Journal of Affective Disorders 118 (2009) 201–204

Contents lists available at ScienceDirect

Journal of Affective Disorders

j ourna l homepage: www.e lsev ie r.com/ locate / j ad

Brief report

Complicated grief in patients with unipolar depression

Anette Kersting a,⁎, Kristin Kroker a, Julia Horstmann a, Patricia Ohrmann a, Bernhard T. Baune b,Volker Arolt a, Thomas Suslowa

a Department of Psychiatry, University of Muenster, Germanyb Department of Psychiatry, School of Medicine, James Cook University, Townsville, Australia

a r t i c l e i n f o

⁎ Corresponding author. Department of Psychiatry,Albert-Schweitzer-Str. 11, D-48129 Münster, Germ8356677; fax: +49 251 8356603.

E-mail address: Anette.Kersting@ukmuenster.de (A

0165-0327/$ – see front matter © 2009 Elsevier B.V.doi:10.1016/j.jad.2009.01.033

a b s t r a c t

Article history:Received 11 November 2008Received in revised form 28 January 2009Accepted 28 January 2009Available online 5 March 2009

Background: The loss of a close family member (e.g. child or spouse) has been shown to be oneof the most stressful life-events increasing the risk of affective disorders. In the present study,we investigated for the first time the frequency of complicated grief in psychiatric inpatientswith unipolar depression. Further, the study was aimed to identify characteristics predicting acomplicated grief reaction in depressed patients.

Methods: In a sample of 73 DSM-IV diagnosed unipolar affective disordered inpatients grief,depression, anxiety and psychological stress reaction were assessed.

Results: A high prevalence of loss and impairing complicated grief was found in this sample ofunipolar depressed patients. Depressed patients with complicated grief were more severelydepressed than depressed patients without complicated grief reactions. Higher traumatic stressand close family membership of the lost person were associated with higher severity of grief.

Conclusions: Comorbid complicated grief appears to contribute to greater severity and poorerfunctioning in unipolar depressed patients and should be specifically addressed inpsychotherapeutic treatment.

© 2009 Elsevier B.V. All rights reserved.

Keywords:DepressionUnipolar affective disorderComplicated griefBereavement

1. Introduction

Bereavement has been shown to be a severe stressful lifeevent, greatly affecting physical, social and psychological wellbeing (Stroebe et al., 2007).Moreover, severe grief increases therisk of major depressive episodes (Li et al., 2005) includingsuicidal ideation (Szanto et al., 2006), and is a risk factor formortality (Li et al., 2003; Tomassini et al., 2002; Lichtensteinet al., 1998). In the bereaved, approximately 40% meet criteriafor major depressionwithin a month of the death and one yearafter the loss approximately 15% of the bereaved persons aredepressed (Hensley, 2006). Research in the area of bereave-ment has distinguished between normal and complicated grief,the latter being a separate form of psychopathology emergingin the context of bereavement (Boelen and vandenBout, 2008).

University Muenster,any. Tel.: +49 251

. Kersting).

All rights reserved.

Complicated grief (CG) has been shown to constitute a one-dimensional symptom cluster comprised of symptoms ofseparation distress and traumatic distress persisting 6 monthsor more after the loss. These symptoms of CG occur in asubgroup of 10% to 20% of bereaved individuals (Middletonet al.,1996), and have proven to be distinct from depressive andanxiety symptom clusters (Prigerson et al., 1995a, 1996;Ogrodniczuket al., 2003). Specific psychotherapeutic treatmentimproves complicated grief (Shear et al., 2005; Wagner et al.,2006), and pharmacologically treatment of bereavement-related depression has a beneficial influence on grief sympto-matology (Hensley et al., 2008). Although we know thatstressful life events such as the loss of a close family membercan be a risk factor for unipolar depression (Li et al., 2005), todate little is known regarding the prevalence or outcome ofbereavement in patients with unipolar depression. Piper et al.(2001) examined the prevalence of complicated grief amongoutpatients of two psychiatric clinics. Psychiatric diagnosescomprised affective disorders (79%) and substanceabuse (15%),with further 6% not being specified. About one third of all

202 A. Kersting et al. / Journal of Affective Disorders 118 (2009) 201–204

patients met the criteria for either moderate or severecomplicated grief. However, Piper et al. failed to detect aspecific relation of CG to any subgroup of specific psychiatricdiagnoses. In a larger study of predictors of suicidality inpatients with bipolar disorder Simon et al. (2005) investigatedthe frequency and implications of the loss of a close familymember. A lifetime history of significant loss was reported by86% of participants with 24% of those meeting the criteria forcomplicated grief.

