1. The programme at a glance
Indication. Prolonged, disabling grief in children and adolescents aged 7 to 18, at least six months after the death. The programme also suits grief complicated by a violent or sudden death, with the reservation set out in section 30.
What this programme is not. Support for every bereaved child. The great majority of children get through a bereavement without treatment, with the people around them, and systematic intervention gives them nothing — the meta-analyses show it (Currier et al., 2007; Rosner et al., 2010), and some universal interventions have null effects, or might even do harm. This manual is addressed to children who are unwell, lastingly, and whose functioning is impaired.
Reference models. Three converging contributions. Cognitive behavioural therapy for prolonged grief in children, evaluated by Boelen, Spuij and their colleagues (2015, 2021), which targets cognitions and avoidance. The multidimensional grief theory of Kaplow, Layne and their colleagues (2013), which distinguishes three forms of distress — separation, the circumstances of the death, and the impact on identity and meaning — and which underpins trauma- and grief-focused therapy for adolescents, evaluated in group format in schools (Saltzman et al., 2001; Layne et al., 2008). And the grief components of Cohen, Mannarino and Deblinger's trauma-focused cognitive behavioural therapy (2017), for cases where the death has been traumatic.
Format. Fourteen individual sessions of 50 minutes with the child, weekly, and five parallel sessions with the parent or guardian, interleaved. The sessions with the parent are not an extra: they are part of the treatment, and their omission is, in clinical experience, a major cause of failure in the child.
Target mechanism. Not forgetting, nor acceptance understood as resignation, but four changes: that the child can think of the person who died without being overwhelmed, that they stop avoiding what recalls them, that the thoughts that weigh on them — the fault, the injustice, the worry about those who remain — are examined, and that they take up the course of their development again.
| Session |
Focus |
What the session produces |
| 1 |
Welcome, assessment, alliance |
What the child knows, in their own words |
| 2 |
Formulation, goals, naming feelings |
Diagram drawn with the child |
| 3 |
When it overflows: the tools |
Three tools tried in session |
| 4 |
Who this person was |
First memories, written or drawn |
| 5 |
Understanding the death, the facts |
The unanswered questions, listed |
| 6 |
The narrative, part one |
Beginning of the narrative, written |
| 7 |
The narrative, the hardest moments |
Complete narrative |
| 8 |
"I should have": the fault |
Belief examined, share reattributed |
| 9 |
Injustice, anger, "why" |
Anger named and expressed in a bounded form |
| 10 |
Taking back what is avoided |
Two avoidances lifted |
| 11 |
What I keep |
Memory object, or letter |
| 12 |
What has changed for me |
Identity, role, future |
| 13 |
The sharing with the parent |
Narrative read or told, prepared |
| 14 |
Stocktake, plan, the dates that come round |
Personal plan written by the child (worksheet 7) |
The last column shows what the session should have produced; the criterion for moving on, at the end of each session (sections 14 to 27), spells it out.
| Parent session |
Placement |
Focus |
| P1 |
Before session 1 |
Assessment, information, their own grief |
| P2 |
After session 5 |
Answering questions, talking about the death |
| P3 |
After session 7 |
Preparing the sharing, tolerating the narrative |
| P4 |
After session 10 |
Routines, boundaries, school, siblings |
| P5 |
After session 13 |
The dates, the future, the family plan |
What the child takes away. Seven printable worksheets, listed in section 33 and downloadable from this page: what I know, my feelings, my tools, my memories, my questions, what I am taking back, my plan.
What sets this programme apart from bereavement support. The target is not sadness, which is not a symptom and does not have to disappear. The target is what prevents grief from running its course: avoidance, the thoughts that weigh, and the absence of an adult able to talk about it. A child who cries for their mother two years on is doing well; a child who refuses to hear her name is not. That difference governs the whole treatment.
2. Before you start
The prior question: should you treat?
This is the most important decision in the manual, and it is taken before session 1.
Do not treat a child who is doing well. The great majority of bereaved children do not need therapy. They need an adult who answers their questions, a school that has been told, rituals they are allowed to take part in, and time. Interventions offered to all bereaved children, regardless of their state, achieve very small effects, and there are grounds for thinking they can do harm: they suggest to a child who was doing well that they ought not to be.
Treat a child whose functioning is impaired beyond six months: school refusal, withdrawal from peers, lasting regression, consistently disturbed sleep, massive avoidance of everything that recalls the person who died, thoughts of death, or separation distress that does not ease.
Wait, in the first few weeks, except where there is risk. The month after a death is not the time for a protocol. What is useful then: information for the parent, an answer to the child's questions, a school that has been told, and an appointment set for three months' time.
How to use it
Read the whole thing before session P1. Sections 3 to 13 lay down the frame; sections 14 to 28 are worth rereading the day before each session.
