Is DSM-IV Bereavement Exclusion for Major Depressive Episode Relevant to Severity and Pattern of Symptoms? A Case-Control, Cross-Sectional Study

Is DSM-IV Bereavement Exclusion for Major Depressive Episode Relevant to Severity and Pattern of Symptoms? A Case-Control, Cross-Sectional Study

Emmanuelle Corruble, MD, PhD; Virginie-Anne Chouinard, MD; Alexia Letierce, PhD; Philip A. P. M. Gorwood, MD, PhD; and Guy Chouinard, MD, MSc Pharmacol

Objective: To assess the DSM-IV major depressive episode (MDE) bereavement exclusion criterion by comparing severity and pattern of symptoms in bereavement-excluded individuals satisfying all other DSM-IV MDE criteria to these same variables in MDE controls.

Method: A case-control, cross-sectional study of self-referred individuals seeking treatment for depressive symptoms was conducted. A total of 17,988 subjects met DSM-IV MDE symptom criteria. Of these, 1,521 individuals (8.5%) met all MDE criteria except the bereavement exclusion. They were matched by age, gender, marital status, and educational level with 1,521 MDE controls. Among the MDE controls, 292 had a recent bereavement and 1,229 did not. Severity of depression was measured by the number of MDE symptoms and the Montgomery-Asberg Depression Rating Scale (MADRS) score. Symptom cues of the bereavement-exclusion criterion were analyzed. The study was conducted between September 2003 and May 2004.

Results: Bereavement-excluded subjects were more severely depressed than MDE controls without bereavement and similar to MDE controls with bereavement. Two symptom cues, suicidal ideation and worthlessness, and the majority of other depressive symptoms were more pronounced in bereavement-excluded individuals than in MDE controls.

Conclusions: Symptom cues of the DSM-IV MDE bereavement exclusion criterion should be modified since they could result in patients failing to be correctly diagnosed and treated.

J Clin Psychiatry 2009;70(8):1091-1097

© Copyright 2009 Physicians Postgraduate Press, Inc.

Submitted: June 17, 2008; accepted September 2, 2008.

Online ahead of print: June 30, 2009 (doi:10.4088/JCP.08m04475).

Corresponding author: Emmanuelle Corruble, MD, PhD, Department of Psychiatry, Bicêtre University Hospital, 78 rue du General Leclerc, 94275 Le Kremlin Bicêtre, France ([email protected]).

Major depressive disorder (MDD)1-21 and suicidal ideation22-24 are frequently associated with bereavement. Predictive factors of major depressive episodes (MDEs) in bereaved individuals, however, are inconsistent across the literature.10-14,20,25-30 Thus, general practitioners and psychiatrists are frequently confronted with the question: is my patient normally bereaved or did he become depressed following bereavement?

Several cues have been proposed by the Diagnostic and Statistical Manual of Mental Disorders, Fourth Edition (DSM-IV)31 to distinguish normal bereavement and major depression in the context of bereavement. These cues are included in the MDE bereavement exclusion E criterion, which is a polythetic criterion. The aim of the DSM-IV bereavement exclusion criterion is to identify individuals who have “normal sadness” in reaction to loss, but do not have MDE. This exclusion criterion could prevent bereaved individuals who suffer from unpleasant, disturbing, but transient and not too severe depressive symptoms from being diagnosed with major depressive episode. The DSM-IV bereavement exclusion criterion selects bereaved individuals according to duration of episodes and level of depressive symptoms, and excludes them from the diagnosis of MDE, unless “the symptoms persist for longer than 2 months or are characterized by marked functional impairment, morbid preoccupation with worthlessness, suicidal ideation, psychotic symptoms, or psychomotor retardation.”31(p327) Therefore, bereaved individuals with 5 or more MDE symptoms (including depressed mood or lack of interest) are excluded from the diagnosis of MDE by the E criterion if their symptoms last less than 2 months and are not severe enough.

