There's a diagnosis that a lot of people clearly have, that describes something very real, and that officially does not exist. That's the shape of what Daniel sent us this week.
The ghost diagnosis.
Here's what he wrote. He starts from what he calls a paradox: chronic PTSD is a controversial diagnosis that has never been accepted into the DSM, and yet the link between early life trauma and later chronic mental health conditions is about as well established as anything in psychiatry. People raised by emotionally abusive parents report difficulties that look a lot like the difficulties of people abused in other ways. The suffering is real, the outcomes are real, and the fight seems to be over which box gets ticked.
Which is the part that would drive a clinician up the wall.
And then he pulls two threads. First, why chronic PTSD is inherently difficult for the way trauma responses are conceived in the narrower, classic PTSD diagnosis. Second, something he said to a friend recently: that many people in these situations can't seal the door on the harmful person. You can learn gray rocking, you can learn every technique going, but if there's periodic contact, the wounds never even begin to heal, because every conversation is another chance for that person to get under your skin and undo the work you did in therapy. So is it reasonable, from a treatment standpoint, to bucket that person together with someone whose traumatic event happened early but is in the past.
That's the question.
His framing, not mine: he isn't trying to be academic and he isn't trying to minimize anyone's suffering. He wants to know whether the past versus ongoing distinction is a valid line that actually exists in the discourse.
It does. It's just not a line anybody has drawn on the official map.
So we'll take it in his order. The diagnostic fight first, then the past versus ongoing question and what it means for recovery.
Start with the terminology, because Daniel's premise is correct in a way that matters. There is no diagnosis called chronic PTSD in DSM-5 or DSM-5-TR. Not as a category. What exists is a chronic specifier, which just means symptoms lasting more than three months. That's a duration stamp on the same diagnosis, not a different animal.
So "chronic PTSD" is a description, not a diagnosis.
Right. The actual competitor is complex PTSD, CPTSD, which went into ICD-11 in 2018 and into clinical practice for WHO member states in 2022. And it is not in the DSM. ICD-11 carved out a separate category for it; the DSM went the other direction and just widened PTSD to swallow the symptoms.
Two manuals, two philosophies, one patient in the middle.
And here's the thing that took me a while to see clearly, because I'd been conflating them too. There are two separate axes in this conversation and they get mashed together constantly. CPTSD is about what kind of trauma. Interpersonal, prolonged, inescapable, usually starting young. Continuous traumatic stress, CTS, is about whether the threat has ended. Those are different questions.
And Daniel's question is the second one.
Almost entirely. Which is the less institutionalized of the two. CPTSD at least made it into a manual. CTS made it into neither.
Scale, before we go further. How big is this?
Seventy point four percent of people globally are exposed to at least one traumatic event in their lifetime. Only five point six percent develop PTSD. And childhood maltreatment accounts for roughly forty-five percent of the population attributable risk for childhood-onset psychiatric disorders. So the exposure is near-universal and the diagnosis is rare, which tells you the diagnosis is doing something specific and narrow.
Narrow is the word. Let's take the first thread, then. Why does the classic model struggle so badly with trauma that doesn't stop?
Because PTSD is unique among psychiatric diagnoses in requiring exposure to a specific event. That's the architecture. DSM-III defined trauma as a stressor outside the range of usual human experience. A thing that happened. A discrete, datable, past thing.
And the whole diagnostic apparatus is built on that assumption.
Built on it and load-bearing. Judith Herman's Trauma and Recovery in 1992 is the original challenge. She defined complex trauma as prolonged and repeated trauma where escape is difficult or impossible. Notice what's missing from that definition. There's no event. There's a condition.
A condition you live inside rather than an event you survive.
And the field split over it. ICD-11 recognized CPTSD as its own category. CPTSD is core PTSD symptoms, so re-experiencing, avoidance, hyperarousal, plus what they call a triad of self-organization disturbances: affect dysregulation, negative self-concept, and interpersonal difficulties. The DSM-5 instead expanded PTSD to cover the same ground without creating a new box.
And that expansion worked out beautifully, I assume.
It produced six hundred and thirty-six thousand, one hundred and twenty ways to have posttraumatic stress disorder. That's Galatzer-Levy and Bryant, counting the symptom combinations that qualify under DSM-5. Six hundred and thirty-six thousand.
That's not a diagnosis. That's a menu.