The aim of the present study was to investigate theprevalence of complicated grief for the first time in psychiatricinpatients with a unipolar depression. Further, we wanted toidentify influential characteristics predicting a complicatedgrief reaction.

2. Methods

2.1. Subjects

Seventy-three inpatients recruited between June 2004 andMay 2005 in the department of psychiatry, University Muen-ster, Germany, and diagnosed with unipolar affective disorderconstituted the study sample. Five patients refused to partici-pate.Meanage of the total study samplewas46.7 (S.D.=16.17),47 patients (64%) were female and 26 (36%) patient partici-pants were male.

This investigation was part of a larger study examiningdeterminants and predictors of affective disorders (Baune et al.,2008). All patients provided written informed consent beforeparticipation. The studywas approved of by the respective localethical committees and has therefore been performed inaccordance with the Declaration of Helsinki ethical standards.

2.2. Measures

Psychiatric diagnosis was established using the StructuredClinical Interview forDSM-IV—patientedition (SCID, First et al.,1996) administered by trained raters. Complicated grief wasassessed using the Inventory of Complicated Grief (ICG:Prigerson et al., 1995b), a scale that assesses a wide range ofcognitions, emotions, and behaviours that define CG and

Table 1Complicated grief and its clinical correlates in patients with unipolar depression.

Complicated grief(N=13)

Non-compli(N=57)

Close family member 92% 50%Impairment today 3.0 1.23 1.3Impairment at loss 4.1 1.24 2.6Grief at loss 4.9 0.29 4.4BDI 29.1 11.60 20.8IES-r Total 41.5 10.19 17.2IES Intrusion 15.2 3.30 6.9IES Avoidance 12.3 4.17 5.3IES Hyperarousal 14.0 3.85 5.0BSI.GSI 1.7 .76 1.2Stai state 56.4 6.89 51.6Stai trait 49.0 8.58 46.2F-Sozu 3.5 .87 3.6

aPearson chi-square, 2×2 table, df=1, bMann–Whitney U-test, ctwo-tailed T-test.⁎⁎Pb0.01; ⁎Pb0.05.

provides good psychometric properties (Cronbach's alpha=0.94, test–retest reliability 0.80; Prigerson et al., 1995b).Following Prigerson et al. (1995b), patients with complicatedgrief were identified as those subjects constituting the upper20% in ICG total score (ICG-score≥18). Psychological stressreaction after the loss was assessed using the Impact-of-Event-Scale — Revised (IES-R) comprising the three subscalesintrusion, avoidance and hyperarousal (Weiss and Marmar,1996; Horowitz et al., 1979). Severity of depressive symptomswasmeasured using thewell validated 21-item self-report BeckDepression Inventory (BDI: Beck et al., 1961, 1988). Further,anxiety (State-Trait Anxiety Inventory, STAI: Spielberger,1983),overall mental health (Brief Symptom Inventory BSI providingthe Global Severity Index (GSI): Derogatis, 1993), and socialsupport (F-SozU, Short form K-22: Dunkel et al., 2005) wereevaluated.

Demographic characteristics, loss-specific information(relationship to the lost person, time since loss, age at loss)were recorded. Intensity of religious faith, of impairment afterthe loss today and, retrospectively, intensity of grief andimpairment shortly after the loss were judged on five-pointLikert scales.

2.3. Statistics

Univariate analyses on categorical measures were done byChi-square tests or Fisher exact test (FET). Group differenceson continuous variables were analyzed with Mann–WhitneyU-tests or 2-sided T-tests. Variables with a relevant (Pb0.15)association with severity of grief were included into ahierarchical regression analysis employing a stepwise regres-sion procedure to explore the best predicting power forseverity of grief. All P values are 2-tailed, and statisticalsignificance was set at the 5% level (Pb0.05). Analysis wasperformed by SPSS (Version 14.0.1 for Windows).

3. Results

Ninety-six percent of the study sample reported the loss of aclose person (n=70). The most meaningful loss occurred 16.4(SD=14.10) years prior to timeof inclusion in thepresent study

cated grief Test-statistic df P

7.751a 1 0.005⁎⁎0.78 77.500b b0.001⁎⁎1.36 153.000b 0.002⁎⁎0.56 256.000b 0.083, n.s.

