The 5 Stages of Grief: What Therapists Need to Know (Beyond the Basics)

Last updated: April 2026 · 10 min read

The five stages of grief are the most widely known psychological model in popular culture. Denial, anger, bargaining, depression, acceptance — most clients walk into a first grief session already familiar with the framework. Many have used it to try to locate themselves on a map of their own bereavement. Some have felt comforted by it. Others have felt judged by it, or confused about why their grief doesn't seem to match.

For clinicians, the five stages present a specific challenge: the model is so culturally dominant that it shapes client expectations before therapy even begins — yet the clinical evidence base has moved significantly beyond it. Understanding what the stage model gets right, where it falls short, and what more current frameworks offer is essential for any therapist doing grief work today.

---

Where the Five Stages Actually Come From

Elisabeth Kübler-Ross introduced the five stages in her 1969 book On Death and Dying, based on her observations of terminally ill patients at the University of Chicago. The stages — denial, anger, bargaining, depression, and acceptance — were originally descriptions of the psychological responses she observed in people facing their own death, not in people bereaved by someone else's.

This is a distinction that is frequently lost in popular use of the model. Kübler-Ross was writing about the dying process, not about bereavement. The stages were later applied to grief more broadly — including by Kübler-Ross herself in her later work — but the original empirical basis was a specific clinical population in a specific context.

The model was also explicitly not intended as a linear sequence. Kübler-Ross was clear that individuals might experience the stages in different orders, skip stages entirely, return to earlier stages, or experience several simultaneously. The popular understanding of the stages as a progression to move through — and acceptance as the destination — was a cultural interpretation rather than the model's original framing.

Kübler-Ross herself wrote in her later work that she regretted the stages being misunderstood as a rigid linear sequence. They were meant as a framework for recognising experiences, not a prescription for how grief should unfold. (Kübler-Ross & Kessler, 2005)

---

What the Stage Model Gets Right

Before examining its limitations, it is worth acknowledging what the stage model genuinely contributed — and continues to offer clinically.

It named grief as a legitimate psychological process. When Kübler-Ross published in 1969, death was a largely medicalised and taboo subject. Her work brought grief into the clinical and cultural conversation in a way that had not previously existed. The stages gave people language for experiences they had been having without a framework to understand them.

It validated the full emotional range of grief. Naming anger and bargaining as normal grief responses — not pathology, not failure to cope — was clinically significant at the time. Many clients still arrive in therapy believing their anger at the deceased, or their guilt, or their relief, is abnormal. The stage model's broad emotional scope remains useful as a starting point for normalisation.

It is immediately accessible to clients. The simplicity that makes the stage model clinically limited also makes it culturally accessible. Clients who have never engaged with psychological frameworks often find the stages easy to understand, and starting from what a client already knows — even if you then complicate it — is good therapeutic practice.

Acceptance remains a clinically meaningful concept. The destination Kübler-Ross proposed — acceptance — is not simply resignation. In her framing, and in more contemporary usage, acceptance means acknowledging the reality of the loss and finding a way to live with it. This is closely aligned with ACT's concept of acceptance and with the adaptation focus of more current grief models. The word endures because the concept is sound.

---

Where the Stage Model Falls Short

The clinical limitations of the stage model have been extensively documented in the bereavement literature over the past three decades. The key problems for practising therapists:

The empirical base is weak. The stages were derived from clinical observation of a specific population, not from systematic research. Multiple studies attempting to confirm the stage sequence in bereaved populations have failed to find evidence that people grieve in the order the model proposes, or that the stages are universal, or that progression through them predicts better outcomes. (Maciejewski et al., 2007; Holland & Neimeyer, 2010)

It implies a finish line. The model's linear structure — even when qualified as non-linear — implies that grief has an end point: acceptance. This creates two clinical problems. First, it can generate shame in clients who feel they have "not reached acceptance" after what they or others consider an appropriate amount of time. Second, it positions the goal of grief as completion rather than integration — which is inconsistent with the continuing bonds evidence base, which shows that maintaining an ongoing relationship with the deceased is adaptive rather than a failure to reach acceptance.

It does not account for the oscillating nature of grief. The Dual Process Model — supported by far stronger empirical evidence — demonstrates that healthy grieving involves moving back and forth between loss-orientation and restoration-orientation, not progressing through discrete emotional stages. Clients who understand the stage model may interpret moving "back" to an earlier stage as regression rather than as the normal rhythm of bereavement.

