The Stages of Grief

In brief

Elisabeth Kübler-Ross (1926 to 2004) introduced the five stages in On Death and Dying (1969), a book drawn from a seminar she ran with terminally ill patients at the University of Chicago's Billings Hospital, beginning in 1965. She proposed that people who learn they are dying tend to pass through denial, anger, bargaining, depression, and acceptance. She was writing about the dying, not about people grieving someone else's death. Within a few decades the model migrated, partly through her own later writing and mostly through popular culture, into the reigning folk theory of how anyone processes any loss at all: a death, a divorce, a layoff. Grief researchers who have since tested the model directly, on people actually bereaved, have found little support for a fixed sequence that everyone climbs in order.

The full treatment

The problem it answers

Mid-century American medicine treated death as a subject to manage around rather than discuss. Physicians routinely withheld terminal diagnoses from patients on the theory that hope mattered more than truth, and dying patients often spent their final months isolated, aware something was wrong but denied any real conversation about it. Kübler-Ross, a Swiss-born psychiatrist working in Chicago, began sitting with dying patients and asking them, on tape, what the experience was actually like. The gesture itself was the innovation: treating the dying as informants about their own condition, at a moment when hospitals rarely did. Her work ran alongside the hospice movement Cicely Saunders was building in London in the same years, both pushing medicine to attend to the end of life rather than manage it away.

How the model works

From those interviews Kübler-Ross drew five recurring responses. Denial: "No, not me, it cannot be true," a buffer that lets the patient absorb shocking news gradually. Anger: "Why me?", directed outward at family, staff, or fate once denial can no longer hold. Bargaining: an attempt to postpone the outcome, often through private deals with God or fate, more time in exchange for good behavior. Depression: grief for losses already suffered and losses still to come. Acceptance: not happiness, in her account, but a kind of quiet expectation, "the final rest before the long journey." Kübler-Ross insisted the stages were not a fixed staircase: patients could skip stages, revisit them, or hold two at once, and she described hope as running through all five. That qualification did not survive contact with the public.

What it claims, and what got added to it

The original claim was narrow: five common responses observed in dying patients coming to terms with their own mortality. The claim that later spread was broader and different: that grief for any loss unfolds through these same five stages, in roughly this order, and that a mourner not progressing through them is grieving abnormally. Kübler-Ross herself contributed to that expansion. On Grief and Grieving (2005), written with David Kessler and published the year she died, applied the framework explicitly to bereavement, the death of someone else, rather than to one's own dying. By then the DABDA acronym (denial, anger, bargaining, depression, acceptance) had already spread through nursing curricula, self-help books, and daytime television as the default account of mourning in general.

The key test

The clearest empirical check came from Paul Maciejewski, Baohui Zhang, Susan Block, and Holly Prigerson, who published "An Empirical Examination of the Stage Theory of Grief" in JAMA in 2007. Drawing on the Yale Bereavement Study, they tracked several hundred recently bereaved people at intervals after a loss and measured the five states directly. They found that acceptance, not denial, was the most commonly endorsed response from the first month onward, that yearning (a state the original model has no separate place for) ran higher than anger or bargaining throughout, and that where the other states did peak, they did not do so in the proposed order or as discrete phases replacing one another. The study did not claim grief has no structure at all, only that this particular staged structure, tested against real trajectories, does not hold.

Related distinctions

"Stages of dying," Kübler-Ross's actual subject, a person's own anticipated death, is not the same claim as "stages of grief," the popular extension to mourning someone else's death, and neither is the same as "grief," the emotional response to loss generally, or "mourning," its outward and culturally shaped expression. Frameworks built to replace the stage model tend to drop staging altogether. J. William Worden's task model (1982) describes mourning as four active tasks, accepting the reality of the loss, processing the pain, adjusting to a world without the person, and finding a way to carry the relationship forward, worked on in any order rather than in sequence. Margaret Stroebe and Henk Schut's dual process model (1999) describes oscillation between confronting the loss directly and attending to ordinary life, rather than any forward march toward a final state.

Lineage

Kübler-Ross's work belongs to a mid-century turn toward treating death as a legitimate subject of psychological and medical attention, alongside Cicely Saunders's hospice movement and Ernest Becker's The Denial of Death (1973). Downstream, it became the template for a far wider genre of "stages of X" folk theories applied to divorce, addiction recovery, job loss, and organizational change, most of them tracing back, loosely, to her five terms.

The strongest case for it

The model's staying power is not accidental. It gave dying patients, grieving people, and the clinicians around them a shared, plain-language vocabulary for feelings that had previously gone unnamed in medical settings; denial, anger, and bargaining are recognizable to almost anyone who has sat with a serious loss, and naming a feeling as a stage rather than a personal failing can be genuinely stabilizing. Kübler-Ross's larger achievement, distinct from the specific taxonomy, was to insist that dying patients have an inner life worth documenting and listening to, and that insistence reshaped palliative care, medical training, and hospice practice in ways that have held up far better than the five stages themselves.

The strongest case against it

Camille Wortman and Roxane Cohen Silver, in "The Myths of Coping with Loss" (1989), identified the stage model as a source of several unsupported assumptions common in clinical practice: that intense distress is inevitable after a loss, that failing to display it signals pathology or denial, and that actively working through grief is necessary for healthy recovery. Their review of the empirical literature found each assumption poorly supported. George Bonanno's longitudinal research, summarized in The Other Side of Sadness (2009), went further: tracking bereaved people over time, he found that resilience, sustained low distress from early on, without any staged unfolding, is the single most common trajectory after loss, followed by smaller groups showing gradual recovery from acute distress or chronic, prolonged grief. A fixed sequence that all mourners are expected to pass through fits none of these trajectories well. Robert Kastenbaum, an early figure in the field of thanatology, raised a related objection as far back as the 1970s, noting that Kübler-Ross's original interviews were never designed or reported as a controlled study: no fixed sample, no independent raters, no tracking of the same patients over time, so the sequence was closer to a clinical narrative than a tested finding. Maciejewski's 2007 data supplied the sharpest empirical blow: acceptance appearing early rather than late, and no clean order among the rest.

Where it stands now

Clinicians in palliative and grief care largely no longer treat the five stages as a diagnostic checklist, and most grief researchers now favor trajectory models like Bonanno's or task-based models like Worden's, which fit the variability the data actually show. But the stages have outlived the debate that discredited their literal claim. DABDA remains, by a wide margin, the public's default vocabulary for loss, taught in nursing programs, cited in eulogies, and invoked well outside its original scope, in careers, breakups, and diagnoses that have nothing to do with a hospital ward in 1960s Chicago.

Test yourself

Think of a real loss in your life and try to place it honestly against the five stages, rather than against what you have been told grief should look like. If it did not arrive in order, if two stages overlapped, or one never showed up at all, that is not a sign you grieved wrong. It is closer to what the actual data on grieving people finds.