Him ORIGINAL CONTRIBUTION An Empirical Examination of the Stage Theory of Grief susan D. Bloek, MD Hol IE NOTION THAT A NATURALPSY- chological response to loss involves an orderly progres- sion through distinct stages of bereavement has been widely accepted by clinicians and the general publi Bowlby and Parkes! were the iso pr pose a stage theory of grief for adjust ment to bereavement that included 4 stages: shock-numbness, yearning- searching, disorganization-despair, and reorganization. Kubler-Ross? adapted Bowlby and Parkes theory to describe a S-stage response of terminally illpaticnts to awareness of their impending death: denia-dissociationisolation, anger bar- gaining, depression, and acceplance. The stage theory of grief became well- known and accepted, and hasbeen gen- crlized oawide variety oflosses includ ing children’s reactions to parental separation,’ adults reactions to marital separation,’ an linia stalls reactions tothedeath ofan inpatient A 1997 sure vey conducted by Dovne-Wamboldt and Tamlyn® documented the heavy relic ance of medical education on the Kubler Ross model of grel. The National Can- cer Institute currently maintainsa Web siteon loss, geet, andl bereavement that describes the phases of grief.” As entrenched asthenotion of phases of grief may be, the hypothesized sequence of get reactions has previously not been snvestigated empirically Several bereavement scholars have in- vestigated particular aspects of, oF diae szammed changesin,grie reactions over {796 JAMA, Fey 21,200 Context The stage theory of grief remains a widely accepted model of bereavement adjustment stil taught in medial schools, espoused by physians, and applied in d- ‘essen Ntethkss the sage Reo of ge fis evn beat tes empirical Objective To examine the relative magnitudes and patterns of change overtime post los of 5 gre indeators for consistency wih the stage theory of grt Design, Setting, and Participants. Longitudinal cohort study (ale Bereavement Study) of 233 bereaved inviduals ving n Connecticut, with data collected between January 2000 and January 2003 Main Outcome Measures. Fiverater-adminstered tems asesing dabei, yeam- ing, anger, depression, and acceptance af the death from 1 to 24 months posts Results Counter to stage theory, dsbelef was not the initial, dominant gre nd cator Acceptance was the mos frequently endorsed em and yearming waste dom rant negatve grief indicator from 1 to 24 months poslos. In mode: that take into account the rse and fall of psychological responses, once rescaled, dibelie de- creased fom an nal igh af 1 month posts, yearing peaked at 4 months post loss, anger peaked a5 months polos, and depression peaked at6 months poss. Acceptance increased throughout the study observation period. The gre indeators achieved ther respective maximum value inthe sequence (sbele, yearning, an- fer, depression, and acceptance) predicted by the stage theory of gi Conclusions dentition ofthe normal stages of gr folowing a death from natu- flcases enhances understanding of how the average person cogil nd em tonal processes the loss fa farm member ive tal the negative grinders all peak within approximately 6 months poss, the who score high on these in- dichtors beyond 6 months pastloss might bene from furler evaluation. saa aoraarne na wacom time-!® Bonanno etal!" found 5 diver- _(yearning-anger-anxiety), depression gent grieving trajectories from prelossto mourning, and recovery. To date, no 18 months postloss (common grief, study has explicitly tested whether the chronic grief, chronic depression, im- normal course of adjustment toa nati provement during bereavement, and re- ral death progresses through stages of Silence). Wortman and Silver exame ‘abs Allon: Depart Poca, Wor ined and disproved the necessity of 1 Sheath esuth and Magee Rea fe stage inthe grief theory when they found — sah Cater ae Uresty Stet Means New seat! when they found eer Con Hasse en atdepresionwasnotaninevablere Qa fate Caen eat rere | Sick and Pagan and Oeparment ot eit Packs asd Kabler Rose theories J Ethan and Wones Hosp aod Hara ed cobs" synthesized and illustrated the hy- cal School Center for Paiaive Care Boston, Mas (Drs Seana rcesen poesia theory of gic nich Sed kaha sy gn. Ce the normal response to loss progresses tee Peke-Onslony tn Pale Cae fe tenor et Pere tame ete Sey Sh Etiam mk 02113 ly eset hnumbnese-disbelef separation distress anudeds No.7 (Reprinted) (©2007 American Medical Association. All ights reserved.disbelief, yearning, anger, depression, and acceptance The identification of the patterns of \ypical grief symptom trajectories is of clinical interest because it enhances the understanding of how individuals cog- nitively and emotionally process the death of someone close. Such knowl- edge aids in the determination of whether a specific pattern of be ment adjustment is normal or not, Once the normal patterns of griefare known, individuals with abnormal bereave- ment adjustment can be identified and referred for treatment when indicated. This study used data froma sample of community-based bereaved individu- als to examine the course of disbelief, yearning, anger, depression, and accep- lance as described by Jacobs" from L 10 24 months postloss, FIGURE 1 illus- trates the hypothesized sequence of stages of grief for this analysis. Be- cause approximately 94% of US deaths result from natural causes (eg, vehicle crashes, suicide), deaths from