Every session is described on the same frame: the objective, a step-by-step run-through, what you say, the common mistakes, and the criterion for moving on. That last point sets the pace: the programme advances by acquisition, not by calendar: you move on to the next session when its criterion for moving on is met, and not because a week has gone by. With children, the gap between the two is wider than with adults.
3. What is normal, and what is not
Ordinary childhood grief
It is intense, discontinuous, and it disconcerts adults. Three features in particular.
It alternates. A child cries for ten minutes then goes off to play. That alternation is neither indifference nor denial: it is the way a child bears a pain they cannot hold for long. Parents often misread it — "he did not understand", "she does not care" — and it has to be explained to them.
It comes back in waves, over years, at anniversaries, at festivals, at milestones — starting a new school year, an exam, a first love, a birth. A child who cries for their mother at sixteen, having lost her at ten, is not relapsing: they are growing up with that loss, and they cross it again at every age with what they newly understand.
It shows itself through the body and through behaviour more than through words: stomach aches, headaches, sleep, tempers, regression, falling school results, repeated questions about details.
What causes needless worry
Saying little about the person who died. Not crying. Playing at death or funerals. Asking the same factual question ten times. Saying blunt things. Laughing at the funeral. Talking aloud to the dead person, or saying that they see or hear them: this is frequent, it is not a psychotic sign, and it is not treated as a symptom.
What should cause concern
After six months, and lastingly:
massive avoidance of everything that recalls the person — their name, their photographs, their room, the places, the subjects;
separation distress that does not ease: the child can no longer be apart from the surviving parent, sleeps in their bed, refuses school, checks that they are alive;
fixed guilt, with an explanation in which the child is the cause;
prolonged withdrawal from peers and activities;
a drop at school that does not recover;
thoughts of death, the wish to join the person who died, or a plan;
a conviction that life no longer has any meaning, in adolescents above all;
lasting regression in a younger child: toileting, language, separation.
These are not signs of weakness, and time alone cannot be counted on to resolve them. This is the indication for this programme.
The ages, and what they change
From 7 to 9. The child understands that death is final and universal, but applies that understanding poorly to themselves and to their own family. Magical reasoning is still active: they may believe that something they thought or wished caused the death (session 8). The questions are concrete and repeated, about the body, the grave, what happens afterwards. Answer them concretely: that is what soothes.
From 10 to 12. Understanding is close to an adult's, and the worry concerns the consequences: who is going to look after me, is the other parent going to die too, are we going to move. Shame appears: the child does not want to be the one whose father died. They begin to protect the surviving parent by not talking.
From 13 to 18. The impact falls on identity and on meaning. The adolescent asks who they are without that person, what becomes of their future, and whether any of it means anything. They share with peers rather than with family, and they may take risks. Suicide risk is real here and must be assessed at every session in severe cases. Alcohol and substance use appears as an avoidance, and it has to be looked for.
4. The clinical picture
The three forms of distress
Multidimensional grief theory, proposed by Kaplow, Layne and colleagues (2013), distinguishes three forms that call for different interventions. This distinction is the best formulation tool we have in children, and it organises this programme.
Separation distress. The child suffers from the absence: they look for the person, miss them, cannot bear the empty places, do not want to be apart from those who remain. It is the most visible form, and the one worked through memories, continued bonds (section 5) and the gradual lifting of avoidance.
Distress related to the circumstances. The child is preoccupied with how the death came about: what they saw, what was kept from them, what was done or not done, who is responsible. It is recognised by insistent questions about details, and it is worked through the narrative and through the examination of beliefs.
Existential and identity distress. The child, and above all the adolescent, is affected in who they are and in what they expect of life: I am not the same, I have no future, nothing means anything, I do not deserve to be happy. It is the least often spotted form, the most closely linked to suicide risk, and it is worked in sessions 11 and 12.
Most children show all three, in varying proportions. At assessment (section 10), estimate the weight of each from the signs described here: it is the proportions that decide how much room each part of the programme will take.
What keeps grief stuck
Avoidance. This is a central factor, as in anxiety disorders. The child avoids the reminders — the name, the photographs, the room, the house, the anniversary — and each avoidance prevents the discovery that they can think about it without being destroyed. Adults, in good faith, join in: the photographs are put away, the name is no longer spoken, the subject is changed.
The thoughts that weigh. Three families recur. The fault: I could have, I said something terrible, it happened because I wished it. Injustice and anger: it is not fair, I was lied to, someone should have done something. And anticipation: the others are going to die too, I am going to be left alone, I will not manage.