It focuses on intrapsychic emotion at the expense of meaning and identity. The stage model is primarily a description of emotional states. It says relatively little about the meaning-making and identity reconstruction that contemporary grief research identifies as central to long-term adaptation. Clients whose grief involves not just emotion but a disrupted sense of who they are — which is most clients following a significant loss — need frameworks that address this dimension.

It can be used to police grief. Perhaps the most clinically significant problem with the stage model is the way it gets weaponised — by clients themselves, by families, and occasionally by clinicians — to judge whether someone is grieving correctly or efficiently. "You should be at acceptance by now" and "you seem stuck in the anger stage" are statements that cause genuine harm to bereaved people. A model that generates these judgements is clinically problematic regardless of its original intent.

---

More Useful Frameworks for Clinical Practice

The bereavement field has produced several models with stronger empirical support and greater clinical utility. These are not replacements for each other — they illuminate different aspects of the grief experience and work best used together.

The Dual Process Model (Stroebe & Schut, 1999)

The Dual Process Model proposes that adaptive grieving involves oscillating between two orientations: loss-orientation (processing the grief, focusing on the loss, yearning) and restoration-orientation (attending to life changes, building a new identity, managing secondary stressors). Neither orientation is better; both are necessary. The movement between them — which Stroebe and Schut call oscillation — is the mechanism of healthy adaptation.

The DPM has strong empirical support and is clinically useful in several ways. It normalises "good days" without framing them as a failure to grieve. It helps clients understand that attending to life — going back to work, socialising, planning for the future — is part of grief, not an avoidance of it. And it gives clinicians a map for assessing whether a client is spending all their time in one orientation at the expense of the other, which is associated with poorer outcomes.

For clients who arrive familiar with the five stages, the DPM is often the most accessible bridging framework — it reframes their oscillating experience as adaptive rather than chaotic, without requiring them to abandon the stage model entirely.

Worden's Tasks of Mourning (1982, revised 2018)

William Worden's Tasks of Mourning model reframed grief from something that happens to a person to something that requires active engagement. His four tasks — accepting the reality of the loss, processing the pain of grief, adjusting to a world without the deceased, and finding an enduring connection with the deceased while embarking on a new life — offer a more agentive framework than the stage model.

The task model is particularly useful clinically because it is not time-bound, it does not imply a linear sequence, and it gives both the therapist and the client something active to orient toward in treatment. The fourth task — finding an enduring connection — was revised from Worden's earlier formulation of "emotionally relocating" the deceased, and now aligns closely with continuing bonds theory.

The task framework also maps naturally onto a treatment structure. Each task has associated clinical tools and can be used to help the client and therapist understand where the work is needed.

Meaning Reconstruction (Neimeyer, 2001)

Robert Neimeyer's Meaning Reconstruction model addresses the dimension that the stage model largely ignores: the disruption to the client's assumptive world — their sense of who they are, what the world is like, and what the future holds. Significant loss does not only produce emotion; it disrupts narrative coherence. The therapeutic task is helping the client construct a new narrative that authentically integrates the loss.

Meaning Reconstruction is particularly important for losses that involve strong identity disruption — the death of a child (disrupting the parent identity), the loss of a spouse of many decades (disrupting the partner and family identity), or the loss of a future that was planned around the deceased. These clients often report that the emotional processing work is manageable but that they genuinely do not know who they are any more — and that is the clinical work that meaning reconstruction addresses.

Continuing Bonds Theory (Klass, Silverman & Nickman, 1996)

Continuing Bonds theory challenged the grief work hypothesis — the idea that the goal of bereavement is severing the emotional bond with the deceased — by demonstrating that maintaining an ongoing, transformed relationship with the person who died is not pathological but normative and adaptive. The clinical implication is significant: the therapeutic goal is not detachment but transformation of the relationship.

This framework is especially important as a corrective to the five stages model's implied destination of acceptance-as-moving-on. Clients who feel they have not "moved on" — who still talk to the deceased, who feel the person's presence, who make decisions with reference to what the deceased would have valued — are not stuck. They are engaging in a healthy and adaptive form of continuing bonds. Naming this explicitly can be profoundly relieving for clients who have been judging themselves against a cultural norm of closure that the research does not support.