unnatu- ral causes (eg, car crashes, suicide) were excluded thereby enabling the results to be generalized to the most common lypes of deaths. Individuals who met the criteria for complicated grief disor- der'®” also were excluded so that the results would represent normal be- reaverent reactions, Although the pro- posed stage theory of grief" does not specily the precise timing of the stages, Jacobs" described the normal griev- ing process and each of is stages as being completed within 6 months fol- lowing the loss of a loved one. How- ever, in the absence of an established, ‘empirical foundation for the length of lime associated with the normal griev- ing process, the normal grieving pro- ‘cess was not assumed to be limited to 66 months postloss in this study. In- stead, the grief indicators were exam- ined as functions of ime up to 24 months postloss, METHODS: ‘study Sample The Yale Bereavement Study, a longitu- dinal examination of grief in a comrn- nity-based sample of bereaved individ (©2007 American Medical Assoc 1, All rights reserved. Figure 1: Hypotiedzed Stage Theory oF ner bkcator Rating Tine als, collected data between January 2000 and January 2003, and was funded by the National Institute of Mental Health. For greater Bridgeport/Fairfield, Conn, the names of the newly bereaved months) were obtained from the div sion of the American Association of Re lured Persons Widowed Persons Ser- vice, a community-based outreach program. For the New Haven, Conn, metropolitan and surrounding areas, names were obtained from obituaries listed in the New Haven Register, through newspaper advertisements, fliers, per- sonal referrals, and referrals from the chaplain’s office ofthe St Raphael Hos- pital. A comparison hetween greater Bridgeport/ Fairfield Bureau of Vital Rec- ords death certificates and the Wid- cowed Persons Service list during the same 3-month period revealed that the Wid lowed Persons Service listings captured 19594 ofall deaths leaving behind a wid cowed individual, suggesting thatthe list- ing provided an unbiased and compre hensive ascertainment of recently widowed individuals in the sampled re- gion. Participants recruited from greater New Haven (37.0%) did not differ sig- nificantly from participants recruited from greater Bridgeport/ Fairfield (©3.0%) with respect sex, income, edu- cation, race/ethnicity, or quality of life. Participants reeruited Irom greater New Haven were significantly younger (mean [SD] age, 59.7 [16-4] years) than par- ‘STAGE THEORY OF GRIEF From Loss Licipants from Bridgeport/ Fairfield. (mean [SD], 63.2 [11.5] years) (P=.05). The institutional review boards of ll pat= Licipating sites approved the research, protocol. Individuals were invited to participate Inthestudy viaaletter that described how their names were obtained, identified the investigators, outlined the aims and pro- cedures, and noted that they would be contacted by study staff in the following ‘week unless they informed us oftheir wish not to be contacted. Of the 575 persons contacted, 317 (55.1%) agreed to part pate, Reasons for nonparticipation i chided reluetance to participate in research (n=11; 43%); being to0 busy (n=46; 17.896); being t00 upset (n=27; 10.5%), “doing fine” (n=23;8.%); not being terested or having no reason (n=145; 50.2%);and having other reasons (n=6; 2.3%). Compared with participants, non- participants weresigificantly morelikely tobe male (25.0% vs37.2%; P<.001) and colder (mean [SD] age, 61-7 [13.1] years vys 68.8 [13.7] years) (P<001). Non English-speaking personsand those con- sidered too frail complete theinterview ‘wereineligible. The 317 parieipants were interviewed ata mean (SD) of 6.3 (7.0) monthsalterthedeath ofalovedone. The frst follow-up interview (n=296;03 4%) ‘was completed atamean (SD) f109(6.1) months postloss; second follow-up inter view (n=263; 83.0%) ata mean (SD) of 19.7 (5.8) months postloss. Written i (Reprinted) JAMA, Februry 21,2007 Vol 297, No.7 717‘STAGE THEORY OF GRIEF Table 1. Demograpine Varabier of the Yale Bereavement Stay sample Compared Win 2005 Data From the US Census Yale Bereavement ‘Study Sample 2005 US (= 289).No. (4) Gensus, % 125 (635) 15 (772 ear 26 7.) Education beyond high sehooT 15 (22 SAT Fosian rousehola income, S E2000 ELITE Measures of Grief The indicators of dishelif, yearning, anger, and acceptance ofthe death were assessed using single items obtained from the rater-adininistered version of the Inventory of Complicated Grief- Revised, formerly known as the Trau- matic Grief Response to Loss." Although it would have been prefer- able to use separatescales for the assess- ment of yearning, disbelief, anger, and acceptance of the death, no such scales exist for each of these grief stages. To sures, single items were used forall grief phase indicators, Single-item inte view screenings have proven remark- ably accurate in the prediction of depres- sion. The frequency, rather than severity, ofeach griefindicator was used as the response format in the Inven- tory of Complicated Grief-Revised because frequency has proven to be a more effective means of evaluating the impact of events.” Grief phase indica ors were measured using a 5-point Likert scalein which 1 equaled less than, once per month: 2, monthly; 3, weekly 4, dally; and 5, several times per day. These items showed moderately high correlations with the total Inventory of Complicated Grief-Revised score at baseline