This DSM-IV MDE bereavement exclusion is a source of perennial controversy. On one hand, some authors11-13,32 such as Clayton have suggested that the bereavement exclusion is appropriate because bereavement is seen as a varied and normal response to attachment losses rather than as a mental disorder and should not be prematurely classified and treated as major depression. On the other hand, others1,33-38 such as Karam or Zisook have argued that the DSM-IV MDE bereavement exclusion has resulted in patients failing to be diagnosed and treated appropriately for major depression, and have emphasized the discrepancy that bereavement is the only stressful life event that can lead to exclusion from DSM-IV MDE diagnosis.

Another issue that should be addressed for the DSM-IV MDE bereavement exclusion is its relevance to office settings. The MDE bereavement exclusion criterion is a polythetic one, including both symptom duration cue and symptom severity cue. The symptom severity cue is also polythetic, referring to the severity of 4 depressive symptoms: worthlessness, suicidal ideation, psychotic symptoms, and psychomotor retardation. To our knowledge, there are no studies with large sample sizes assessing the relevance of the DSM-IV MDE bereavement exclusion criterion regarding both severity and pattern of symptoms.

Our hypothesis was that the bereavement-excluded subjects would be less severely depressed than MDE controls and would present fewer of the symptoms listed in the DSM-IV MDE bereavement exclusion criterion.

The aim of this case-control study was to compare the severity and pattern of symptoms in bereavement-excluded individuals satisfying all other DSM-IV MDE criteria to these same variables in MDE control subjects.

METHOD

Study Design

A French national cross-sectional study was designed to assess major depressive symptoms and stressful life events.39 It was carried out in 17,988 self-referred adults aged 18 years or older who met DSM-IV MDE symptom criteria with at least 5 of the 9 MDE symptoms including depressed mood or diminished interest or pleasure. Patients with schizophrenia or other psychotic disorders and symptoms, bipolar disorders, cognitive disorders, or current substance abuse or dependence (DSM-IV) were excluded. The MDE module of the Mini-International Neuropsychiatric Inventory (MINI) structured interview,40 which assessed DSM-IV MDE symptom criteria, does not include the bereavement exclusion-criterion.

The study was conducted between September 2003 and May 2004 by 2,408 physicians (74.3% general practitioners, 25.7% psychiatrists) uniformly distributed throughout France.39 The acceptance rate for participation was over 90%. Each physician was asked to include the first 6 consecutive depressed patients fulfilling DSM-IV MDE criteria with at least 5 MDE symptoms. The French national ethics committee (Commission Nationale de l’ Informatique et des Libertés [CNIL]) approved the study. All participants gave informed written consent. Confidentiality and anonymity were maintained.

As part of the French national cross-sectional study assessing major depressive symptoms and stressful life events, this case-control study was designed to evaluate the severity and pattern of symptoms in bereavement-excluded individuals satisfying all other DSM-IV MDE criteria and in MDE controls.

Study Participants

From the total sample of 17,988 subjects, 1,521 subjects (8.5%) were identified as bereavement-excluded individuals, meeting all DSM-IV criteria for MDE except the bereavement E criterion (BE group). The BE group were patients considered as not depressed but bereaved by the evaluating physicians. These subjects were identified by physicians on the basis of the DSM-IV clinical description of the bereavement E exclusion criterion. In order to more closely simulate clinical practice, no instruction was given to physicians on the bereavement exclusion prior to the study, and no specific procedure was used to record the presence or absence of the bereavement exclusion E criterion.

The 1,521 subjects of the BE group were matched by age, gender, marital status, and educational level with subjects from the same sample satisfying all the criteria for MDE (MDE group).

Assessment Procedure

The severity of depressive symptoms was assessed by the number of the MDE symptoms and the Montgomery-Asberg Depression Rating Scale (MADRS)41 score.

Three symptom cues of the bereavement exclusion criterion—suicidal ideation, worthlessness, and psychomotor retardation—were analyzed on the basis of MDE symptoms and MADRS item scores. The fourth one, psychotic symptoms, was not considered since it is not a MDE criterion symptom and is not included in the MADRS items.