10.45 2.540c 68 0.013⁎14.96 5.574c 68 b0.001⁎⁎4.86 5.877c 68 b0.001⁎⁎5.92 4.047c 68 b0.001⁎⁎5.30 5.772c 68 b0.001⁎⁎0.62 2.049c 68 0.04⁎11.91 1.403c 68 0.165, n.s.11.31 0.829c 68 0.410, n.s.0.84 0.384c 68 0.703, n.s.

Table 2Stepwise regression analysis of the independent variables on severity of grief(N=70).

Model Variable B SEB β R2 ΔR2

1 (Constant) −0.490 1.503 (n.s.) 0.514 0.514IES-R 0.451 0.054 0.717⁎⁎

2 (Constant) −5.456 2.131 ⁎ 0.578 0.064IES-R 0.428 0.052 0.681⁎⁎BDI 0.244 0.078 0.255⁎⁎

3 (Constant) −7.726 2.203 ⁎⁎ 0.621 0.043IES-R 0389 0.051 0.619⁎⁎BDI 0265 0.075 0.277⁎⁎Significance of loss 4.679 1.742 0.217⁎⁎

⁎⁎Pb0.01, ⁎Pb0.05: IES-R: Impact of Event Scale, revised version; BDI BeckDepression Inventory; adjusted R2 for the final model: 0.603.

203A. Kersting et al. / Journal of Affective Disorders 118 (2009) 201–204

at amean ageof 31.4 (SD=16.91) years. In 93%, deathsweredueto natural causes (29% suddendeath, 64%death after a long termillness), in 7% of the cases, the loss was due to traumaticcircumstances (homicide/accident/suicide). Fifty-eight percenthad lost a close family member (48.0% parent, 4% spouse, 4%sibling, 2% child), 38% other relatives and 4% good friends. Griefscore ranged from 0 to 43 (mean 9.4, S.D.=10.81). Complicatedgrief was diagnosed in thosewith an ICG score above the cut-offpoint (CG: n=13, ICG=28.8, S.D.=7.93, Non-CG: n=57,ICG=5.0 S.D.=4.81). Patients with CG and patients withoutCG did not differ regarding age (T=1.183, df=68, P=0.341, n.s.)or sex (Chi2=0.170, df=1, P=0.680, n.s.). No significantdifferences between the groups were found for time since loss(T=1.028, df=68, Pb0.308, n.s.), age at loss (T=1.980, df=67,P=0.052, n.s.) or conditions of loss (Chi2=4.351, df=2,P=0.114, n.s.). However, patients with CG more frequently losta close family member (parent, sibling, spouse) and reported ahigher impairment at time of testing and, retrospectively judged,a higher impairment shortly after the loss. No difference wasfound according to retrospectively judged intensity of griefshortly after the loss (see Table 1). Depressed patients with CGshowedhigherdepressive symptoms(BDI), higherposttraumaticstress concerning the significant loss and a significantly higheroverall level of psychopathological symptom severity thandepressed patients without CG (see Table 1).

Finally, variables with a relevant (Pb0.15) associationwithseverity of grief (Time since loss r=−0.182, P=0.133; age atloss r=0.206, P=0.090; grief at loss r=0.406, P=0.001;impairment at loss r=0.547, Pb0.001; significance of lossr=0.365, P=0.002; posttraumatic stress r=0.726,Pb0.001; severity of depression r=0.364, P=0.002; overallmental health r=0.363, P=0.002) were included into ahierarchical regression analysis to examine predictor vari-ables for complicated grief. Default values were PIN=0.05and POUT=0.10. Higher posttraumatic stress, higher depres-sion and close family-membership best predicted severity ofgrief explaining 60% of its variability (see Table 2).

4. Discussion

To the best of our knowledge this study is the firstinvestigation that was set out to examine the prevalence ofloss and complicated grief in a sample of psychiatricinpatients with unipolar depression. The most striking resultof our study was that patients with complicated grief weresignificantly more depressed, showed significantly higher

symptoms of posttraumatic stress and higher levels ofpsychopathological symptom severity compared to depressedinpatients without complicated grief. Our findings extendedthe results of Piper et al. (2001) who found that psychiatricoutpatients with severe complicated grief had significantlyhigher levels of depression, anxiety and general symptomaticdistress compared to psychiatric outpatients without com-plicated grief. Piper et al. investigated a heterogeneouspsychiatric sample using a compound of related character-istics for the assessment of grief. In our study a homogeneoussample of unipolar depressed patients was investigated usingstandardized measuring instruments for the diagnoses ofcomplicated grief. In contrast to the findings of studies inpsychiatric outpatients (Piper et al., 2001) and bipolarpatients (Simon et al., 2005), in our study unipolar depressedpatients with complicated grief did not show significantlyhigher levels of anxiety and the extent of social supportcompared to depressed patients without complicated grief.