interview, which ranged in magnitude from 0.47 to 0.57 (all com- parisons yielded P<.001). Toenhance comparability in the measurement of cach indicator, depression was assessed using the single-item depressed mood inthe Hamilton Rating Scale for Depres- sion.” The correlation between depressed mood and the total Hamil- ton Rating Seale for Depression score formedconsent was obtained fromallin- dividuals enrolled in the study. Of the 317 individuals identified, 58 were excluded because they met criteria for complicated grief disor- der, 19 because they survived trau- matic deaths, and 14 because they hhad missing data on examined mea- sures. The study sample (N=233) consisted of individuals who did not meet eriteria for complicated grief disorder'*"” during the study; had a family member or loved one who died [rom natural not traumatic causes; and had at least 1 complet assessment of the 5 grief indicators included in the stage theory of gris ‘within 24 months postloss. The par- cipants were significantly older (mean [SD] age, 62.9 [13.1] years: 53.5% aged =05 years) and more likely to be white (97.0%) than the excluded individuals (mean [SD] age, 58.5 [15.0] years; 90.4% white) but did not significantly differ with respect (0 sex, income, education, and relationship to the deceased. The vast majority of participants were spouses of the deceased (83.8%). The remaining participants (16.2%) were adult children, parents, or siblings of the deceased, The data from the participants were compared with data [rom the 2005 US Census (Taste 1)."* Compared with the Uswidowed population, the study par- Lcipants were younger, more likely tobe male, and ahigher proportion were white Compared with the US general popula tionaged 25 yearsorolder, thestudy par- Licipants werebetter educated and hada higher median household income. 798 JAMA, Feb 21,200 yo 297, No. 7 Reprinted athaseline interview was0.65 (P<.001) To be consistent with the scale levels of other grief indicators, all levels of depressed mood were increased by 150 that 1 indicated “absence of depressed mood” and 5 indicated “patient reports virally only these feeling states in his spontaneous verbal and non-verbal Individuals selt-idenufied their rcial/ cethinic status according to the racial/ ethnic eategories defined in the US Cen- sus." They also reported the cause of death for the family member or loved cone, For deaths due to a terminal ill ness, the date of the diagnosis was r corded. Diagnoses of the terminal ill ness within 6 months (52/109; 26.1%) ‘were compared with those 6 months or longer (147/199; 73.9%) prior to the death. Six months was used as the threshold because terminal diagnoses of less than 6 months resulted in smaller, less reliable groupings and else- where!” 1 has been determined that ‘6 months is the time alter which nor- mal grief ean be distinguished from complicated grief disorder. Statistical Analyses Statistical analyses were conducted to lest for significant differences in the magnitude of each of the 5 grief indi- ceators within each of the 3 postloss pe- ods (6 months [1-6 months cat- egory], >6 to =12 months [6-12 months category], and >12 to =24 months [12-24 months category]); 10 ‘compare the pattern of changes in the absolute levels of each ofthe 5 grief in- dicators over time; and to determine when each of the 5 grief indicators achieved its maximum value. Specifically, single-sample (testsand nonlinear, ordinary least squares re- {gression analyses were used to exam- ine the differences in magnitude he- tween grief indicators at a given time postloss and changes in grief indica lors as @ function of time postloss. Single-sample ¢ tests were used (0 ex- amine within-person differences in magnitude between the 5 grief indica tors postloss at to 6 months, 6 to 12 months, and 12 to 24 months and (©2007 American Medical Association, All rights reserved.‘STAGE THEORY OF GRIEF ‘Table 2, Gret inciestors Asersed Dung 3 Porfos: Penode Poriod of Posioss Assessment mo) 16 2 1a.28 rit No.of Mean No.of Mean No. of Mean Indicator? Parielpants (0) Participants (0) Participants (80) Diabet 77 Zari, Zit 807.05) 205, Tar Os: Yearning Wi Ir 2 S819) 205, 2a (12, ager TE TET TOL 2 TBOT-Oe) 25, 15 08 Depresson We 22025) 2 225 TAT 25, TaD Ta Recep Tas aT TT 20 £29 0.03), 205) 70 08} within-person temporal changes in magnitude of each grief indicator post- loss between 1 to months and 6 to 12 months and hetween 6 to 12 months and 12 to 24 months. Nonlinear, ordinary least squares re- gression analyses were used to model the trajectory of each grief indicator as ‘function of time postloss. Because the stage theory of grief predicts the se- {quential rise and fall of each ofthe grief indicators as a function of time post- loss (ie, phase), we chose the follow- ing parametric functional form that ‘would capture such phases: Yo[A+ BOUT + 1] exp-Yatlt) + € where ¥ represents the value of the grief indicator and the term Ut represents lume postloss with t sealed by the model parameter £, The expression [A + B (Cut-+ D] expVa/o) represents, line ear combination of normalized (weighted) zero-order and first-order Laguarre polynomials, scaled by the model parameters A and B, respec Lively, ineluded to capture the anticl- pated rise and fall in the data. Model parameter C represents the asymp- totic value that the grief indieator ap- proaches as time postloss increases to infinity. One observation per