Stressful life events occurring in the 3 months preceding the current depressive syndrome were identified using the Life Events Inventory (LEI).42 Events were rated as present or absent. Each item has a weighted score (1-100). Raters were blind to the weighted scores. Life events of loss were analyzed in order to distinguish MDE patients who had or did not have a recent loss (MDE with bereavement or MDE without bereavement).

The number of lifetime MDEs was recorded in 6 categories (0, 1, 2, 3, 4, and > 4). Standardized training and procedures were used to ensure interrater reliability.

Statistics

The MDE control group was separated into 2 subgroups, MDE controls with and without bereavement, on the basis of the presence/absence of a recent loss based on the LEI.

Univariate analyses were performed to examine differences between the BE and MDE groups using 2-sample t tests and Cochran-Mantel-Haenszel χ2 tests as appropriate. First, bereavement-excluded individuals were compared to matched MDE controls with bereavement in order to assess if the bereavement exclusion criterion discriminates these 2 subgroups correctly. Second, bereavement-excluded individuals were compared to matched MDE controls without bereavement. Third, the overall BE and MDE groups were compared in order to assess whether the bereavement exclusion criterion is easy to use and discriminative. Five variables of interest were determined a priori: the number of DSM-IV MDE symptoms, the MADRS total score, and item scores on 3 symptom cues of the bereavement exclusion criterion: suicidal ideation, worthlessness, and psychomotor retardation. In order to control for type I error, the other DSM-IV symptoms and MADRS items were considered when the total number of DSM-IV symptoms and MADRS total score comparisons were significant.

Finally, within the BE and MDE groups, conditional logistic regression analyses were conducted to identify predictors of classification into the BE group using the MDE control group as the reference group. Number of MDE symptoms, MADRS total score, and number of previous MDEs were the independent variables.

All tests were 2-tailed. Due to the number of comparisons, Bonferroni corrections were applied, and statistical significance was set at an α level of less than .01.

RESULTS

Patient Characteristics

The 17,988 participants who met DSM-IV MDE criterion symptoms (Part A of the DSM-IV MDE definition: at least 5 of the 9 MDE symptoms including depressed mood or lack of interest or pleasure) had a mean (SD) age of 48.6 (14.8) years, 11,352 (63.1%) were women, and 11,710 (65.1%) had at least 1 previous MDE.

Of the 17,988 participants, 13,377 (74.4%) met the full criteria for MDE (Parts A, B, C, D, and E of the DSM-IV MDE definition).

From the total sample of 17,988 subjects, 4,044 (22.5%) had experienced a loss of a loved one within 3 months as identified by the LEI. The losses of a loved one experienced by all bereaved participants (n = 4,044) were spouses (n = 1,145, 28.3%), children or parents (n = 1,634, 40.4%), close friends (n = 838, 20.7%), or others (n = 427, 10.6%).

Of the 17,988 participants, the BE group included 1,521 subjects (8.5%), fulfilling all MDE criteria except the E criterion. They had a mean (SD) age of 51 (14.8) years, 1,106 (72.7%) were women, and 1,107 (72.8%) had at least 1 previous MDE. The bereavement-excluded group included individuals who lost a variety of loved ones: spouses (n = 420, 27.6%), children or parents (n = 547, 36.0%), other family members (n = 325, 21.4%), close friends (n = 167, 11.0%), and others (n = 62, 4.1%).

Among the MDE controls, 292 (19.2%) had experienced the loss of a loved one within the last 3 months according to the LEI assessment: spouses (n = 131, 44.9%), children or parents (n = 79, 27.1%), other family members (n = 32, 11.0%), close friends (n = 31, 10.6%), and others (n = 19, 6.5%). Whereas no significant differences were shown between MDE controls with and without bereavement for MDE symptoms and for gender, MDE controls with bereavement were older (mean [SD] age = 56.9 [13.6] vs 49.6 [14.7] years, t1517 = −7.79; P < .0001) and, as expected, had significantly higher LEI scores (mean [SD] = 223.1 [137.6] vs 160.6 [106], t1519 = −8.51; P < .0001) than MDE controls without bereavement. Compared to MDE cases without bereavement, those with bereavement had a higher mean (SD) MADRS total score (31.4 [6.9] vs 29.6 [6.7], t1515 = −3.94; P < .0001).