Unipolar depressed patients in our study showed a highprevalence of loss (96%), which is comparable to a sample ofpatients with bipolar disorders (86% Simon et al., 2005). Fifty-eight percent of the patients in our study lost a close familymember, comparable to 55% in general psychiatric outpatients(Piper et al., 2001). Compared to other studies examining griefinpsychiatric patients, the event of loss dated back longer in ourstudy (15.6±13.3 years) compared to 12.3±11.3 years (Simonet al., 2005), 9.7±10.2 years (Ogrodniczuk et al., 2003) and10.4±9.7 years (Piper et al., 2001). The high prevalence oflosses in our study could be due to the higher age of our studysample (46.7±16.2 years) compared to 42.2±11.9 years inthe study of psychiatric outpatients (Piper et al., 2001) and to44.1±13.3 years in the sample of patientswith bipolar disorder(Simon et al., 2005). Higher age naturally elevates the risk toexperience the loss of a person.

In contrast to bipolar patients (Simon et al., 2005), inunipolar depressed inpatients the nature of relationship to thelost person influenced development of CG. In contrast tofindings of Simon et al. patients with complicated grief did nothavemore comorbiddiagnoses than thosewithout complicatedgrief.

Some limitations of the study should also be noted. Sincepatients were asked to recall their previous traumatic loss,recall and suspicion bias confounding the present resultscannot be ruled out. Moreover, since presently no controlgroupwas investigated, themessage of thepresent studywith apatient-only design is limited to the notion of complicated griefas a potential risk factor for greater severity and poorer functionin unipolar depressed patients. Longitudinal studies compara-tively analyzing patient and control groups will additionallyallow for estimation of complicated grief as a risk factor fordepression. In this context it has to be considered that thefactors that increase vulnerability to unipolar depression mayalso increase vulnerability to bereavement.

In summary the present results for the first time providepreliminary support for a potential role of comorbid compli-catedgrief as a risk factorof greater severityandpoorer functionin unipolar depressed patients. Clinicians should assessdepressive patients for loss and complicated grief on a routinebasis. Those with ongoing complicated grief should be treatedusing specific psychotherapeutic concepts (Shear et al., 2005;Wagner et al., 2006).

204 A. Kersting et al. / Journal of Affective Disorders 118 (2009) 201–204

Role of funding sourceNothing declared.

Conflict of interestNo conflict declared.

Acknowledgement

We would like to express our appreciation to the patientswho participated in the present study.

References

Baune, B.T., Hohoff, C., Berger, K., Neumann, A., Mortensen, S., Roehrs, T.,Deckert, J., Arolt, V., Domschke, K., 2008. Association of the COMTval158met variant with antidepressant treatment response in majordepression. Neuropsychopharmacology 33, 924–932.

Beck, A.T., Ward, C.H., Mendelson, M., 1961. An inventory for measuringdepression. Arch. Gen. Psychiatry 4, 561–571.

Beck, A.T., Steer, R.A., Garbin, M.G., 1988. Psychometric properties of the BeckDepression Inventory: twenty-five years of evaluation. Clin. Psychol. Rev.8, 77–100.

Boelen, P.A., van den Bout, J., 2008. Complicated grief and uncomplicatedgrief are distinguishable constructs. Psychiatry Res. 157, 311–314.

Derogatis, L.R., 1993. Brief Symptom Inventory (BSI), Administration, Scoring,and ProceduresManual, third ed. National Computer Services,Minneapolis.

Dunkel, D., Antretter, E., Fröhlich-Walser, S., Haring, C., 2005. [Evaluation ofthe short-form social support questionnaire (SOZU-K-22) in clinical andnon-clinical samples] [Article in German]. Psychother. Psychosom. Med.Psychol. 55, 266–277.

First, M.B., Gibbon, M., Spitzer, R.L., Williams, J.B.W., 1996. User'. s Guide forthe Structured Clinical Interview for DSM-IV Axis 1 Disorders— ResearchVersion. American Psychiatric Press, Washington, DC.

Hensley, P.L., 2006. Treatment of bereavement-related depression andtraumatic grief. J. Affect. Disord. 92, 117–124.

