Depressive Symptoms After Loss

May 13, 2009

To improve validity, we proposed extending the current MDD bereavement exclusion-which excludes “uncomplicated” (relatively brief, lacking certain severe symptoms) depressive bereavement from diagnosis-to also exclude uncomplicated reactions to other major stressors, such as romantic breakups, job loss, and serious medical diagnoses.

Depressive Symptoms After Loss

Response: An Epidemic of Depression

Psychiatrists Should Not Fall Back on DSM

The Lure of the Context-Dependent Psychiatric Diagnosis

Depression or Major Loss, or Both?

In “Major Depression After Recent Loss Is Major Depression-Until Proved Otherwise” (Psychiatric Times, December 2008, page 12), Dr Ronald Pies critiques our earlier article (“An Epidemic of Depression,” Psychiatric Times, November 2008, page 44) and our book, The Loss of Sadness: How Psychiatry Transformed Normal Sorrow Into Depressive Disorder (Oxford University Press). We had argued that DSM’s reliance on decontextualized symptomatic criteria to define major depressive disorder (MDD) confuses disorder with intense normal sadness in response to loss, which frequently includes DSM/MDD–like symptoms. To improve validity, we proposed extending the current MDD bereavement exclusion-which excludes “uncomplicated” (relatively brief, lacking certain severe symptoms) depressive bereavement from diagnosis-to also exclude uncomplicated reactions to other major stressors, such as romantic breakups, job loss, and serious medical diagnoses. The evidence (presented in our book) suggests that such uncomplicated responses are frequent but mostly transient and are likely normal–range responses.

Dr Pies, to the contrary, proposes eliminating the bereavement exclusion and maintains that even uncomplicated depressive reactions to loss are disorders. However, many of his arguments fail to support his position:

1. Pies questions whether there are controlled, prospective data showing self–limited course and lack of treatment response in uncomplicated bereavement–related depression. A vast amount of evidence cited in our book, including many prospective studies, indicate high MDD–level symptom rates immediately following major loss that decline steeply over time. This is why the bereavement exclusion was introduced and why its extension is warranted.

2. Pies’ focus on treatment studies confuses drug response with disorder status. Ritalin helps with attention–deficit/hyperactivity disorder and with studying for examinations, and Valium helps with generalized anxiety disorder and normal performance anxiety. Even if antidepressants help with normal grief (the jury is still out), that wouldn’t imply that the recipient has a disorder. We are not opposed to medicating patients with normal distress; rather, we object to mislabeling conditions as disorders, thus biasing prognosis, informed consent, and treatment planning.

3. The 2 studies Pies cites in his commentary as support reveal the poverty of evidence for his position. In the first study, Karam and colleagues1 showed uncomplicated MDD to have recurrence rates similar to complicated MDD but, as they acknowledge, their uncomplicated samples were “too small to allow for generalization.” The second study is irrelevant because Brent’s2 adolescent sample had depressions of 8 months’ average duration, outside the “uncomplicated” domain. In addition, Pies did not consider our own empirical study that showed that the incidence of uncomplicated MDD following bereavement and other stressors is markedly lower on pathology indicators than complicated cases.3

4. Pies also fallaciously concludes that, because errors of causal attribution can occur, “trigger” is a nebulous notion. In fact, the role of stress in triggering depressive episodes-with additional causal factors, of course-is one of the best–confirmed hypotheses in psychiatry. Despite the inevitable possibility of error, physicians throughout history have considered context crucial in MDD diagnosis.

5. Pies quotes a historian’s opinion that physicians helped people without attending to normality versus disorder. While it is true that physicians have always helped suffering people-with or without a disorder-they have also explicitly addressed the diagnostic issue of whether the patient’s distressed condition is normal or disordered. The historian notes “no clear, bright line between disease and health.” This is true, but neither is there a bright line between, say, child and adult. The distinction between normal and disordered sadness is similarly real despite considerable boundary fuzziness, with clear cases on both sides-some of which are misclassified by DSM criteria.

6. Pies objects to assessing whether MDD symptoms are “proportional” to context without a research–validated scale. However, many criteria sets require clinicians to judge without formal scales whether symptoms are “excessive” or “unreasonable”-proxies for proportionality-to distinguish mental disorders from normal responses. MDD is the exception.

7. Pies argues (citing Zisook4) that myocardial infarctions (MIs) are triggered by stress but are disorders nonetheless. We know MIs are biological dysfunctions. Genuine disorders, whether MIs or “complicated” depressive reactions, can of course be caused by environmental stress. But sadness, unlike MI, is biologically designed to occur in reaction to specific environmental events, so judging whether MDD symptoms are pathological requires reference to context.