Comparison of Bereavement-Excluded Individuals to Matched MDE Controls With Bereavement

MDE controls with bereavement (n = 292) and the matched BE subgroup were not significantly different for the number of MDE symptoms (mean [SD] = 6.59 [1.18] vs 6.69 [1.27]; t = −0.84; P = .39), for the MADRS total score (mean [SD] = 31.1 [6.38] vs 31.9 [7.21]; t = −1.19; P = .23), and for the 3 symptoms of interest: suicidal thoughts (DSM-IV: 35.1% vs 37.4%, χ2 = 0.26, P = .6; MADRS mean score: 2.15 [1.27] vs 2.11 [1.29], t = 0.32, P = .7), worthlessness (62.3% vs 64.3%, χ2 = 0.19, P = .65), and psychomotor retardation (65.7% vs 68.0%, χ2 = 0.30, P = .58). These results show that physicians failed to discriminate between these 2 subgroups on the 3 symptom cues of the DSM-IV MDE bereavement exclusion criterion.

Comparison of Bereavement-Excluded Individuals to MDE Controls Without Bereavement

Compared to MDE controls without bereavement (n = 1,229), the matched BE subgroup had a greater number of MDE symptoms, higher MADRS total scores, and higher levels of suicidal thoughts and worthlessness (Table 1). The BE subgroup also had more apparent sadness, reported sadness, reduced sleep, reduced appetite, weight disturbance, inability to feel, and pessimistic thoughts (Table 1).

Global Comparison of Bereavement-Excluded Subjects to MDE Controls With and Without Bereavement

The BE group had more MDE symptoms and higher mean MADRS total scores than the MDE control group (Table 2). The BE group also had higher levels of 2 of the 3 symptoms of the bereavement exclusion criterion: suicidal thoughts and worthlessness (Table 2). The BE group also had significantly higher levels of apparent sadness, reported sadness, reduced sleep, reduced appetite, weight disturbance, inability to feel, and pessimistic thoughts (Table 2).

Conditional logistic regression analysis showed that the BE group (n = 1,521) was more likely to have a higher number of MDE symptoms (OR = 1.10, 95% CI = 1.03 to 1.18, χ2 = 8.01, P = .004) and a higher MADRS total score (OR = 1.02, 95% CI = 1.01 to 1.03, χ2 = 10.01, P = .001) than the MDE control group (n = 1,521). These groups were not significantly different in terms of number of previous MDEs (OR = 0.89, 95% CI = 0.76 to 1.05, χ2 = 1.88, P = .17).

DISCUSSION

The present study was designed to evaluate the relevance of the symptom cues of the DSM-IV MDE bereavement exclusion criterion. Individuals satisfying all DSM-IV MDE criteria except the bereavement exclusion criterion were compared to matched MDE controls satisfying all of the DSM-IV MDE criteria. Our hypothesis was that the bereavement-excluded individuals would have less severe depressive symptoms than MDE controls, particularly less suicidal ideation, worthlessness, and psychomotor retardation since these symptoms are part of the exclusion criterion.

Bereaved individuals who were excluded from the DSM-IV MDE diagnosis but satisfied all other MDE criteria (BE group) represented 8.5% (n = 1,521) of 17,988 individuals who satisfied DSM-IV MDE criterion symptoms.