person (N=233), selected randomly among, those observations that contained com= plete data for each of the 5 grief indi- cators within 24 months postloss, was used to fit these regression models. For each grief indicator, the model para- meters t, A, B,and Cwere estimated by means of nonlinear, ordinary least squares regression implemented using PROC MODEL in SAS version 9.1 (SAS (©2007 American Medical Assoc ton, Al rights reserved. Institute Ine, Cary, NC). P<.05 was considered significant. A series of multivariable analyses of variance were conducted to evaluate whether demographic variables and re- port of diagnosis of terminal illness \within 6 months of the death were s nificantly related to the 5 grief indica tors of to the within-person diffe ences between of temporal changes in the 5 grief indictors, RESULTS The means and SDs for the 5 gr dicators of disbelief, yearning, anger, de- pression, and acceptance postloss at 1 to 6 months, 6 to 12 months, and 12 to24 months appearin Tapte 2, Within, cach period, acceptance is greater than disbelief, yearning, anger, and depres- sion; yearning is greater than disbe- lief, anger, and depression; and depres- sion is greater than anger. Between 1 and 6 months postloss and 6 and 12 months postloss, disbelief and yearn- ing decline and acceptance increases. From 6 to 12 months postloss and 12, to 24 months postloss, disbelief, yearn- ing, anger, and depression decline and acceptance increases. More specifically, acceptance is si nificantly greater than disbelief (1-6 months postloss:(,4.=10.79, P<.001: 6-12 months postloss: t4o.=23.16, P<.001; 12-24 months postloss: tayj= 31.88, P<001), yearning (1-6 months postloss:f4.-2.11, P=.04; 6-12 months postloss: ljq.=10:80, P<.001: 12-24 months postloss: (95=19.39, P<.001), anger (1-6 months postloss: 4:=12.66, P<001; 6-12 months post- loss: tyy=23.14, P<001; 12-24 months postloss: fn4=35.24, P<.001), and de- pression (1-6 months postloss: Luu= 11.64, P<.001; 6-12 months post- loss: f.=18.84, P< 001; 12-24 months postloss:t59;=29.97, P=,001) Yearning issignificantly greater than, disbelief (1-6 months postloss: fo" 13.57, P=.001; 6-12 months post- loss: (5-15.57, P<001; 12-24 months postloss: t4.4=12-49, P<.001), anger (2-6 months postloss: t,;0=16.43, P<.001; 6-12 months postloss: 19=15.10, P<.001; 12-24 months post- loss: tny= 12.43, P.001), and depre sion (1-6 months postloss: (;5= 1440, P<.001; 6-12 months postloss: 75, P<.001; 12-24 months post- ‘041, P<.001). Depression is significantly greater than anger (1-6 months postloss: 1123.61, P<.001; 6-12 months post- loss: {= 5.32, P<001; 12-24 months postloss: t50'=3.16, P=.002). Disbelief is significantly greater than anger at 1 to 6 months postloss (in1=3.22, P=.002); depression is sig- nificantly greater than disbelief at 6 10 12 months postloss(t,5=5.22, P<.001) and at 12 to 24 months postloss (y=2-19, P=03) Between Land 6 months postlossand 6 and 12 months postloss, disbelief (s:=4.78, P<.001) and yearning (i5o27.89, P< 001) decline and accep- lance increases (tj0=3.91, P<.001). Be tween 6 and 12 months postloss and 12 and 24 months postloss, disbelief ((o=2.84, P=.005), yearning (y)=5.96, P<.001), anger (l)y:=3.91, P<.001), and depression (tyy:=5.60, P<.001) decline and acceptance increases (y23.37, PX.001). (Reprinted) ANA, Fsbnunty 21,2007 Vol 297, No.7 79‘STAGE THEORY OF GRIEF Ficure 2 displays the results of the tween 1 and 4 months postloss, de- nonlinear regression analyses. Accord- creases between 4 and 24 months post- ing to the models displayed in the top loss, andis greater than disbelief, anger, part of Figure 2, acceptance increases and depression between 1 and 24 monotonically (uniformly in 1 diree- months postloss. Disbelief decreases tion), and is greater than each of the monotonically between 1 and 24 other grief indicators between land 24 months, is greater than anger between, months postloss. Yearning increases be- 1 and 6 months postloss, and is greater Figure 2. Empiial and Rescaled Model for iit Indicators as Functions of Time (rie nestor | Aescaled incite Rating “ina Fem Loss, mo “Top, The carves eprezel pet indeator a unos of tne baad on nonneareesion mages etimated ‘rom the data (2253). Te dala mater long the ans are cetemaned by the mean value of be Fore Testor the indus incude in the 10 groups of senators (n= 73 abseratons pe group for ups 6 observations for 1 oup). The caresponding eer bas ndeate SDs. These 10 ops ebstvabans ‘ere fomed by exening alo he cbservaons use Inte regression analyses (N=255)by ceasing te Poses: (bseraon that occured a the same tne posts wee randomly signed a peston ne o (ered sequencea servations that ne), ad then sstnng the st23cbsevars on ts odeed ist {othe iat group the next 23 ebservains tothe second group, te The regression cuwes we based onthe ‘alyss of mind data pnts (N=233) fr which time fom ls aes fom 15 to 23 moths Bottom, The ‘ves rspresen gre indators uncon of te bed on nein epeson rade athe loving Feseaing procedure (10 ~ Yd! a Yo) here YO ste ade vale forthe get dest at tne [Rand Yecand Yo ae the minimum and maar mode! ues ofthe gre nakeator respecte, beeen ‘and 27 month porto. The greta: achieve th respcive masimu values ine enats quence sett, yearning anger depesson, and acceptance) preted by he hyplheszed Meo of het Proceed Figure than depression between 1 and 4 months postloss. Depression in- creases between 1 and 6 months post- loss, decreases