---

How to Use the Five Stages Clinically — Without Being Constrained by Them

Given how culturally embedded the stage model is, abandoning it entirely in clinical practice is neither practical nor necessary. The more useful clinical approach is to work with what clients bring while expanding their framework.

Start where the client is. If a client arrives referencing the five stages, meet them there. Acknowledge what's useful in the framework — that it names grief as a real psychological process, that it validates strong emotions including anger. Then open it up.

Introduce the Dual Process Model as a complement, not a correction. Rather than telling clients the stage model is wrong — which can feel dismissive of the framework they've found meaningful — present the DPM as additional information. "A lot of people find the stages helpful as a starting point. There's also some more recent research that describes grief a little differently — would it be useful to look at that together?" This preserves the therapeutic alliance while introducing a more current framework.

Address the linear expectation directly. Many clients are suffering partly because they believe their grief should be progressing toward acceptance in a way it isn't. Explicitly naming that grief is not linear — and that oscillating, returning to intense grief after a period of relative calm, and continuing to feel the loss years later are all normal — is often among the most therapeutic things a clinician can do in early grief work.

Reframe acceptance. When acceptance comes up — as it will — it's worth exploring what the client means by it, and what they think it requires of them. Many clients believe acceptance means being okay with what happened, or no longer missing the person. A more clinically accurate reframe: acceptance means acknowledging the reality of the loss and finding a way to carry it forward — not being glad about it, and not leaving the deceased behind.

Use the stages as a normalisation tool, not a progress map. The most clinically appropriate use of the stage framework is as a catalogue of normal grief emotions — useful for validating that what the client is experiencing is a recognised grief response, not a sign that something is wrong with them. Not useful as a roadmap for where the client should be going next.

---

Talking to Clients About the Five Stages

A few common clinical conversations around the stage model, and how to navigate them:

"I feel like I should be at acceptance by now."

This is an invitation to explore what the client believes about how long grief should take, and where that belief comes from. Gently normalise that there is no universal timeline, explore what "acceptance" means to them, and introduce the idea that adaptation — rather than acceptance as an endpoint — is a more useful frame. The DPM is helpful here: attending to life and having periods of relative calm is not the same as having finished grieving.

"I keep going back to anger — I thought I'd already done that stage."

A direct opportunity to reframe the non-linear nature of grief. The wave model is useful — grief comes in waves, triggered by different things at different times, and returning to an emotion that felt processed is not regression. It is how grief works. Normalise, then explore what has triggered the return to anger and what it might be communicating.

"My family says I'm stuck in the depression stage."

This one requires more careful handling. The family's concern may or may not be clinically warranted. Validate the client's experience first. Then explore both the family's concern and the clinical picture — duration, functioning, specific PGD features if relevant. Psychoeducation about the difference between grief and clinical depression may be useful here, as may a gentle discussion of what Prolonged Grief Disorder actually involves if the presentation warrants it.

"I've accepted it but I still feel sad — does that mean I'm not really at acceptance?"

A useful reframe opportunity. Acceptance and sadness are not mutually exclusive — this is one of the most common misconceptions the stage model generates. A client can fully accept the reality of their loss and continue to feel its weight. That is not incomplete grief; it is the ongoing experience of loving someone who is no longer here.

---

What Therapists Need to Know: A Summary

The five stages of grief are culturally ubiquitous and clinically limited. They name real emotional experiences and give bereaved people a framework for normalisation — but they imply a linearity, a finish line, and a universality that the evidence does not support.

Contemporary grief practice draws on frameworks with stronger empirical foundations: the Dual Process Model for understanding oscillation between loss and restoration, Worden's Tasks for providing an active and agentive structure, Meaning Reconstruction for addressing identity and narrative disruption, and Continuing Bonds for reframing the goal from detachment to transformation of the relationship.

The clinical skill is not replacing the stage model but expanding beyond it — working with what clients bring, meeting them where their understanding is, and introducing more nuanced frameworks as the therapeutic relationship supports it.

• Grief & Loss Worksheet Bundle — 20 done-for-you worksheets grounded in current grief frameworks
• How to Support Clients Through Grief: A Therapist's Clinical Guide — the full clinical arc from assessment to rebuilding
• Best Grief Counselling Worksheets for Therapists — curated tools across all frameworks
• What Is Prolonged Grief Disorder? — DSM-5-TR criteria, differential diagnosis, and treatment