First, our results showed that physicians failed to discriminate between the BE group and MDE controls with bereavement for depression severity and pattern of symptoms. It is worth noting that the BE and MDE subjects with bereavement were quite similar. Second, the BE group was identified as more severely depressed and, unexpectedly, with more suicidal thoughts and worthlessness than nonbereaved MDE controls. The BE group also had higher levels of apparent sadness, reported sadness, reduced sleep, reduced appetite, weight disturbance, inability to feel, and pessimistic thoughts. Third, our results revealed that the BE group, ie, bereaved individuals meeting all DSM-IV MDE criteria but considered by the clinician as not having an MDE because of the bereavement exclusion criterion, were more severely depressed and had more suicidal thoughts and worthlessness than the matched MDE controls with or without bereavement. The BE group also had more apparent sadness, reported sadness, reduced sleep, reduced appetite, weight disturbance, inability to feel, and pessimistic thoughts compared to MDE control subjects. These results are consistent with those for global severity and symptoms of interest. However, despite high statistical significance in the context of a large sample size, the observed differences were relatively small and their clinical relevance could be discussed.

Our results are in agreement with studies showing that bereavement can lead to MDE1-21 and that suicidal ideation is more likely in bereaved individuals and is associated with severe depressive scores.22,24 However, our bereavement-excluded group (n = 1,521) differs from the bereaved individuals included in some other studies. The majority of bereavement studies11-20,24-26,30,41 showing high rates of major depression symptoms in bereaved individuals have focused on spousal bereavement in elderly populations, while our sample included different types of loss in a younger population (mean age of 51 years). Moreover, our bereavement-excluded group (n = 1,521) differs from the small sample of 56 subjects with depression triggered by uncomplicated bereavement studied by Wakefield et al.32 One explanation for the greater severity of depression in our bereavement-excluded group is that our sample consisted of bereaved individuals seeking treatment for depression, while Wakefield et al32 analyzed the 1990-1992 National Comorbidity Survey data derived from a community-based administered survey.

Regarding the controversy about the MDE bereavement exclusion, some authors such as Clayton28 have suggested that the bereavement exclusion is appropriate because bereavement is a varied and normal response to attachment losses. Our results argue against this viewpoint. Other authors, such as Kendler et al,4 Zisook et al,33 or Karam et al,38 have argued that the MDE bereavement exclusion results in patients failing to be diagnosed and treated appropriately for major depression and have emphasized the discrepancy that bereavement is the only stressful life event that can lead to exclusion from DSM-IV MDE diagnosis. Our results argue for this viewpoint.

There are limitations to the present study. First, there is prior evidence that patients with depressive episodes after grief are less likely to seek treatment than those with other kinds of depression.37 Thus, we could have a biased assessment since participants included were all individuals seeking treatment for depression. Our findings would therefore apply to bereaved individuals seeking treatment in physicians’ offices. As mentioned previously, this phenomenon could explain differences in illness severity between our bereavement-excluded group and the Wakefield et al32 study group. Second, data on comorbidity and functional impairment were not considered, in contrast to the Wakefield et al32 study. Third, psychotic symptoms were not analyzed because they are not included in the DSM-IV MDE criterion symptoms nor in the MADRS. Fourth, no causality relationship can be inferred from our results. The fact that the depressive episode occurred in the context of a loss does not mean that that the episode was caused by the loss. Nevertheless, our study is clinically relevant because it is based on the clinician’s judgment. Moreover, the period of 3 months before the onset of depression explored with the LEI may be a confounder as compared to the period of 2 months used in the DSM-IV bereavement exclusion criterion. Fifth, several points, such as differences between general practitioners and psychiatrists, the potential role of traumatic/complicated grief,43,44 and the longitudinal outcome of bereavement-excluded individuals, should be investigated in further studies.

The use of the DSM-IV MDE bereavement E criterion led unexpectedly to the exclusion of individuals more severely depressed than MDE controls without bereavement and not different in terms of severity compared to MDE patients with recent bereavement. Two symptom cues defining the bereavement exclusion, suicidal ideation and worthlessness, were more pronounced in bereaved individuals excluded from the diagnosis of MDE by the E criterion. The evidence from this study argues against the relevance of the symptom cues of the MDE bereavement exclusion E criterion. Instead of identifying a modest self-limited “normal” depressive syndrome, this criterion leads to the exclusion of individuals with symptoms of major depression that are more severe than those of typical MDE patients.

