The stage theory of grief has survived decades despite limited and only recent empirical support, ^{Reference Maciejewski, Zhang, Block and Prigerson2} and commentaries still attempting to discredit it. ^{Reference Bonanno and Boerner3,Reference Silver and Wortman4} Notions of stages of grief for dying patients ^{Reference Kubler-Ross1} and for bereaved survivors ^{Reference Maciejewski, Zhang, Block and Prigerson2,Reference Bowlby5–Reference Prigerson, Jacobs, Stroebe, Hansson, Stroebe and Schut8} live on in popular culture, ^9,10 remain taught in medical schools, ^{Reference Downe-Wamboldt and Tamlyn11} and continue to be cited by health authorities as established fact. ¹² What might explain the sustained, widespread and uncritical endorsement of the stage theory of grief? From a human interest perspective, it may reflect a desire to make sense of how the mind comes to accept events and circumstances that it finds wholly unacceptable. Answers to the question of whether or not discernible patterns emerge in psychological reactions to loss may reveal normative bereavement responses and identify processes necessary for promoting positive adjustment to the loss. From a clinical perspective, knowledge of how people grapple with objectionable realities such as their own or a close other's death could inform interventions designed to ameliorate loss-related distress.

Results from our study, ^{Reference Maciejewski, Zhang, Block and Prigerson2} together with enduring popular and scientific interest in the topic, suggest that it may be time to reevaluate stage theories of grief and consider their potential clinical utility.

States, not stages, of grief

Stage theories posit that grief, whether among terminally ill individuals or bereaved survivors, progresses through a sequence of distinct psychological phases. Kubler-Ross ^{Reference Kubler-Ross1} describes each stage as a discrete phase of grief that is separate from, if not conditioned on the resolution of, prior grief stages. During each stage's period of ascendance, it is expected to prevail over the other stages. We found ^{Reference Maciejewski, Zhang, Block and Prigerson2} that this may not necessarily be the case for the typical bereaved survivor. Yearning was found to be the predominant distressing sentiment throughout the acute bereavement period (i.e. our 1–23 months post-loss observation period). Less frequently experienced reactions such as disbelief, anger and sadness declined on average over time from loss. When each grief indicator was rescaled to enable a comparison of peaks, all peaked within 6 months post-loss and in the exact sequence proposed by Bowlby ^{Reference Bowlby5} and Parkes, ^{Reference Parkes6} and illustrated by Jacobs. ^{Reference Jacobs7} All of these grief indicators were highly correlated with one another, which is understandable given that three of the four grief stage indicators were extracted from an internally consistent grief inventory. ^{Reference Prigerson, Jacobs, Stroebe, Hansson, Stroebe and Schut8,Reference Prigerson, Maciejewski, Reynolds, Bierhals, Newsom, Fasiczka, Frank, Doman and Miller13} These findings suggest that disbelief, yearning, anger and sadness may represent aspects of a single underlying psychological construct – grief.

As grief decreases, acceptance increases

Figure 1 plots the average of each of the four negative grief indicators, and the combined average of all four indicators (i.e. the ‘grief’ curve). After about 4 months, the constellation of grief indicators declines through 23 months, with parallel shifts downward in all four curves. Importantly, as grief falls, acceptance of the loss rises, suggesting that grief and acceptance may be opposite sides of the same coin. Grief and its associated features may largely reflect an emotional inability to accept the loss of something cherished. At its core, grief may be the state of emotional unrest and frustration associated with wanting what one cannot have. Acceptance, by contrast, may represent emotional equanimity – a sense of inner peace and tranquillity that comes with the letting go of a struggle to regain what is lost or being taken away.

Peaceful acceptance as a goal for those confronting loss

We recognise that some individuals will neither want nor have the capacity to accept loss peacefully. We are not suggesting that all dying patients or bereaved survivors be implored to confront death with peace and equanimity, nor that complete death acceptance is a realistic goal. What we are suggesting is that enhanced degrees of acceptance, and reduced grief, appear associated with less suffering, implying that there may be benefits to promoting acceptance.

Normal grief and prolonged grief disorder

By contrast, intense, prolonged grief has been shown ^{Reference Prigerson, Jacobs, Stroebe, Hansson, Stroebe and Schut8,Reference Prigerson, Maciejewski, Reynolds, Bierhals, Newsom, Fasiczka, Frank, Doman and Miller13–Reference Prigerson, Vanderwerker, Maciejewski, Stroebe, Hansson, Schut and Stroebe16} to constitute ‘a clinically significant behavioral or psychological syndrome or pattern that occurs in an individual and that is associated with present distress or disability’ – meeting the (DSM–IV) ¹⁷ definition of a mental disorder.

Fig. 1

Grief and acceptance of loss over time.

Specifically, criteria for prolonged grief disorder ^{Reference Prigerson, Vanderwerker, Maciejewski, Stroebe, Hansson, Schut and Stroebe16} require bereaved individuals to have severe levels of yearning, and five of the following nine symptoms for ≥6 months post-loss: disbelief and bitterness over the loss, confusion about one's identity, an inability to trust others, numbness (absence of emotion) and feeling that life is empty and meaningless since the loss, difficulty accepting the loss and moving on with life (e.g. making new friends, pursuing interests), and feeling stunned by the loss. Individuals who meet criteria for prolonged grief disorder have been shown to be at heightened risk for present and future major depressive disorder, post-traumatic stress disorder, and generalised anxiety disorder, suicidal ideation, functional disability and diminished quality of life relative to individuals who do not meet criteria for prolonged grief disorder. ^{Reference Prigerson, Vanderwerker, Maciejewski, Stroebe, Hansson, Schut and Stroebe16} Thus, just as acceptance is associated with less distress, an inability to resolve grief over time, as in the case of prolonged grief disorder, is associated with more distress and dysfunction, making it a worthwhile target for psychotherapeutic intervention.

