After a major loss: when grief turns into depression

“It has been six months,” she said. “Everyone else has moved on. I still cook for two.”

She sat in my armchair with her hands in her lap, apologising. Not to me — to someone in her own head who had decided, somewhere along the way, that grief comes with an expiry date. Her husband died in February. It was September. And the question she brought was not “am I hurting too much”, it was “am I ill now”.

I hear that question often, and almost always late — after someone has spent months privately judging whether something is wrong with them. So this time I want to write about the boundary. Not about sadness not being depression; I have written about that already. About when grief after a major loss really does turn into something else, how to recognise it, and what we do then.

Grief is not a fault in the system

I will start where most people go wrong. Grief is not a disorder to be cured. It is the response to the fact that someone woven into your daily life, your habits and your picture of the future is suddenly not there. The pain is not a sign that you are weak. It is a measure of how much room that person took up.

So with grief my first question is not “how much does it hurt” but “where is it going”. Grief has movement. It arrives in waves that are almost continuous at first, and over time gaps open between them. In the gaps a person cooks again, laughs at a joke, gets annoyed at the neighbour’s dog. Then a new wave comes — on an anniversary, at a song in a shop, at the empty half of the bed — and takes them back. That is not a relapse. That is the shape grief has.

What is ordinary in grief, even when it is unbearable

People are frightened by things that are entirely expected in grief. Hearing his voice in the hallway. Ringing her number, because you always rang at five. Being sure you saw her coat in the supermarket. Such experiences in bereaved people are not rare and by themselves mean nothing pathological.

Anger is expected too — at the doctors, at the person who died, at yourself. So is the numbness of the first weeks, when someone runs a funeral like a project and only falls apart a month later. So is the body giving out: no sleep, an aching back, no appetite. Grief is a physiological event, not only an emotional one.

And guilt is expected. Almost everyone finds something they could have done differently. That is the mind trying to force logic onto a loss, because without logic it is unbearable. On when guilt is part of the process and when it becomes a burden of its own, I wrote in the post on healthy and toxic guilt.

Where the boundary is — and why a calendar cannot draw it

The honest answer: the boundary is not in the calendar. It is not that six months is fine and seven months is a diagnosis. What I look at is not time itself but direction.

If the gaps between the waves slowly widen — if somewhere in the first year there is a week with three bearable days — grief is moving. If the curve after twelve months looks the same as it did in the second month, if there is not a single day that is different, if life stopped on the day of the funeral and has gone nowhere since, then I am looking at something that has become stuck.

Prolonged grief disorder: what the manuals say

This stuck form has a name. The international classification of diseases has listed it since 2019 as prolonged grief disorder; the American manual added it in 2022. The description is surprisingly down to earth: longing for the dead person, or preoccupation with them, lasting considerably longer than is expected in your surroundings — at least six months by the international classification — and is strong enough that the person cannot work, look after themselves or be with others.

Alongside the longing, other features often appear: an inability to accept that the person is really dead; the sense that part of you died too; numbness; avoidance of everything that reminds you; a feeling of meaninglessness. The key word is frozen. It is not that someone grieves too much, it is that the grief goes nowhere.

How common is it? Reviews of studies in people bereaved in ordinary circumstances give roughly one in ten. After a violent or sudden loss the proportion is considerably higher. Nine in ten, then, come through without a diagnosis — which is both a good reason not to pathologise every long pain, and a good reason not to overlook the tenth.

Depression after a loss is a third thing

This is where things get mixed up, so let me separate them slowly.

In grief, and in prolonged grief, the person is in the foreground. Thoughts circle around them. The pain has an address. Self-regard usually stays intact — someone does not hate themselves, they simply miss.

In depression the damage spreads out. It is no longer focused on the loss but on everything. There is no pleasure anywhere, not even in the gaps. A steady thought appears that you are worthless in yourself, that you are a burden, that you should not exist — and that is a qualitatively different thought from “I want to be where he is”. The first is against yourself; the second is against separation.

Depression can be set off by a loss; the death of someone close is one of the strongest known triggers. It can also occur alongside prolonged grief. So I do not care about the distinction for the sake of a label, but because the two call for different answers.