Pies, lacking scientific evidence, disparages commonsense observation of human nature at psychiatry’s peril. If DSM–V eliminates the bereavement exclusion, thereby making intense depressive grief after loss a disorder, not only the validity but the credibility of psychiatric diagnoses will be undermined, playing into the hands of the antipsychiatry movement.

Jerome C. Wakefield, PhD, DSW New York

Allan V. Horwitz, PhD New Brunswick, NJ

Dr Wakefield is university professor, professor of social work, and professor of the conceptual foundations of psychiatry, as well as affiliate faculty in bioethics and in the Center for Ancient Studies at New York University. He is the author of more than 150 publications on the conceptual foundations of the mental health professions. He is author with Allan Horwitz of The Loss of Sadness: How Psychiatry Transformed Normal Sorrow Into Depressive Disorder (Oxford, 2007), named best psychology book of 2007 by the American Association of Professional and Scholarly Publishers.

Dr Horwitz is professor of sociology and dean of social and behavioral sciences at Rutgers University in New Jersey. In addition to The Loss of Sadness, he is the author of 4 other books and over 75 articles and book chapters on social aspects of mental illness.

Neither author has any disclosures to report of conflicts of interest or financial support.


1. Karam EG, Tabet CC, Alam D, et al. Bereavement related and non–bereavement related depressions: a comparative field study. J Affect Disord. 2009;112:102–110.
2. Brent DA, Perper JA, Moritz G, et al. Major depression or uncomplicated bereavement? A follow–up of youth exposed to suicide. J Am Acad Child Adolesc Psychiatry. 1994;33:231–239.
3. Wakefield JC, Schmitz MF, First MB, Horwitz AV. Extending the bereavement exclusion for major depression to other losses: evidence from the National Comorbidity Survey. Arch Gen Psychiatry. 2007;64:433–440.
4. Zisook S. Commentary on chapter 1, “diagnosis of depressive disorders.” In: Herrman H, Maj M, Sartorius N, eds. Depressive Disorders. 3rd ed. Vol 9. Hoboken, NJ: John Wiley & Sons. In press.

Drs Pies and Zisook respond:

[Editor–in–Chief’s note: Since my editorial drew heavily on the work of Sidney Zisook, MD, and colleagues, I have invited Dr Zisook to serve as coauthor of this response.]

We appreciate the detailed rejoinder from Professors Wakefield and Horwitz (henceforth, W–H). Obviously, no competent psychiatrist believes that mere “sadness” is a pathological condition that requires professional treatment or considers uncomplicated grief an illness. We no more want to “medicalize” a normal condition (grief) than W–H want to “normalize” a pathological one (eg, MDD). But if-absent appropriate data-W–H are claiming that “non–disordered sadness” is the proper designation for a syndrome that meets full severity and duration criteria for MDD but which occurs within 2 months of a major loss, then they are simply engaging in an elaborate renaming exercise. We can rename pancreatic cancer “pancreatic hyperplasia,” but it remains a fatal disease. Renaming as “appropriate sadness” DSM–IV major depression that occurs within 2 months of bereavement does not show that this condition is less debilitating than major depression without recent loss. Indeed, most data suggest that MDD with loss is every bit as debilitating, “biological,” and recurrent as major depressive syndromes that occur in any other context, with or without clear–cut environmental precipitants.1–3

We certainly agree that attention to context may help formulate personalized treatment interventions and influence outcome. But context should not define diagnosis. In many ways, W–H’s thesis harks back to the pre–DSM–III categories of type A depression (severe illness, akin to W–H’s complicated MDD) and type B depression (mild illness, more like W–H’s “uncomplicated” group).4 Other terms for this dichotomy included autonomous–reactive, endogenous–exogenous, psychotic–neurotic, and endogenomorphic–nonendogenomorphic.5,6 Yet features intended to differentiate type A and B depressions, including the presence of precipitating events, failed to validate these subtypes.7–10 Indeed, “it is difficult in clinical practice to discriminate between different categories of depression on the basis of presence or absence of a psychic trauma or a stressful life event … the inherent problem may be that it is difficult to decide whether a present life event … constitutes a psychic trauma for the individual patient or not.”9

The Wakefield study,11 which supposedly supported the W–H thesis, retrospectively evaluated individuals who met MDD symptom criteria and whose MDD episodes were triggered by either bereavement or other loss. Subjects were subdivided into “uncomplicated” and “complicated” cases. Essentially, those with “uncomplicated” cases lacked suicidal ideation, marked functional impairment or psychomotor retardation, feelings of worthlessness, or prolonged duration of symptoms. The study found that:

• MDD after bereavement was far more similar to than different from depressive syndromes that occurred after other stressful life events.

• Those with “complicated” cases showed more severe pathology than those with “uncomplicated” cases (a finding we consider essentially tautological).