Cognitive and emotional acceptance: clinical and public heath significance

Results from our research on patients with cancer confronting terminal illness indicate that cognitive and emotional acceptance are distinct but related phenomena. ^{Reference Mack, Nilsson, Balboni, Friedlander, Block, Trice and Prigerson18,Reference Ray, Block, Friedlander, Zhang, Maciejewski and Prigerson19} By cognitive acceptance, we mean the dying patients' understanding or recognition that their illness is terminal (defined as a life expectancy of <6 months). Cognitive acceptance is associated with patients' reports of having a discussion of end-of-life treatment preferences with their physician, suggesting that clinicians may be able to influence cognitive acceptance. ^{Reference Ray, Block, Friedlander, Zhang, Maciejewski and Prigerson19–Reference Wright, Ray, Zhang, Mack, Mitchell, Nilsson, Trice, Block, Maciejewski and Prigerson21} It is also related to higher rates of ‘do not resuscitate’ order completion and with the prediction of greater use of palliative care services. ^{Reference Ray, Block, Friedlander, Zhang, Maciejewski and Prigerson19–Reference Wright, Ray, Zhang, Mack, Mitchell, Nilsson, Trice, Block, Maciejewski and Prigerson21} In cross-sectional analysis, patients who had cognitively accepted their terminal illness had more difficulty emotionally accepting it than those who had not achieved cognitive acceptance.

The acceptance to which Kubler-Ross ^{Reference Kubler-Ross1} refers is essentially a state of emotional acceptance of impending death. We developed the Peace, Equanimity, and Acceptance in the Cancer Experience (PEACE) ^{Reference Mack, Nilsson, Balboni, Friedlander, Block, Trice and Prigerson18} scale to measure the patients' emotional acceptance of their life-threatening illness. This is a 12-item questionnaire assessing the patient's sense of acceptance, calmness and peace, as well as their sense of struggle or desperation about their illness. We found that bereaved individuals who reported that they had known of the patient's terminal illness >6 months prior to the death had significantly higher levels of emotional acceptance in bereavement. ^{Reference Maciejewski, Zhang, Block and Prigerson2} Unlike cognitive acceptance, emotional acceptance of a terminal prognosis was associated with concurrent feelings of being less terrified and more supported. ^{Reference Mack, Nilsson, Balboni, Friedlander, Block, Trice and Prigerson18} Patients with cancer who were peacefully aware (cognitively and emotionally accepting) were more likely to engage in advance care-planning, had better mental and physical health in the last week of life, and their surviving relatives had significantly better quality of life 6 months after their death. ^{Reference Ray, Block, Friedlander, Zhang, Maciejewski and Prigerson19}

An agenda for future research

Longitudinal data with multiple assessments prior to death for dying patients and their caregivers and, following death, for bereaved individuals are needed to plot the course of grief. In principle, analysis of longitudinal data could determine whether the course of grief is better characterised in terms of stages or states, and whether each stage or state must become fully expressed before moving on to the next stage or state. Alternatively, such data could determine that the course of grief is more accurately characterised in terms of co-occurring symptoms of grief that evolve in concert rather than in terms of either stages or states.

Longitudinal studies could clarify the way in which grief resolution relates to acceptance of dying and death, and whether grief relates differentially to cognitive as compared with emotional acceptance. Within the individualistic psychology of Kubler-Ross, ^{Reference Kubler-Ross1} future research could determine how personality traits (e.g. Erikson's ^{Reference Erikson22} ego-integrity) influence acceptance. Social psychological studies might explore the ways in which physicians and other healthcare providers influence the patient's acceptance of impending death and surviving family members' bereavement adjustment. Preliminary work has shown that cognitive and emotional acceptance relate cross-sectionally to advance care-planning, preferences regarding aggressiveness of end-stage cancer care, and predict use of aggressive and palliative care and quality of life near death. ^{Reference Mack, Nilsson, Balboni, Friedlander, Block, Trice and Prigerson18–Reference Wright, Ray, Zhang, Mack, Mitchell, Nilsson, Trice, Block, Maciejewski and Prigerson21} A thorough analysis of the outcomes of acceptance – both positive (better psychosocial functioning, less symptom burden) and negative (premature abandonment of treatment that prolongs life) – is needed before acceptance can be recommended as a goal of terminal illness and ‘after’ (bereavement) care.

Conclusions

Studies demonstrating how the mind comes to comprehend, cope with and accept death address core psychological issues that need to be better understood before they can inform interventions to promote adjustment. Rather than distinct, sequential stages of grief, it may be more accurate to conceptualise proposed stages as multidimensional grief states that evolve and diminish in intensity over time. Decline in grief-related distress appears to correspond with an increase in peaceful acceptance of loss. This suggests a need for studies to advance understanding of how the resolution of grief may facilitate acceptance. The potentially therapeutic role of clinicians and family members in advancing acceptance should be examined and inform interventions to promote the mental health of those confronting death. Research that determines ways to promote peaceful acceptance offers the promise of offsetting the pain and misery frequently associated with dying and death.