between 6 and 24 months, is greater than disbelief post- loss between 4 and 24 months, and is jgreater than anger between 1 and 24 months postloss. Anger increases be- tween 1 and 5 months postloss and di creases between 5 and 24 months post- loss. The close agreement between the models and the data in the top part of igure 2 indicates that the phasic func ional form specified in the regression models adequately represent the data, The bottom part of Figure 2 dis- plays the regression models following a rescaling procedure that constrains each fgiet indicator to fall within the inter- val of O through 1, In the top part of Figure 2, the relative locations in time ‘of the peaks of the grief indicators are ‘obscured because the curves are not side by side, thereby making comparisons difficult. Those compatisons ae faci tated in the bottom part of Figure 2 by placing all ofthe indicators on the same scale. The 5 grief indicators achieved their respective maximum values in the sequence (disbelief, earning, anger, d pression, and acceptance) predicted by the stage theory of grief. Given that there are 120 possible sequences of these 5it- dicators, the probability that the ob- served sequence is exactly the si quence predicted by the stage theory of, sriet by chance alone is P=.008, Based on the results of the multivari- able analyses of varianc graphic Tactors of age, sex, race/ ethnicity (white/nonwhite), education, and income and a terminal illness diag” nosis reported within 6 months of the death were largely unrelated to within- person dilferences and temporal changes in the grief indictors throughout the study observation period (1-6, 6-12, and 12-24 months postloss). Education be- yond high school was significantly as- sociated with grief indicators 12 to 24 months postloss (Wilks \=0.94, P,,yo=2.52; P=.03), due to its signifl- ccant associations with lesser disbelief (P2.05) and depression (P=.003), and ‘with greater acceptance (P=.02) dur- the demo- 720 JAMA, Febsny 21,2007 297, No.7 (Reprinted) (©2007 American Medical Association, All ights reserved.Ing that period. Education beyond high school was also significantly associated ‘with within-person differences in grief indicators 6 to 12 months postloss (Wilks A=0.95, F, .=2.51; P2.04) and 121024 months postloss (Wilks A=0.04, Pyag9= 3.115 P2.02) due to its signifi- ceant associations with greater diffe cences between acceptance and each of the other grief indicators during each of those periods. Widowhood (compared. swith loss of a parent, child, or sibling in this study group) was significantly as- sociated with within-person differ- fences in grief indicators | 19 6 months postloss (Wilks N=0.3, Pyys=2.51; P=.05), due to its significant associ lions with a greater difference between yearning and depression (P2.02) and a lesser difference between acceptance and yearning (P=.01) during that period. Re- port of terminal liness diagnosis within ‘6 months ofthe death was significantly associated with grief indicators 12 to 24 months postloss (Wilks \=0.93, P,,y.=2.62; P=.03), due to its signifi. ccant association with lower acceptance of the death (P=.008) during that period ‘COMMENT Results of this study identify normal patterns of grief processing over time following the natural death of a loved. fone. Given that the vast majority (04%) of deaths in the United States are the result of natural causes,” the Lindings reflect how the average pet son psychologically processes «typical death of a close family member, Although the temporal course of the absolute levels of the 5 grief indicators did not follow that proposed by the stage theory of grief," when rescaled and examined for each indicator’s peak, the data fit the hypothesized sequence exactly. In terms of absolute frequency, and ‘counter to the stage theory, disbelief was not the initial, dominant grief indica lor. Acceptance was the most often en- dorsed item. Evidently, high degree of acceptance, even in the initial month postloss, isthe norm in the case of nat ral deaths. This conteasts with individ (©2007 American Medical Assoc 1, All rights reserved. als who survived a family member's trau- ‘matic death and those who met criteria for complicated grief disorder,” both groups of whom were found in prelirsi- nary analyses to have significantly lower levels of acceplance relative to the study sample. The lower frequency of accep- tance of the death among participants who reported that the patients termi- nal illness diagnosis was within 6 months compared with 6 months or longer prior to the death suggests that prognostic awareness may promote acceptance of the death, This result is consistent with findings reported elsewhere indicating that preparation for the death is associ- ated with better psychologteal adjust- ‘ment to the loss Future research that examines the elects of prospective rather than retrospective reports of prognostic awareness on the bereaved survivor's ac- ceplance are needed before definitive conclusions can be drawn, Yearning was the most frequent nega- Live psychological response reported throughout the study observation pe- riod (1-6, 0-12, and 12-24 months post- loss). Yearning was significantly more common than depressed mood despite the exclusive focus in the Diagnostic and Statistical Manual of Mental Disorders, Fourth Edition" bereavement section on depressive symptomatology: "As part of their reaction