MJ, Martinez-Aran A, Pifarre J, Vieta E: Long-term stability of cognitive impairment in bipolar disorder: A 2-year follow-up study of lithium-treated euthymic bipolar patients. J Clin Psychiatry. 2008;69:712.

Murray CJL, Lopez AD, eds: The Global Burden of Disease. Geneva: World Health Organization; 1996.

Oedegaard KJ, Fasmer OB: Is migraine in unipolar depressed patients a bipolar spectrum trait? J Affect Disord. 2005;84:233.

Perugi G, Musetti L, Simonini E, Piagentini F, Cassano GB: Gender mediated clinical features of depressive illness: The importance of temperamental differences. Br J Psychiatry. 1990;157:835.

Petty F: GABA and mood disorders: A brief review and hypothesis. J Affect Disord. 1995;34:275.

Post RM: Transduction of psychosocial stress into the neurobiology of recurrent affective disorder. Am J Psychiatry. 1992;149:999. 

Prange AJ Jr, Wilson IC, Lynn CW, Alltop LB, Stikeleather RA: L-Tryptophan in mania: Contribution to a permissive hypothesis of affective disorders. Arch Gen Psychiatry. 1974;30:56.

Puig-Antich J: Affective disorders in children and adolescents. In: Meltzer HY, ed: Psychopharmacology: The Third Generation of Progress. New York: Raven; 1987.

Recupero PR, Harms SE, Noble JM: Googling suicide: Surfing for suicide information on the Internet. J Clin Psychiatry. 2008;69:878.

Rihmer Z, Rutz W, Pihlgren H: Depression and suicide on Gotland: An intensive study of all suicides before and after a depression-training programme for general practitioners. J Affect Disord. 1995;35:147.

Rihmer Z, Akiskal H: Do antidepressants t(h)reat(en) depressives? Toward a clinically judicious formulation of the antidepressant-suicidality FDA advisory in light of declining national suicide statistics from many countries. J Affect Disord. 2006;94:3.

Schildkraut JJ: The catecholamine hypothesis of affective disorders: A review of supporting evidence. Am J Psychiatry. 1995;22:509.

Seligman MD: Helplessness: On Depression, Development and Death. San Francisco: Freeman; 1975.

Siever LJ, Davis KL: Overview: Toward a dysregulation hypothesis of depression. Am J Psychiatry. 1985;142:1017.

Van Praag HM, Kahn RS, Asnis GM, Wetzler S, Brown SL: Denosologization of biological psychiatry or the specificity of 5-HT disturbances in psychiatric disorders. J Affect Disord. 1987;13:1.

Wehr TA, Rosenthal NE: Seasonality and affective illness. Am J Psychiatry. 1989;146:829.

Weller EB, Weller RA: Bipolar disorder in children: Misdiagnosis, underdiagnosis, and future diagnosis. J Am Acad Child Adolesc Psychiatry. 1995;34:709.

Winokur G, Clayton PJ, Reich T: Manic-Depressive Illness. St. Louis, MO: Mosby; 1969.

Zis KD, Zis A: Increased adrenal weight in victims of violent suicide. Am J Psychiatry. 1987;144:1214.