The silence of adults. This is the maintaining factor specific to childhood. A bereaved parent who cannot talk, who cries as soon as the subject comes up, or who lied about the circumstances, leaves the child alone with their own hypotheses — which are often worse than the reality. The child, for their part, keeps quiet to protect the parent. This pact of mutual silence is extremely common, and it is a direct target of treatment.
The breakdown of routines. A bereaved child whose meals, bedtimes, school and activities have fallen into disorder is unwell independently of their grief. Restoring routines is a simple and direct intervention, and it goes through the parent.
The state of the surviving parent. This is one of the best established predictors of how the child fares. A parent who is depressed or in severe prolonged grief struggles to support their child, and their own treatment is part of the child's care plan.
Epidemiology, in two useful figures
About 4 to 5 % of children in Western countries lose a parent before eighteen (Melhem et al., 2011), and a far larger proportion lose a close grandparent, a brother or sister, or a friend.
Among bereaved children, a minority develop full prolonged grief. Estimates are scarce and depend on the sample and on the criterion used (DSM-5-TR, ICD-11 or a questionnaire cut-off): after the sudden death of a parent, about one child in ten still had intense and almost unchanged grief nearly three years after the death, more often after an accident or a suicide than after a natural death (Melhem et al., 2011). It is that minority this programme concerns.
5. The model that guides this programme
Why this one
Three lines of work converge, and this programme assembles them.
The cognitive behavioural model of prolonged grief (Boelen et al., 2006) attributes the blockage to three mechanisms. The first is insufficient integration of the reality of the loss into autobiographical memory: the loss is not yet connected to what the child knows about themselves and their life, and it keeps an unreal quality. The second consists of negative beliefs about oneself, the world and the future. The third groups together avoidance strategies: avoidance of reminders, but also rumination and seeking proximity to the person who died. This model has been specifically adapted and evaluated in children and adolescents, in an open study, without a control group, and then in a controlled trial (Spuij et al., 2015; Boelen et al., 2021).
Multidimensional grief theory (Kaplow et al., 2013) supplies the three-domain formulation set out in section 4, and the treatment principle that follows from it: you do not attack at the same point depending on the dominant form.
The grief components of trauma-focused therapy (Cohen, Mannarino and Deblinger, 2017) contribute two elements that are essential in children: the gradual narrative, written or drawn, as a way of approaching what is unbearable in small doses; and the joint parent-child work, with a prepared sharing of the narrative.
The five levers
Lifting avoidance. It is done gradually, concretely, and by negotiating each step with the child: the photographs, the name, the room, the places, the subjects (session 10).
The narrative. It is built in sessions 6 and 7. It is not a catharsis, it is a work of ordering: what happened, in what order, with the facts corrected where the child had got them wrong.
The beliefs. The fault above all (session 8), then injustice (session 9). Anticipation, the fear that the others will die too, is addressed from session 5 onwards, together with the fear of contagion and repetition: that the same thing will happen to the child themselves or to those who remain.
The continued bond. The aim is not to detach the child from the dead. It is to turn a bond made of absence and pain into a bond made of available memories. A child who can talk about their father and laugh is doing better than a child who no longer talks about him.
The parent. The lever is their capacity to talk about the death, to bear their child's grief, and to hold the routines; the five parent sessions (section 28) work on it. Without this lever, the other four do not hold.
What the model implies you should not do
Do not aim for the end of grief. The aim is not that the child should stop being sad, nor that they should "move on". Say so to the child and the parent in session 1: nobody here will ask you to forget, and there is no moment at which one is supposed to have finished.
Do not force the narrative. A child made to tell their story before they have regulation tools leaves the session dysregulated and does not come back. Session 3 comes before session 6 for that reason.
Do not reassure about what you do not know. To "is Mum going to die too?", the honest answer is bounded, not reassuring at any cost: it sticks to what is known, for example that Mum is in good health today, without promising what nobody can guarantee.
Do not work without the parent. A child treated alone, who goes home to a house where the name is forbidden, loses their gains between sessions.
Do not use the stages of grief, the model that describes a succession of phases, from denial to acceptance. They have received no empirical validation, and they create false expectations.
Frequently asked questions
From when should you treat, and when should you wait?
Wait during the first few weeks, except where there is risk: what is useful then is information for the parent, answers to the child's questions, a school that has been told, and an appointment set for three months' time. Treat beyond six months when functioning is impaired: school, peers, sleep, or massive avoidance. And do not treat a child who is doing well, even if their loss was terrible.
The parents say the child "showed nothing". Should we be worried?
Not in itself. Many children do not cry, play on the same day, and appear indifferent. It is the way a child bears a pain they cannot hold for long. What should cause concern is not the absence of tears, it is the avoidance, the withdrawal, the drop at school and the duration. Explain it to the parent: it is often the first thing that relieves them.
Should a child be told their parent died by suicide?