to the loss, some griev- ing individuals present with symptoms characteristic ofa Major Depressive Epi- sode (eg. feelings of sadness... The be- reaved individual typically regards the depressed mood as ‘normal’... The di- agnosis of Major Depressive Disorder is generally not given unless the symp- Comes arestill present 2 months aller the loss."2*" Findings from this report demonstrate that yearning, not depres- sive mood, isthe salient psychological response to natural death, They indi cate that depressive mood in normally bereaved individuals tends to peak at ap- proximately 6 months postloss and does not occur prior to 2 months postloss Findings elsewhere® chronically elevated levels of yearning are a cause for clinical concern. Taken together, these results imply anced for revision of the Diagnostic and Statisti- * indicate that ‘STAGE THEORY OF GRIEF cal Manual of Mental Disorders, Fourth Edition with respect to bereavement Models that tested for phasic epi- sodes ofeach grief indicator revealed that disbelief about the death is highest ini- Lally. As disbelief declined from the first, month postloss, yearning rose until 4 months postloss and then declined. An- ‘ger over the death was fully expressed ‘aU 5 months postloss. Alter anger d clines, severity of depressive mood peaks Atapproximately 6 months postloss and thereafter diminishes in intensity through 24 months postloss. Accep- lance increased steadily through the study observation period ending at 24 months postloss, Because of the mints cule probability that by chance alone these 5 grief indicators would achieve their respective maximum values in the precise hypothesized sequence," these results provide at least partial support for the stage theory of grief The results also offer a point of ref erence for distinguishing between nor- mal and abnormal reactions to loss. Given that the negative grief indica- torsall peak within 6 months, those i dividuals who experience any ofthe in- dicators beyond 6 months postloss ‘would appear to deviate from the nor- ral response toloss. These findings also support the duration criterion of 6 months postloss for diagnosing com- plicated grief disorder,!*"”""* or what is now referred toas prolonged grief dis- order. Unlike the term complicated, which is defined as “difficult to ana- lyze, understand, explain," pro- longed grief disorder accurately de- seribes a bereavement-specific mental disorder based on symptoms of grief that persist longer than is normally the cease (ie, >6 months postloss based on the results of the present study). Pur- thermore, prolonged grief disorder per mits the recognition of other psyehiat- ric complications of bereavement, such ‘as major depressive disorder and post- traumatic stress disorder. Additional analyses are needed to examine grief tra- jectories among those meeting criteria, Tor prolonged grief disorder. The mode of death may be an m= portant factor that influences the course (Reprinted) JAMA, Febrory 21, 2007 Vol 297, No.7 724‘STAGE THEORY OF GRIEF ‘of bereavement adjustment. Inthe pre- sent study, individuals bereaved by traumatic deaths (eg, vehicle crashes suicide) were removed. Bereavement adjustment following deaths from trat- matic causes may be more difficult 10 process and demonstrate higher de- ‘grees of disbeliet and anger and lower levels of acceptance than those re- ported herein, A recent study found that those bereaved by traumatic vs nati ral deaths had greater difficulty in mak- ing sense of the loss.™ Participants who reported that the family member or loved one’s terminal illness was diag- nosed within 6 months of the death did not differ significanily from other par- Licipants with respect to their level of frief indicators, However, the partici- pants who reported the diagnosts within {6 months of the death did report ac- cceplance of the death significantly less ‘often, Subanalyses revealed that dish lief within 6 months postloss was also significantly higher in those for whom the patients terminal illness diagnosis, was reported to be within @ months prior to death. Thus, the manner and Torewarning ofthe death appear to affect the processing of grief. Studies are needed to explore the pattern of grief! lrajectories among the survivors be- reaved by traumatic causes of death, The results should be understood in light of several study limitations. Id ally, all individuals would have been as- sessed itnmediately after the loss rather than beginning at month | postloss. Due to respect for the initial mourning pe- riod and institutional review board con ccems about harm to participants, we did not interview individuals within a month, of the death, In addition, it would have been better to analyze data that reas- sessed individuals each month [rom 0 to24 months postloss. However, ostich data exist nor does the stage theory"? specify in what month postloss each stage would predominate. And, al- though we acknowledge that other grief indicators might have been used, the various proxy measures (eg, stunned for disbelief bitterness for anger, hopeless- ness for depression, quality of ile scores for acceptance/recovery) all revealed re- 22 JAMA, Fey 21,2007 297, No.7 (Reprinted) ‘Downloaded From: https:/jamanetwork.com/ on 11/24/2020 smarkably similar patterns to those pre sented herein, We chose to present the ftems that ft most closely with the stage {indicators illustrated in the literature." U should be noted that participants were younger and less likely to be male compared with the study nonpartict- pants, and that the study sample may be more resilient than is typically the case given the low prevalence of de- pression (8.0% of the individuals had ‘a Hamilton Rating Seale for Depre: sion summary score of =17) com- pared with other samples of bereaved individuals.