Since the bereavement exclusion criterion is the fifth and last criterion of the DSM-IV MDE definition, it might be relatively neglected in office-based practice as compared to the A, B, C, and D criteria. In this context, the complexity of the DSM-IV MDE bereavement polythetic exclusion criterion, which includes not only symptom cues but also a functional impairment cue and a time cue, may explain some of our results.

Our study would argue for the modification of the DSM-IV MDE exclusion E criterion because keeping it as it is could result in patients failing to be correctly diagnosed, living with prolonged and unnecessary suffering, and not getting appropriate treatment. A revision of this criterion would also allow a further step toward the integration of bereavement as part of stressful life events and a step away from the old DSM–II classification of endogenous versus neurotic/reactive depression (related to the disorder being induced or not being induced by a stressful life event).

Author affiliations: INSERM U 669, Department of Psychiatry (Drs Corruble, V.-A. Chouinard, and G. Chouinard), and Clinical Research Unit, Biostatistics Department (Dr Letierce), Paris XI University, Bicêtre University Hospital, Assistance Publique-H×´pitaux de Paris, Le Kremlin Bicêtre, France; INSERM U675, Paris VII University, Assistance Publique-H×´pitaux de Paris, Paris, France (Dr Gorwood); and Fernand-Seguin Research Centre, H×´pital Louis-H Lafontaine, Department of Psychiatry, University of Montreal, and Departments of Psychiatry and Medicine, Clinical Psychopharmacology McGill University Health Centre, Montreal, Quebec, Canada (Drs V.-A. Chouinard and G. Chouinard).

Financial disclosure: Dr Corruble has received consulting fees and honoraria within the last 5 years from Wyeth, Lilly, Servier, Pfizer, Bristol-Myers Squibb, and Janssen-Cilag. Dr Gorwood has, within the last 5 years, received research grants from Eli Lilly and Servier and speakers honoraria from Eli Lilly, Lundbeck, Servier, Janssen, and UCB-Pharma and has served on the advisory boards for Janssen, Servier, Bristol-Myers Squibb, and Wyeth. Dr G. Chouinard has received consulting fees and honoraria within the last 5 years from Janssen-Cilag, Neuro3d, Organon, Solvay, Schering Plough, Pfizer, and BioLineRx. Drs V.-A. Chouinard and Letierce report no additional financial or other relationships relevant to the subject of this article.

Funding/support: Supported by research grants from Paris XI University, INSERM U669, and Servier. Servier had no involvement in the design, organization, analysis, or preparation for publication of the study.

REFERENCES

1. Akiskal HS, McKinney WT Jr. Overview of recent research in depression: integration of ten conceptual models into a comprehensive clinical frame. Arch Gen Psychiatry. 1975;32(3):285-305. PubMed

2. Paykel ES, Tanner J. Life events, depressive relapse and maintenance treatment. Psychol Med. 1976;6(3):481-485. PubMed doi:10.1017/S0033291700015920

3. Kendler KS, Karkowski LM, Prescott CA. Causal relationship between stressful life events and the onset of major depression. Am J Psychiatry. 1999;156(6):837-841. PubMed

4. Kendler KS, Hettema JM, Butera F, et al. Life event dimensions of loss, humiliation, entrapment, and danger in the prediction of onsets of major depression and generalized anxiety. Arch Gen Psychiatry. 2003;60(8):789-796. PubMed doi:10.1001/archpsyc.60.8.789

5. Fava GA, Munari F, Pavan L, et al. Life events and depression: a replication. J Affect Disord. 1981;3(2):159-165. PubMed doi:10.1016/0165-0327(81)90040-9

6. Costello CG. Social factors associated with depression: a retrospective community study. Psychol Med. 1982;12(2):329-339. PubMed doi:10.1017/S0033291700046663

