Bereavement, Complicated Grief and DSM Part 1: Depression

“To spare oneself from grief at all costs can be achieved only at the price of total detachment, which excludes the ability to experience happiness.”

Erich Fromm

The case of Mr. A

Mr. A is a 73-year-old male whose wife of 50 years died 5 weeks previously. He has no appetite, has lost 8 lbs in the past month, consistently awakens at 4 a.m., “can’t think straight,” and no longer takes any pleasure in customary activities. He denies feelings of guilt or worthlessness. Although he denies suicidal intent, he confides that he wishes to join his dearly departed wife. When discussing his wife, he shows moderate psychomotor agitation and spends most days mindlessly sitting in front of his television.

According to the DSM-IV-TR, Mr. A does not have a psychiatric condition; consequently, treatment with antidepressants or formal psychotherapy is not indicated. Mr. A would be given the V-code, bereavement, and reassured that nothing is wrong. However, according to the ICD-10, Mr. A has MDD. Ostensibly, decisions about treatment would be made just as they would after any other, non-bereavement related, episode of MDD. The question to be addressed in the following discussion is whether the best available evidence more strongly supports the DSM-IV-TR or the ICD-10.

History of the DSM Bereavement Exclusion

Prior to the 1980 publication of the Diagnostic and Statistical Manual-Third Edition (DSM-III), bereavement was not part of psychiatry’s official nomenclature. DSM-III introduced recent bereavement as an exclusion for the diagnosis of major depressive episode (MDE) and as a V-Code (other conditions that may be a focus of clinical attention but are not themselves instances of mental disorders): As part of their reaction to the loss, some grieving individuals present with symptoms characteristic of a MDE.…the duration and expression of ‘normal’ bereavement vary considerably among different cultural groups.…the diagnosis of Major Depressive Disorder (MDD) is generally not given unless the symptoms are still present 2 months after the loss…however, the presence of certain symptoms (guilt, suicidal thoughts, worthlessness, psychomotor retardation, marked functional impairment and psychotic features) that are not characteristic of a “normal” grief reaction may be helpful in differentiating bereavement from a MDE”. Thus, according to the DSM-IV-TR, an individual who meets all symptomatic, duration and impairment criteria for MDD but is recently bereaved may not have MDD; in contrast, a non-bereaved individual with the same clinical constellation of symptoms, who is recently divorced, impoverished or disabled, or who cannot identify any recent adversity, does have MDD. Does the preponderance of available data support this distinction?

In the most current edition, DSM-IV-TR, bereavement remains an exclusion for the diagnosis of MDE and continues as a V-Code, but bereavement also is mentioned as an exclusion for the diagnosis of Adjustment Disorder and, under certain circumstances, for the diagnosis of Posttraumatic Stress Disorder (PTSD). Of interest, the International Statistical Classification of Diseases and Related Health Problems-10th Revision (ICD-10) does not exclude the diagnosis of MDE or Adjustment Disorder based on recent bereavement, and does not have a V-Code for bereavement. That there are differences in the ways the DSM-IV-TR and the ICD-10 deal with the nosological status of bereavement speaks to uncertainty, disagreement among “experts” and the lack of reliable data to guide diagnostic and treatment decisions in the context of bereavement. Given that one of the goals of the DSM-5 development committees is to better align diagnostic criteria with ICD-10 and 11, the time has come to evaluate relevant evidence for and against present diagnostic conventions and, and make changes based on the best available evidence. This two-part series focuses on the two most controversial issues regarding the role of bereavement in psychiatric diagnoses: 1) should the recent death of a loved one continue to exclude the diagnosis of MDE (part 1); and 2) is there a point at which grief fails to be adaptive and should be diagnosed as a clinical condition requiring formal treatment (part 2)? Based on the best available data, the authors will conclude with recommendations for DSM-5.