That's the argument. A diagnosis with six hundred thousand valid presentations is arguably less coherent than the one it replaced.
Why did the DSM reject CPTSD in the first place? There must have been a reason.
The story that gets repeated is that DSM-IV rejected it because research showed ninety-two percent of people with CPTSD also met criteria for PTSD. So the reasoning was, why add a category when the existing one catches almost all of them. I'd flag that as a secondary-source claim, I haven't seen the original working papers, but it's the version that circulates.
It's a strange argument, though. If ninety-two percent overlap, that's a reason to ask whether your categories are cutting reality at the joints.
Lehrner and Yehuda made the sharper version of that in 2020. Their claim is that the ICD-11 dichotomy suggests a difference in severity rather than a difference in category. And then they land the line that I think is the most useful thing in this whole literature: separating CPTSD perpetuates the expectation that simple PTSD can be easily treated with brief trauma-focused therapy.
Meaning the clean category was always a bit of a fiction.
Meaning the patients who fit the tidy three-symptom picture primarily exist in clinical trials. Their words. Patients with PTSD who are both eligible for and willing to embark on brief manualized trauma-focused therapy represent a minority. Dropout from that therapy is cited as high as fifty percent in some work.
Half the people walk out.
And the counter-critique is real too, I should say. Resick and colleagues in 2012 argued CPTSD lacked a clear definition, reliable and valid assessment measures, and support for convergent, discriminant, and incremental validity. So you had two camps in American psychiatry. The DESNOS people on one side, the PTSD-alone-is-adequate people on the other. It wasn't a settled question that got decided. It was a fight that one side won institutionally.
Which brings us to emotional abuse, because that's the cleanest case of real suffering with no clean box.
It's the purest example. Dye, 2020, studying seven hundred and forty-eight college students, found that people reporting emotional abuse had higher scores for depression, anxiety, stress, and neuroticism than people reporting only physical abuse, only sexual abuse, or combined physical and sexual abuse.
Higher than the combined category.
Higher than all of them. Hart and colleagues back in 1996 went further and said emotional abuse may be the most damaging form of maltreatment, causing adverse developmental consequences equivalent to or more severe than other forms. And Glaser in 2002 gave the reason, which I think is the key structural insight: it's most damaging because the perpetrator is almost invariably the person responsible for enabling the child to fulfil their developmental tasks. The primary carer.
The person you need is the person doing it.
That's the whole thing in one sentence. Stranger violence is terrible but it's external. This is the attachment figure. There's nowhere to go.
And the brain evidence backs the equivalence?
Teicher and Samson, 2016. Parental verbal abuse, witnessing domestic violence, and sexual abuse appear to specifically target brain regions and pathways that process the aversive experience. Auditory cortex for the verbal abuse, visual for the witnessing, somatosensory for the physical. The maltreatment shapes the system that perceives it.
The instrument gets tuned to the abuse.
And the epidemiology is the cleanest demonstration. Rehan and colleagues, 2017, ten thousand nine hundred and eighty Finnish adults. Severe emotional abuse associated with depression and anxiety at an odds ratio of three point eight zero. Severe physical abuse, three point nine seven.
Those are the same number.
Within noise of each other. Martins and colleagues found the same pattern in a smaller psychiatric sample, emotional abuse tracking with psychopathology, particularly mood disorders. So the suffering is equivalent, the outcomes are equivalent, and the diagnosis is not.
So the kind of trauma is one axis, and on that axis the field has basically conceded the point everywhere except the DSM.
ICD-11 conceded it. The research literature conceded it decades ago. The DSM has not moved.
But Daniel's second thread is about a different axis entirely. Not what kind of trauma, but whether the threat has actually ended.
And that's where it gets interesting, because this axis is even less formalized. The construct is continuous traumatic stress. Coined by Gillian Straker and colleagues in South Africa, developed by Eagle and Kaminer in 2013. Their framing: CTS captures a domain of traumatic stress experience not adequately formulated in the existing repertoire of traumatic stress responses.
Not adequately formulated. That's a polite way of saying the vocabulary doesn't have the word.
It doesn't. And the core distinction is stated cleanly by Goral and colleagues in 2022. In single traumatic exposures, stress symptoms relate to a past, no longer dangerous event. In ongoing exposure to threat, stress symptoms may be related both to the past and to potential future events.