Why the bereavement exclusion was removed, and why that was contested

Until 2013 the diagnostic manual contained a rule: if someone had lost a close person in the previous two months, you did not usually diagnose a depressive episode. Then the exclusion was removed. The row that followed still echoes in the field today.

The argument for removing it: research showed that depression triggered by a loss resembles depression triggered by anything else in its course, its heritability, the risk of recurrence and the response to treatment. If we deny it in bereaved people, the ones who need help most are the ones left without it.

The argument against: two weeks is an absurdly short threshold after the death of a partner. We risk turning ordinary human pain into a diagnosis and a prescription. That worry was not empty, and it still stands.

I do not have a clean side in this dispute. What I have is a stance: I use a diagnosis as a map and not as a name — I wrote about that in a diagnosis is a map, not a name. If the name opens the door to the right help, it is useful. If it becomes a description of the person, it does harm.

What makes grief more likely to get stuck

It is not random who this happens to. From the research, and from what I see in my room, a few patterns keep coming back.

A sudden, violent or senseless death. The death of a child. A relationship that was very close and very dependent — or, the opposite, full of resentment you can no longer talk through with someone who is dead. Previous depression or trauma. Isolation. And a setting that expects you to be “functioning” again after three weeks.

I would add one that gets less mention: losses the people around you do not recognise. A miscarriage. A former partner. Someone you were secretly involved with. A dog that was, for ten years, the only reason you got up in the morning. When grief is given no public room, it more often gets tangled.

Signs not to wait on

Whatever anyone would call the state, there are things I do not recommend waiting on: thoughts of taking your own life, or a plan; a fast physical decline — significant weight loss, weeks without sleep; alcohol or tablets becoming the only way through the evening; no longer being able to look after your children or yourself; months of being unable to return to work.

In such a case the first step is often a doctor rather than a therapist. When it is worth going to a psychiatrist first, I set out in a separate post.

What actually helps

With grief that is moving, what helps most is that someone carries it alongside you. That they do not fix it. That they do not offer the bright side. That they can sit with silence.

Rhythm helps too: eating, going outside once a day, one fixed appointment a week. Not as a treatment, but because grief demolishes the scaffolding of a day, and an empty day makes the pain larger. And it helps not to add the reproach that you are grieving wrongly, too long or too loudly. That reproach is often what exhausts a person most, and it can be worked on; more on that in the post on self-compassion.

With grief that has got stuck, ordinary support is usually not enough. The best researched approaches are purpose-built: working with the avoidance, gradually returning to the memories a person keeps dodging, and at the same time setting very concrete goals for the life that continues. Where depression is also present, a combination with medication often works — on when that makes sense, I wrote about antidepressants and therapy. Interestingly, medication on its own does little for the longing while doing well for the depressive part. That is one more argument that the distinction is not academic.

Where this post is wrong

Now, against myself.

First: the six-month threshold is an agreement, not a discovery. Nothing tips over in the body at the end of the sixth month. The line was drawn because a system needs a line, and some researchers argue, with reason, that it is too early.

Second: the criteria were developed in a largely Western, urban, individualistic setting. Where the boundary of “expected” lies is a cultural question. A widow who wears black for a year and tends the grave every week is respectful in one setting and stuck in another. I do not judge that from a table but from one question: does this form of grieving help her live, or does it stop her.

Third, and this matters most: nothing I have written reliably sorts one individual. The curves in the picture are averages. I know people who cried every day for three years and never needed a diagnosis, and people who looked tidy after three months and were deep in trouble. The criteria are useful as questions, not as verdicts. If something here does not fit your experience, your experience is probably not the thing that is wrong.

Fourth: there is an opposite error. Not only pathologising, but its mirror image — “of course you are grieving” used as a reason for nobody to do anything. Both sentences leave a person alone.

What to remember

Grief is not measured in months but in whether it moves. Waves are expected; the gaps between them should widen over time.

Prolonged grief disorder is longing that has got stuck — pain that keeps the same address and lets no life through. Depression after a loss is something else: a spreading numbness that attacks your relationship to yourself as well. The two can appear together.

The thought “I want to be where he is” is not the same as “I should not exist”. Take the second one seriously and say it out loud — to anyone.

And this: if it feels as though your life stopped on the day of the funeral, that is not a lack of strength. It is something with a name, a description and a form of help that works. You have not missed a deadline. There is no deadline.