Bereavement and Depression

Bereavement is a universal stressor that is one of the most likely to precipitate an episode of major depression^{1, 2}. Studies show that approximately one-third of all widows or widowers manifest a full major depressive episode 1 month after the death of a spouse, approximately one-fourth at 7 months, approximately 15% at 1 and 2 years, and up to 10% may meet criteria for MDE for the entire year^3-5. Yet many clinicians are confused by the relationship between grief and depression, and are uncertain about when to make the diagnosis of MDE in bereaved individuals. The principal source of diagnostic confusion is the common occurrence of low mood, sadness, and social withdrawal in both bereavement and MDE. The DSM-III and its subsequent iterations have attempted to prevent over-diagnosis of MDD in bereaved persons who are sad and withdrawn by excluding recently (less than 2 months after the death) and acutely (duration less than 2 months) bereaved individuals from the diagnosis of MDE, unless they also met certain other conditional criteria (worthlessness, psychomotor retardation, suicidal ideation, psychotic features, severe distress or dysfunction). However, many grief experts, rather than considering a full major depressive syndrome in the context of grief to be “normal”, question the DSM-IV-TR convention to wait a full 2 months before making the diagnosis.

The bereavement exclusion was originally introduced immediately after publication of a series of reports by Paula Clayton, et al^4,5 documenting the high prevalence of major depressive syndromes occurring during bereavement. Because these depressive syndromes tended to be relatively mild, usually dissipated over time, without treatment, and “differed” from clinical depression in several ways, Clayton cautioned against over-diagnosing major depression during the first year of bereavement. Since then, however, two reviews^{3, 6} have noted the similarities between bereavement related depressive syndromes and other nonbereavment related MDD, in terms of clinical and biological characteristics, common comorbidities, course and treatment response. In addition, three large secondary analyses of large population based data bases^7-9 have demonstrated similarities between bereavement-related major depressive syndrome and other life-event related depressions, with respect to demographic and clinical characteristics, intensity, familiality, course, associated features and treatment responses. In addition, in a large population-based prospective study, Karam et al¹⁰, reported that the global symptom profile of depressed individuals and their risk for depressive recurrence was similar in bereaved and non-bereaved subjects, and the duration of illness was actually longer in the bereaved group. Further, in a large, case-control, cross sectional study of a national data base, Corruble et al¹¹ found that subjects who are excluded from the diagnosis of MDE on the basis of current DSM-IV-TR conventions are, if anything, even more severely depressed than MDD controls without bereavement. None of these reviews or studies provides support for the special treatment given to bereavementrelated depression in the DSM. The conclusion is either that all depressive episodes that occur soon after a stressful life event and that are not associated with the conditional features of morbid feelings of worthlessness, psychomotor retardation, suicidal ideation, psychotic features or marked and prolonged functional impairment should not be given the diagnosis of MDE (the position of Wakefield et al⁷); or that the bereavement exclusion should be eliminated from the DSM-5 (the conclusion of Zisook and Kendler³, Zisook et al⁶, Kendler et al⁸, Kessing et al⁹, Karam¹⁰, Corruble et al¹¹ and possibly Clayton⁵).

Returning to the case of Mr. A

The preponderance of available evidence supports the ICD-10 convention of diagnosing MDD when all symptomatic, duration and severity criteria are met. It does not support the exclusivity of bereavement as the only life event that negates the diagnosis of MDD. Mr. A would be ill-served if his MDD were “explained away” by his grief, and if he were thereby denied the best available treatment for his depression.

Recommendations for DSM-5

Based on the best available data as briefly reviewed in this manuscript, we recommend DSM-5:

1. Eliminate the bereavement exclusion for the diagnosis of MDD.

2. Either eliminate the V-Code Bereavement, or specify that it should not be used when symptoms can be better explained by MDD, Adjustment Disorder (space does not permit discussion of why we believe the bereavement exclusion for Adjustment Disorder also should be eliminated), PTSD or Complicated Grief. More useful than the V-Code as now conceived would be a fuller description of uncomplicated grief and of the phenomenological distinctions between the dysphoria associated with grief and with major depression¹².