7. Kessler RC. The effects of stressful life events on depression. Annu Rev Psychol. 1997;48:191-214. PubMed doi:10.1146/annurev.psych.48.1.191

8. Thomson KC, Hendrie HC. Environmental stress in primary depressive illness. Arch Gen Psychiatry. 1972;26(2):130-132. PubMed

9. Surtees PG, Miller PM, Ingham JG, et al. Life events and the onset of affective disorder: a longitudinal general population study. J Affect Disord. 1986;10(1):37-50. PubMed doi:10.1016/0165-0327(86)90047-9

10. Zisook S, Shuchter SR. Depression through the first year after the death of a spouse. Am J Psychiatry. 1991;148(10):1346-1352. PubMed

11. Clayton PJ, Halikas JA, Maurice WL. The depression of widowhood. Br J Psychiatry. 1972;120(554):71-77. PubMed doi:10.1192/bjp.120.554.71

12. Clayton PJ. The clinical morbidity of the first year of bereavement: a review. Compr Psychiatry. 1973;14(2):151-157. PubMed doi:10.1016/0010-440X(73)90007-2

13. Clayton PJ. The sequelae and nonsequelae of conjugal bereavement. Am J Psychiatry. 1979;136(12):1530-1534. PubMed

14. Bruce ML, Kim K, Leaf PJ, et al. Depressive episodes and dysphoria resulting from conjugal bereavement in a prospective community sample. Am J Psychiatry. 1990;147(5):608-611. PubMed

15. Breckenridge JN, Gallagher D, Thompson LW, et al. Characteristic depressive symptoms of bereaved elders. J Gerontol. 1986;41(2):163-168. PubMed

16. Futterman A, Gallagher D, Thompson LW, et al. Retrospective assessment of marital adjustment and depression during the first 2 years of spousal bereavement. Psychol Aging. 1990;5(2):277-283. PubMed doi:10.1037/0882-7974.5.2.277

17. Harlow SD, Goldberg EL, Comstock GW. A longitudinal study of the prevalence of depressive symptomatology in elderly widowed and married women. Arch Gen Psychiatry. 1991;48(12):1065-1068. PubMed

18. Turvey CL, Carney C, Arndt S, et al. Conjugal loss and syndromal depression in a sample of elders aged 70 years or older. Am J Psychiatry. 1999;156(10):1596-1601. PubMed

19. Onrust SA, Cuijpers P. Mood and anxiety disorders in widowhood: a systematic review. Aging Ment Health. 2006;10(4):327-334. PubMed doi:10.1080/13607860600638529

20. Zisook S, Paulus M, Shuchter SR, et al. The many faces of depression following spousal bereavement. J Affect Disord. 1997;45(1-2):85-95. PubMed doi:10.1016/S0165-0327(97)00062-1

21. Maciejewski PK, Zhang B, Block SD, et al. An empirical examination of the stage theory of grief. JAMA. 2007;297(7):716-723. PubMed doi:10.1001/jama.297.7.716

22. Stroebe M, Stroebe W, Abakoumkin G. The broken heart: suicidal ideation in bereavement. Am J Psychiatry. 2005;162(11):2178-2180. PubMed doi:10.1176/appi.ajp.162.11.2178

23. Prigerson HG, Bridge J, Maciejewski PK, et al. Influence of traumatic grief on suicidal ideation among young adults. Am J Psychiatry. 1999;156(12):1994-1995. PubMed

24. Szanto K, Prigerson H, Houck P, et al. Suicidal ideation in elderly bereaved: the role of complicated grief. Suicide Life Threat Behav. 1997;27(2):194-207. PubMed

25. Hays JC, Kasl S, Jacobs S. Past personal history of dysphoria, social support, and psychological distress following conjugal bereavement. J Am Geriatr Soc. 1994;42(7):712-718. PubMed

26. Nuss WS, Zubenko GS. Correlates of persistent depressive symptoms in widows. Am J Psychiatry. 1992;149(3):346-351. PubMed