Hensley, P.L., Slonimski, C.K., Uhlenhuth, E.H., Clayton, P.J., 2008. Escitalopram:an open-label study of bereavement-related depression and grief. J. Affect.Disord. Jul 1.

Horowitz, M.J., Wilner, N., Alvarez, W., 1979. Impact of Event Scale: a measureof subjective stress. Psychosom. Med. 41, 209–218.

Li, J., Precht, D.H., Mortensen, P.B., Olsen, J., 2003. Mortality in parents afterdeath of a child in Denmark: a nationwide follow-up study. Lancet 361,363–367.

Li, J., Laursen, T.M., Precht, D.H., Olsen, J., Mortensen, P.B., 2005. Hospitalizationfor mental illness among parents after the death of a child. N. Engl. J. Med.352, 1190–1196.

Lichtenstein, P., Gatz, M., Berg, S., 1998. A twin study of mortality after spousalbereavement. Psychol. Med. 28, 635–643.

Middleton, W., Burnett, P., Raphael, B., Martinek, N., 1996. The bereavementresponse: a cluster analysis. Br. J. Psychiatry 169, 167–171.

Ogrodniczuk, J.S., Piper, W.E., Joyce, A.S., Weideman, R., McCallum, M., Azim,H.F., Rosie, J.S., 2003. Differentiating symptoms of complicated grief anddepression among psychiatric outpatients. Can. J. Psychiatry 48, 87–93.

Piper, W.E., Ogrodniczuk, J.S., Azim, H.F., Weideman, R., 2001. Prevalence ofloss and complicated grief among psychiatric outpatients. Psychiatr. Serv.52, 1069–1074.

Prigerson, H.G., Frank, E., Kasl, S.V., Reynolds, C.F., III, Anderson, B., Zubenko,G.S., Houck, P.R., George, C.J., Kupfer, D.J., 1995a. Complicated grief andbereavement-related depression as distinct disorders: preliminaryempirical validation in elderly bereaved spouses. Am. J. Psychiatry 152,22–30.

Prigerson, H.G.,Maciejewski, P.K., Reynolds, C.F., III, Bierhals, A.J., Newsom, J.T.,Fasiczka, A., Frank, E., Doman, J., Miller, M., 1995b. Inventory ofComplicated Grief: a scale to measure maladaptive symptoms of loss.Psychiatry Res. 59, 65–79.

Prigerson, H.G., Bierhals, A.J., Kasl, S.V., Reynolds, C.F., III, Shear, M.K., Newsom, J.T.,Jacobs, S., 1996. Complicated grief as a disorder distinct from bereavement-related depression and anxiety: a replication study. Am. J. Psychiatry 153,1484–1486.

Shear, K., Frank, E., Houck, P.R., Reynolds, C.F., III, 2005. Treatment ofcomplicated grief: a randomized controlled trial. JAMA 293, 2601–2608.

Simon, N.M., Pollack, M.H., Fischmann, D., Perlman, C.A., Muriel, A.C., Moore,C.W., Nierenberg, A.A., Shear, M.K., 2005. Complicated grief and itscorrelates in patients with bipolar disorder. J. Clin. Psychiatry 66,1105–1110.

Spielberger, C.D., 1983. Manual for the State-Trait Anxiety Inventory: STAI.Consulting Psychologists Press, Palo Alto, CA.

Stroebe, M., Schut, H., Stroebe, W., 2007. Health outcomes of bereavement.Lancet 370, 1960–1973.

Szanto, K., Shear, M.K., Houck, P.R., Reynolds, C.F., III, Frank, E., Caroff, K.,Silowash, R., 2006. Indirect self-destructive behavior and overt suicidalityin patients with complicated grief. Clin. Psychiatr. 67, 233–239.

Tomassini, C., Rosina, A., Billari, F.C., Skytthe, A., Christensen, K., 2002. Theeffect of losing the twin and losing the partner on mortality. Twin Res. 5,210–217.

Wagner, B., Knaevelsrud, C., Maercker, A., 2006. Internet-based cognitive–behavioral therapy for complicated grief: a randomized controlled trial.Death Stud. 30, 429–453.

Weiss, D.S., Marmar, C.R.,1996. The Impact of Event Scale— Revised. In:Wilson,J.P., Keane, T.M. (Eds.), Assessing Psychological Trauma and PTSD: AHandbook for Practitioners. Guilford Press, New York, pp. 399–411.