Yes, with words adapted to their age, by the surviving parent, prepared, and before the narrative (section 30). The question is not whether to tell them but when and how: the child will almost always find out, often from a third party or from a search, and discovering a lie damages trust lastingly. A usable formulation: "he died because he had an illness in his head that made him suffer a great deal, and that illness made him believe there was no other way out." Never the details of the method.
A child says they see or hear the person who died. Is that worrying?
No, in the great majority of cases. These experiences are frequent in bereaved people of all ages, they are not psychotic, and they are often consoling. Do not treat them as a symptom and do not try to correct them. What does warrant assessment: hallucinations unconnected with the person who died, disorganisation, or a delusional conviction that the person is not dead.
The child refuses to talk. What do I do?
Do not force it, and do not make the silence the issue. Three things work: begin with what is not the death — their life, their memories, session 4 before session 6; go through a medium that does not require speaking — drawing, writing, making a book; and work with the parent in the meantime. An adolescent who comes under duress deserves an explicit negotiation: find a goal that is theirs, however modest, and remind them of the confidentiality rule (section 13): you will report nothing of what they tell you to their parents, unless they are in danger or someone is hurting them.
Should a child be taken to the funeral?
Offer it, explain precisely what they will see, and let them choose. A prepared child who chooses to go is the better for it; a child taken by force or kept away without explanation keeps a wound. When the funeral has passed and the child was kept away, it cannot be made up for — but something else can be built: a prepared visit to the grave, a family ceremony made for them.
The parent is very unwell and cannot support their child. What do I do?
Say so, with tact and without making it a condition of treatment: "what you are going through deserves care of your own, separate from your son's, and it is also one of the things that will help him most." Then adapt: strengthen the parent sessions, mobilise another reliable adult — a grandparent, an aunt, a school link person — and concentrate the work on what does not depend on the parent. And be aware that the child's prognosis is tied to the adult's: that is one more reason to press.
Should the narrative be done with every child?
With those whose distress bears on the circumstances of the death — the insistent questions about details, the blanks, what was seen — yes, and it is the heart of the treatment. In a child whose distress is mainly separation distress, with no trauma and no dark areas, the narrative can be shorter and more room given to sessions 4, 10 and 11. Adapt according to the three forms of distress in section 4.
The child is worse after the narrative. Is that normal?
Yes, for a few days: nightmares, irritability, a return of avoidance. It is expected, and that is why the family has to be warned beforehand. A parent who has not been warned concludes that the therapy is doing harm and stops the treatment — a classic cause of dropping out, and entirely avoidable. If the deterioration lasts more than two weeks, review the pace and the return to calm at the end of the session.
What do I do when a child asks "why"?
Do not manufacture an answer. The formulas of consolation — "he is better off where he is", "it was his time", "you have to be strong now" — are reported years later, word for word, with anger. Say what is true: "there is no reason. It is not fair, and it never will be." If the family has a religious answer, you can refer to it without endorsing it.
An adolescent has been drinking and going out at night since his father died. Where do I start?
With the drinking, which prevents everything else, and with the risk. Name it as an avoidance, without moral judgement, and set a reduction goal before entering the narrative. Assess suicide risk and make a safety plan (session 3, step 6). Then take up the programme, giving more room to sessions 11 and 12: in adolescents, it is the impact on identity and meaning that feeds the behaviour.
Is medication needed?
It has no indication of its own in childhood grief. It is addressed to a comorbid depression or anxiety that meets criteria, assessed as such, and prescribed by a child and adolescent psychiatrist. Do not treat sorrow.
Bereavement groups for children, yes or no?
They are probably useful against isolation and for normalisation — meeting other children in the same situation often brings relief. They do not replace treatment in a symptomatic child, and they are little evaluated. A child who is unwell needs both, in this order: the treatment first, the group as a complement.
Should the room be cleared, the belongings given away, the house sold?
These are not clinical decisions, and it is not for you to take them. What you can lay down is a principle: tell the child in advance, leave them a choice where possible — keeping an object, choosing what they want — and never let them discover a fait accompli. A room cleared in their absence is a frequent and lasting wound.
What do I do when the parent has a new partner and the child reacts badly?
That is frequent, and it is not a relapse of the grief. Two useful points: the child often feels it is a betrayal of the dead person, and they are afraid of losing the place they have left. Both can be voiced, and are worked on in session 12 and, with the parent, in sessions P4 and P5 — asking the parent above all to announce things in advance and to preserve a time that belongs to the child alone.
How long does a child take to get better?
The programme lasts four months, and the improvement is generally clear by the end. But it does not close the grief: the child will cross it again at every stage of their life. Say so to the family in session 14, failing which the first difficult anniversary will be experienced as a failure of the treatment.