27. Clayton PJ, Herjanic M, Murphy G. Mourning and depression: their similarities and differences. Can Psychiatr Assoc J. 1974;19(3):309-312. PubMed

28. Clayton PJ. Bereavement and depression. J Clin Psychiatry. 1990;51(suppl):34-38. PubMed

29. Chen JH, Bierhals AJ, Prigerson HG, et al. Gender differences in the effects of bereavement-related psychological distress in health outcomes. Psychol Med. 1999;29(2):367-380. PubMed doi:10.1017/S0033291798008137

30. Thompson LW, Gallagher-Thompson D, Futterman A, et al. The effects of late-life spousal bereavement over a 30-month interval. Psychol Aging. 1991;6(3):434-441. PubMed doi:10.1037/0882-7974.6.3.434

31. American Psychiatric Association. Diagnostic and Statistical Manual of Mental Disorders. 4th ed. Washington, DC: American Psychiatric Association; 1994.

32. Wakefield JC, Schmitz MF, First MB, et al. Extending the bereavement exclusion for major depression to other losses. Arch Gen Psychiatry. 2007;64(4):433-440. PubMed doi:10.1001/archpsyc.64.4.433

33. Zisook S, Shear K, Kendler KS. Validity of the bereavement exclusion criterion for the diagnosis of major depressive episode. World Psychiatry. 2007;6(2):102-107. PubMed

34. Zimmerman M, McGlinchey JB, Chelminski I, et al. Diagnosing major depressive disorder: applying the DSM-IV exclusion criteria in clinical practice. J Nerv Ment Dis. 2006;194(7):530-533. PubMed doi:10.1097/01.nmd.0000224882.67660.3d

35. Zimmerman M, Spitzer R. Psychiatric classification. In: Sadock BSV, ed. Kaplan and Sadock’s Comprehensive Textbook of Psychiatry. Philadelphia, PA: Lippincott Williams &Wilkins; 2005:1003-1034.

36. Zisook S, Kendler KS. Is bereavement-related depression different than non-bereavement-related depression? Psychol Med. 2007;37(6):779-794. PubMed doi:10.1017/S0033291707009865

37. Karam EG. The nosological status of bereavement-related depressions. Br J Psychiatry. 1994;165(2):48-52. PubMed doi:10.1192/bjp.165.1.48

38. Karam EG, Tabet CC, Alam D, et al. Bereavement related and non-bereavement related depressions: a comparative field study. J Affect Disord. 2009;112(1-3):102-110. PubMed doi:10.1016/j.jad.2008.03.016

39. Corruble E, Falissard B, Gorwood P. Life events exposure before a treated major depressive episode depends on the number of past episodes. Eur Psychiatry. 2006;21(6):364-366. PubMed doi:10.1016/j.eurpsy.2006.03.006

40. Lecrubier Y, Sheehan DV, Weiller E, et al. The Mini-International Neuropsychiatric Interview (M.I.N.I.): a short diagnostic structured interview: reliability and validity according to the CIDI. Eur Psychiatry. 1997;12(5):224-231. doi:10.1016/S0924-9338(97)83296-8

41. Montgomery SA, Asberg M. A new scale designed to be sensitive to change. Br J Psychiatry. 1979;134:382-389. PubMed doi:10.1192/bjp.134.4.382

42. Holmes TH, Rahe RH. The social readjustment rating scale. J Psychosom Res. 1967;11(2):213-217. PubMed doi:10.1016/0022-3999(67)90010-4

43. Prigerson HG, Shear MK, Jacobs SC, et al. Consensus criteria for traumatic grief: a preliminary empirical test. Br J Psychiatry. 1999;174:67-73. PubMed doi:10.1192/bjp.174.1.67

44. Prigerson HG, Frank E, Kasl SV, et al. Complicated grief and bereavement-related depression as distinct disorders: preliminary empirical validation in elderly bereaved spouses. Am J Psychiatry. 1995;152(1):22-30. PubMed