The researchers reached 2 main conclusions:

• Bereavement–related depression does not deserve special diagnostic status compared with other life–event triggered depression. We agree with this conclusion.

• All “uncomplicated,” “triggered depressions” should be exempt from the diagnosis of MDD. We disagree with this conclusion. This study did not compare depression “with and without cause” (indeed, fewer than 5% of all cases lacked an identifiable environmental trigger!); nor was it prospective.

Moreover, as the authors noted, “the accuracy of the … respondents’ self–reports of triggering events is unknown; respondents may misremember whether there was an event, or whether the timing of an event was before an episode. Further, they may misattribute the cause of an episode to an event when they were coincidental.” We fully agree! Furthermore, as Dr Glen Gabbard observes:

Often the stressor identified by the patient (or therapist) is retrospectively assumed to be the cause [of the depression] because it fits a particular intrapsychic narrative-often one involving victimization. It’s rarely that simple. Often there are multiple stressors; issues involving adult developmental phases; dashed fantasies and hopes; and failures to live up to the expectations of internalized parents (personal communication, December 31, 2008).

Another large community epidemiological study also found many more similarities than differences between bereavement–related depressive syndromes and depressions related to other, nonbereavement life events.3 This study failed to support the special status accorded “bereavement” in DSM–IV. The authors concluded, “bereavement–related depression often is recurrent, genetically influenced, impairing, and treatment responsive. These are all characteristics that are likely to be more associated with ‘MDD’ than with ‘normal sadness.’”3

We acknowledge that the data are not conclusive and allow room for debate. Opinions regarding the bereavement exclusion or the exclusion of “triggered” depression from the category of MDD are based largely on “clinical wisdom” and interpretations of studies that were not designed to answer these questions. A definitive study would require prospective comparison of persons whose depressions occur in the context of several common life stressors with persons whose depressions appear unrelated to environmental triggers on symptom profile and duration, hospitalization rates, suicide attempts, social and vocational impairment, course, and so on. We do not believe there are any current studies that conform to these parameters. Nevertheless, based on our clinical experience and the preponderance of available evidence, we conclude that continuing the bereavement exclusion in DSM–V would be a serious error. It would encourage bereaved individuals, their families, and health care providers to ignore signs and symptoms of a potentially debilitating, life–threatening-yet treatable-disorder. Extending this exclusion to still other loss events could create a public health disaster. Our patients deserve better.

Ronald Pies, MD Boston

Sidney Zisook, MD San Diego

Dr Pies is professor of psychiatry and lecturer on bioethics and humanities at State University of New York (SUNY) Upstate Medical University in Syracuse and clinical professor of psychiatry at Tufts University in Boston.

Dr Zisook is professor and director of residency training, department of psychiatry, University of California San Diego and staff physician, VA San Diego Health Science Center.


1. Zisook S, Shear K, Kendler KS. Validity of the bereavement exclusion criterion for the diagnosis of major depressive episode. World Psychiatry. 2007;6:102–107.
2. Karam EG, Tabet CC, Alam D, et al. Bereavement related and non–bereavement related depressions: a comparative field study. J Affect Disord. 2009;112: 102–110.
3. Kendler KS, Myers J, Zisook S. Does bereavement–related major depression differ from major depression associated with other stressful life events? Am J Psychiatry. 2008;165:1449–1455.
4. Kendell RE. The classification of depressions: a review of contemporary confusion. Br J Psychiatry. 1976;129:15–28.
5. Akiskal HS, Bitar AH, Puzantian VR, et al. The nosological status of neurotic depression: a prospective three– to four–year follow–up examination in light of the primary–secondary and unipolar–bipolar dichotomies. Arch Gen Psychiatry. 1978;35:756–766.
6. Klerman GL, Endicott J, Spitzer R, Hirschfeld RM. Neurotic depressions: a systematic analysis of multiple criteria and meanings. Am J Psychiatry. 1979;136: 57–61.
7. Leff MJ, Roatch JF, Bunney WE Jr. Environmental factors preceding the onset of severe depressions. Psychiatry. 1970;33:293–311.
8. Paykel ES, Myers JK, Dienelt MN, et al. Life events and depression. A controlled study. Arch Gen Psychiatry. 1969;21:753–760.
9. Kessing LV. Endogenous, reactive and neurotic depression-diagnostic stability and long–term outcome. Psychopathology. 2004;37:124–130.
10. Kessing LV. Epidemiology of subtypes of depression. Acta Psychiatr Scand Suppl. 2007;(433):85–89.
11. Wakefield JC, Schmitz MF, First MB, Horwitz AV. Extending the bereavement exclusion for major depression to other losses: evidence from the National Comorbidity Survey. Arch Gen Psychiatry. 2007;64: 433–440.