®">! Samples with more males or with older and more di tuessed individuals might reveal a dif- ferent pattern of grief trajectories than those presented herein, Although the study sample does show some gross slimiarities with the US widowed pop lation in terms of age, sex, and race/ ethnicity, and with other comparable groups in terms of education and me- dian houschold income, itis not di- rectly representative of either the US widowed oF US general population, Nevertheless, age, income, race/ ethnicity, and sex were not signifi- cantly associated with the magnitude or course of grief and the representa- liveness ofthe Yale Bereavement Study would not appear to restrict the gen- cralizability of the results tothe US wid ‘owed population. Despite these limi tations, given that the Yale Bereavement Study provides one of the most com- prehensive longitudinal assessments of grief, these data are as adequate as any available for testing the stages of grief In conclusion, the results of this study provide what appears to be the first empirical examination of the stage theory of grief. They indicate that in the circumstance of natural death, the nor- smal response involves primarily accep- tance and yearning for the deceased, Each grief indicator appears to peak in the sequence proposed by the stage theory. Regardless of how the data are analyzed, all of the negative grief indh cators are in decline by approximately 6 months postloss, The persistence of these negative emotions beyond 6 months is therefore likely to reflect a more difficult than average adjust ment and suggests the need for fur ther evaluation of the bereaved survi- vor and potential referral for treatment. The results provide an evidence base from which to educate clinicians (eg, primary and palliative care phys Gians, geriatricians, psychiatrists, on- cologists, related hospital and hospice stl, bereavement counselors) and lay- persons (eg, patients, family mem- bers, friends) about what to expect fol- lowing the death of a family member or loved one. ‘Author Cotibutins Dis Mackie and Pigeon al seces to al of he data te toy and ake ‘espns forthe nto the data an the ac (ayo the ta sass Sti) concep and design: Maciewshi Pigason, ‘enuiston of dat Prgercon, ‘nati and interpretation of data: Maciej “hang oc, Prgeson. Drafting of the manuscript: Mackjewsh, Zrang, Prierse, Chica revision ofthe mars or inpartant int Fetal content Macejews, rang lek Pigeon, Satstal ana Madeenst, Zang bane inane: Mactjensk, Pigeon, ‘Somistatve techn ormatenal support: Zhang, Prigerse. Sti) sipevision: Macgjews lock, Pigeson Financial Disclosures: Nave parted Fandng/Sopor: hs work was uperteby sane M09 vara to Dr Pigeon) and MMGSE92 (awardedtobrPtgeson fambreNatond stte of Manta Heath and gart CATOGS7O (aad fo Dr Pigeon) fom he Natona Canernstute: and rant NSO 16 awards oO Macs fo he Ne Sel eo NeuelegyalDeoreand Stoke Fd. Ingasowasprodedbytecentrfo Pycho Onegy ‘stdPalaie are RezatenDanFarer Canc {ute ang Womens Hest esearch at Yale Une. Ral ofthe sponsors: The National tute of Mer {al Hea, Naioral Cancer state, Natal lst {ute ot Nevolgtal Disorders and Stoke, Cente fr Psjto- Oncology and Pababve Car Research Das Farber Cancer nattut, ang Women's Heath Re Search a Yale Univetsty had node nput ito the Aesign or conduct of he study, coecton, manage rent arly, or ntepretton ofthe dao prea "aon eve raptor tthe manus EARNS 4. Bony) Processes of mouring Its Psycho aera2i7-339 2: Paes C Bereavrnent Studie in Grit in Aclt Tite London, Enlane Tasted: 1972 3. Bowlby Ataciment and Los: New York, NY: Ba Se Boots 1880 4 Paes CM. WiebsRS.Recovey from Bereavement New var, Ni ase Books 1983 5: Kabler Ror On Desth and Dying, New York, vs ara: 1969, 6 GrayC KoopmanE, Hut The emotional phases i mart epuration an empialwvesbgaton. Am Torthopsychity. 199156113643 7- Leben Green Sk Giese AA Mouring and Tle staf reaction he dest ofan mpi Py Chiat Hosp. 1985 195169-173. (©2007 American Medical Association, All rights reserved.8 Downe-Wamboldt 8, Taiyn D.Anittratonl Survey of death eduation tends nfacltes of m- ing ard medene Death Sud 199721:177 188 5" Naoal Caner nttute, Cancer ope om, rit and bereavement. http://www nci-nih gov ‘Jeancertopies/ pl supperovecarelvercaveent ‘aent/pages. Accessed November 28, 2008 40. Bonanno GA Wertman Ch, Leth DR, eta Reine too and von hit: prspectve Sy ome tet months poss Per Soc schol gomasstis0-1164, 1M. tgaeton W Rphae 8 Bumet, Martek N ‘ongtuainal tay comparing berenverent phe ‘omenain recent bres spouses adult ciren nd parent Aust NZ Pychiay. 199832235" a8, 52. OW CH, LuegerRPatemso changein mental eat status during the fest two Yeats of spousal Dereavement Death Stu 2002:26387-411 43, Wertman CB ser RC TRerytisaf coping with los J Const ln Pychl. 