Fourteen months after her husband died, a woman apologizes to her therapist for crying. She has read that grief comes in five stages and that the final one is acceptance. By her arithmetic, she is overdue. She wants to know what’s wrong with her.
Nothing is. But the framework she’s been measuring herself against was never built for her.
The five stages of grief — denial, anger, bargaining, depression, acceptance — may be the most widely circulated idea in popular psychology. They turn up in hospital pamphlets, HR bereavement policies, sitcom subplots, and until recently, medical school lectures. A 2021 review of 72 grief websites found that 61% addressed the model, most with almost no critical commentary. The authors’ concern was blunt: an uncritical portrayal “may give the impression that experiencing the stages is the only way to grieve.”
Which leaves a lot of grieving people quietly convinced they’re doing it wrong.
Where the stages of grief actually came from
Elisabeth Kübler-Ross published On Death and Dying in 1969, drawing on interviews with roughly 200 terminally ill patients at a Chicago hospital. Her subject was people confronting their own deaths — not the bereaved. Not widows, not orphaned adults, not parents who had buried a child.
Over the following decades the model was quietly transplanted into different territory — the experience of surviving someone else — and stuck there, unexamined. As the JAMA researchers who eventually tested it noted, stage theory remained “a widely accepted model of bereavement adjustment still taught in medical schools, espoused by physicians, and applied in diverse contexts” despite never having been “tested empirically.” A remarkable thing to be able to say about an idea nearly forty years old.
What happened when someone finally tested it
In 2007, a Yale team published the first empirical examination of stage theory in JAMA. They followed 233 bereaved adults for two years after a death from natural causes, measuring five things at intervals from 1 to 24 months: disbelief, yearning, anger, depression, and acceptance.
Two findings upended the popular picture. First, acceptance was the most frequently endorsed of the five responses, and it increased steadily across the whole two years. It wasn’t waiting at the end of a corridor. It was already in the room at the first assessment, one month after the death. Meanwhile disbelief — the study’s stand-in for denial, and stage theory’s supposed opening act — was, in the authors’ words, “not the initial, dominant grief indicator.”
Second, the strongest negative feeling was not anger or depression. It was yearning. Plain, unglamorous missing. From month one through month twenty-four, wanting the person back was the engine of grief — not the bargaining or rage the model advertises.
The five responses did peak in roughly the predicted order: disbelief at one month, yearning around four, anger around five, depression around six. So the sequence wasn’t nonsense. But the architecture was. Grief is not a relay race where one emotion hands off and leaves the track. Everything is present at once, in shifting proportions. Acceptance and devastation share a room from the beginning.
If you’ve ever laughed at a funeral and then felt monstrous about it, this is the research that lets you off the hook.
Most people are more resilient than the model assumes
In 2002, George Bonanno and colleagues published a study with an unusual design: data on 205 older married adults collected before their spouses died — on average three years ahead — then again at 6 and 18 months after. Without that baseline, you can’t tell someone flattened by a death from someone who was struggling long before it.
The patterns they found looked nothing like a single grief curve. Just under half — 45.9% — showed a resilient pattern: stable, low distress throughout. Chronic grief accounted for 15.6%. The pattern most people assume is normal, elevated distress that gradually declines, described only 10.7% of the sample.
The limits matter: these were older adults, mostly widowed after an expected death. Sudden loss, the death of a child, and grief in younger people all look different, and the researchers said so. But one conclusion deserves wider circulation. The team looked hard for evidence that resilient grievers were secretly repressing something and didn’t find it. Their warning ran the other way: “Offering treatment to individuals who are coping effectively is not likely to be helpful and might produce some harm.”
When grief becomes something with a name
For a minority it isn’t all right, and for a long time there was no clinical language for that. In March 2022, the American Psychiatric Association added prolonged grief disorder to the DSM-5-TR — the newest condition in the manual.
The criteria are narrower than people expect. The death must have occurred at least 12 months earlier for adults (6 months for children and adolescents). There must be intense longing or preoccupation with the person who died, present “most of the day, nearly every day for at least a month,” plus significant impairment in functioning. The APA lists eight associated symptoms, including identity disruption (“feeling as though part of oneself has died”), avoidance of reminders, emotional numbness, and a sense that life is meaningless. A meta-analysis of 14 studies put the pooled prevalence at 9.8% of bereaved adults after non-violent deaths — roughly one in ten.
The Yale researchers offered a softer marker for everyone else: because the negative indicators all peaked within about six months, people still scoring high beyond that point “might benefit from further evaluation.” Six months is not a deadline. It’s a checkpoint.
One thing to say directly, because grief and despair are not strangers: if you’re in the US and having thoughts of suicide or self-harm after a loss — or you’re worried about someone who is — call or text 988 to reach the Suicide & Crisis Lifeline. Free, confidential, staffed around the clock.
What the evidence says actually helps
Here the research gets genuinely useful. A 2016 four-site randomized trial in JAMA Psychiatry assigned 395 bereaved adults with complicated grief to four arms: an antidepressant (citalopram), placebo, 16 weeks of targeted grief therapy plus citalopram, or that therapy plus placebo.
Therapy plus placebo produced an 82.5% response rate, against 54.8% for placebo alone. Adding the antidepressant to therapy barely moved grief outcomes — 83.7% versus 82.5% — and the antidepressant on its own was not significantly better than placebo at week 12 or week 20.
The medication did one thing well: where depression was also present, adding it reduced depressive symptoms more than therapy alone. Which is the cleanest available demonstration that grief and depression, whatever their overlap, are not the same machine.
The trial found something else worth knowing: suicidal ideation fell substantially more among people receiving the grief-focused therapy than among those who didn’t. Structured, time-limited, grief-specific work — the kind of targeted approach specific therapy modalities are built to deliver — moved the thing that mattered most.
If you’re the one grieving
Stop checking your progress against a staircase. The more useful question isn’t how long has it been but is my life getting larger or smaller?
Yearning is not a malfunction. Missing someone acutely, years later, in a grocery store aisle, is what the data says grief mostly is. The signal that something else is happening is when the longing fills most of the day, nearly every day, for months, and forecloses everything else. Loss also shrinks the circle of people who check in, usually right as the numbness wears off; if that’s where you are, the isolation is worth treating as its own problem.
And if you do talk to someone, bring the specifics rather than the label. “I haven’t been able to open his closet in a year” tells a clinician far more than “I think I’m stuck.” A first session is mostly them trying to understand the shape of what you’re carrying.
The stages were a real observation about dying people, made with real compassion, in a hospital where nobody talked about death at all. They were never a map for the people left behind. Nobody needs permission to grieve in the wrong order — but if it helps to know the evidence is on your side, it is.
This article is for general information and isn’t a substitute for professional diagnosis or treatment. Prolonged grief disorder can only be diagnosed by a qualified clinician who has evaluated you directly. If grief is interfering with your daily life, please reach out to a licensed professional. If you’re in the US and experiencing thoughts of suicide or self-harm, or you’re worried about someone who is, call or text 988 to reach the Suicide & Crisis Lifeline — free and available 24/7. Outside the US, Find A Helpline lists crisis services by country.