1989.57 348-397 1 lcbs. Patol Grit Malctapatin oss. Washington: Oc: Amencan Peete res 1983 15 HoyertDL Rung. sa. Deals relay dhitaor 2003. Vt Heah St 18 2005(83)-1618 46. Pigeon HG, Mace. A cal fr sound mpl testing and evan of rer for ea pleated gre proposed for DSW. Omega J Death Dying 2605323 47, Thang, aA Prigeon HG, Update on bereavement recency evidence-based gules forthe dagnose and treatment of complicated Bereavement alt Med 2005911881208, 48. US Census rea, Populaton, housing, eco roi, and geographiecta np acts cneue (ov/serleuNPTabe? bay geo 1401000054 ffame=ACS. 2005 E500. NPOI eds name= redologetake Accessed November 2, 2006 48" Pigeon HG, Jacbs SC. Traumatic gitar acdi- ‘ictdsorderaraonseconeensur ere snd te liar empieal est Sboube Mi, Hansson RO, StrosbeW, Shut ede Handbook of eranvement Research Consequences Coping and Care, Wash ington DC: Ameen Psycelopal Asati, 2001 sees 20. Chochinav HM, Wizon KG, Enns Mt, "Ave you ‘epressed?” szeening for cepresin nthe te tally Am Psychiat, 1997 134 6-676 2s Heo ner Aare Wimp of vert Sele ameaswet subjective ses Paychosom Med israraosi8 22. Hamton MA rating sale oc depeson. Neu {olNeurosng Poy. 1960235662 23. Bary LC Kasl SV, Prigeson HG, Psychics ‘ede among bereaved persons te fle of pe. ‘ve cestancs of esth and peparednes for eat. J Gant Psa 20021087 457. Dat AnercanPjciatneAsocabon, Bereavement ‘STAGE THEORY OF GRIEF In: Diagrstcand Stasi Manual of Meta Dis ‘es, eur Eater, Washngton, DC Ameian Py ‘hc Asocton 1998 688 35. gaan HG, lacbe Sc, Caing for bereaved pa {ens athe cos stsuddely ga." JAMA. 201 23613001376 26. Pigason HG, Vandererer, Macs PR Prologes ge dsorcerasa mental dserde nc Sonn DSA tebe lS, Hanston RO, Stoebe iW sthut HA ede Handbootof Bereavement Re Search and Paes Canty Parpectves Wash ington DC: American Psjcologia Asciaton Pes 207 27. Dictionary com Wel ste Deftions forthe word Cempeata hip /Sconayretrencecon/browse Yeampleated. Accesibity vetted January 18 ow. 28. Curr IM, Holand JM, Neimeyer RA, Sense Inakng gre and the experience of tlet ios to tad sredational model Desh Stud 20063003 be 29. Gayton Haas A, Mauce WL. The depres: Son of downood Bryn. 1972 2071-77 50. Zsoo4s, shuchie SR Bepesion toughest pera tne death of aspouse. A Poh. 19 Teeet352 31. Zonk’, Shuler SR, Sedge PA etal Thespc ium of depressive phenomena efter spousal Deeavement ln ena. 19945 sup 29-36 | find that a great part of the information I have was acquired by looking up something and finding some thing else along the way. —Franklin P. Adams (1881-1960) (©2007 American Medical Association, All rights reserved. ‘Downloaded From: https:/jamanetwork.com/ on 11/24/2020 (Reprinted) JAMA, Februry 21,2007 Vol 297, No.7 7BTerres Drafting ofthe manus: Croft, Darbysire, van Tit Cite evision ofthe muse for important nalctual content Cio, Da- bye, Jackson, van Ti! ancl Disclosures: None repartee Funding Support No ouside undng or support was eceived for this study, Discllmer eviews opment, andor ndings contained inte stay ae hose tothe autor and shoud not be conve othe offal North Aan Treaty ‘xganzaton or non rt Alan Tety Orgaizabon mllary poston pole, tor desion urls so designated by other ofa dacmentaton ‘Adnowledgment We tank Rati Bence, MD (SAF Cro), Roman lato [MO'ISAF vai), Serdor Kava ID (SAF Turkey), Lopes Poves, MO (SAE Spal) and Ca Gustav Schutz, MO (SAF Sweden), for ther hlp acting ‘hssuivey They received no compensation fr parpaton nfs ud 41. Funk Baumann M. Geographic dsbuton of malaria at waver esnations In-Shagenhat ed Tele” Mala Hanon, Ont: 8 Deca, 20" 5693 2, Malaita personnel etueing om Afghanistan, ConemanDisRep COR ‘iy 20021245 Mt ww oa orale rhe! 2002 /cd2702 pa Be cessed Ferry 26,2007 SS Boecken GH Pathogenesis and management ofa ae maiestation of uax alata afer depioyment to Afghanistan consions for NATO we ces ed (Glee Mi Med 2008;170488-49% 4. Ciminr , undage Malin US mite frcs a description of depoy ‘peri exposres rom 2003 tough 2008 AJ Trop Hed Hy. 20077627527. 5 Croft AM. Malar revebon in taeler In Tovey Ded lina dence. tendon, Engng Bh Pubihing Grup, 2005 958-972 6 Centr Disease Conta and Preven. The elon beck, avlrs he, ‘gional mala injmation hp! ww ede gv /tavelrepondinalsalindag Wimtmalrie Accessed February 26,2007 7a Lawrarce CE, Cot AM. Do mest als prevent mala? a sstematicre ew of ta. Travel ed 2008.11 32:96 CORRECTION Inconect Example Inthe Origal Contution ened “An Empl Barns tor af the sage Theery of Cre” pushed in the Febraary 2, 2007 sue of JANA (2007:357 716-723), anincofet xamge nae provided fort case, ‘Onpage 17 couma 1, second ul parayanh, the tra sentence should be "Be ‘ue sporxinatly 94% of US dens es rom natural ease (ear di ase, cancer” det tom unnatural causes (eg ca ashes, Sie wee x ‘ide thereby enabling the resus fo be generated ote most commen pes of eeath Medicine requires not only the intellectual cultiva- lion of a science, but the patience and the practical skill of an art, AL the bedside we must be animated by the feeling of faithful artisans, of men whose ob- ject and duty is practical work; for when the art of medicine is needed by the suffering and the dying it {fs no question of mere theoretical knowledge and ex- lancous acquirement. But skill in the commonest art, iis not to be attained without much practice, far less in the complicated and difficult ar of healing, where every case presents some peculiarities. To practice it suceessfully, we must have made our home atthe bed- side, and, if may say so, have lived with disease, ob- serving it in all its forms and changes, Sir William Withey Gull (1816-1890) 2200. JAMA stay 2730, 2007 Vo 297, No, 20 (Refined) (©2007 American Medical Association, All ights reserved. Downloaded From: https:/