The past and the future at the same time.
Which is a different psychological structure. Classic PTSD is a memory problem. This is a memory problem plus a vigilance problem plus an anticipation problem, all braided together.
And here's the part that I think is the direct answer to what Daniel was getting at.
It is. Same paper. In ongoing exposure to threat, stress symptoms may be relieved when the source of threat is discontinued. A phenomenon not commonly seen in typical, chronic PTSD.
Read that back to me, because that's the whole argument.
The symptoms lift when the threat stops. That's not how classic chronic PTSD behaves. Classic chronic PTSD persists after the event, that's the defining feature. So the ongoing-threat population isn't just more severe. It's different in kind. The symptom is tracking a live variable.
The wound is being reopened on a schedule.
And the measurement work supports it. The CTSR scale, Goral and colleagues, PLOS One, 2021. Validated on three hundred and thirteen Israeli adults living near Gaza, where the shelter warning time was fifteen to thirty seconds.
Fifteen seconds to get to a shelter.
Fifteen to thirty. Three factors came out of that validation: exhaustion and detachment, rage and betrayal, and fear and helplessness. Related to but distinct from PTSD. That rage and betrayal factor doesn't map onto the classic clusters at all.
Rage at who?
That's the question, isn't it. In the Gaza sample it's plausibly the state, the situation, the people who are supposed to protect you. In an abusive-parent context it maps almost too neatly onto the parent.
And the construct is traveling. It's not just a South African or Israeli thing anymore.
Avramchuk and colleagues, this year, adapted the CTSR for one thousand nine hundred and two Ukrainian military personnel. Their conclusion: the traditional PTSD paradigm is insufficient for repetitive threat. That's the construct proving it generalizes.
So we have a named phenomenon, validated scales, replication across three countries, and it's in neither manual.
In neither manual. Maercker and Brewin in the British Journal of Psychiatry last year took what they called a look at the potential diagnosis of continuous traumatic stress reaction. Potential. Future tense. That's the most official attention it has received.
And the title of the Goral paper is almost too on the nose.
In the Middle, between Anxiety Victims and PTSD, There Are People That Have Some Kind of a Disorder That Has No Name Yet. That's the actual title.
That's Daniel's paradox with a citation attached. It's not that the patients fail to fit the boxes. It's that the boxes fail to fit the patients.
And that reframing matters, because the default assumption when someone doesn't fit is that something is wrong with the someone.
So now the treatment question. Is it reasonable to bucket these two people together?
From a treatment standpoint, the honest answer is that the field has not tested it. I looked for a head-to-head trial, ongoing exposure versus resolved past trauma, same treatment, compared outcomes. It doesn't exist. That's a genuine gap, not a gap I'm papering over.
So Daniel's question is open in the literature.
Open. But there are pieces that bear on it. The ISTSS phase-based model, 2012. Phase one stabilization and safety, phase two trauma memory processing, phase three reintegration. Eighty-four percent of fifty expert trauma clinicians originally advocated for that structure. ISTSS has since moved toward a more personalized approach, but the phase model is still the spine of a lot of practice.
And phase one is safety.
Phase one is explicitly ensuring the person's safety. Which is the point where the abstract distinction becomes a concrete problem. If the abusive person is still in periodic contact, you cannot complete phase one. You're being asked to stabilize someone who is still being destabilized.
You can't build the floor while someone's still pulling up the boards.
Here's the counterweight, because I don't want to pretend the field agrees with Daniel. NICE, 2018, is explicit: treatment for PTSD should not be delayed or withheld in cases where there are ongoing court proceedings or applications for compensation.
They've already faced the ongoing-stressor question.
They faced it and they came down against deferral. The logic being that waiting for the stressor to end can mean waiting forever, and untreated PTSD compounds. So the official guidance is, treat anyway.
Which cuts against the intuition that you should wait until the door is sealed.
It cuts against the deferral version of it. It doesn't answer the bucketing question. Treating someone with an ongoing stressor isn't the same as claiming they're identical to someone without one.
What about the outcome predictors? Is ongoing trauma more or less treatable?
Worse outcomes track with more severe symptoms before treatment and comorbid depression. Better outcomes track with more recent trauma. So recency helps. Which is interesting, because ongoing trauma is by definition recent, but it's also by definition not over, and those pull in opposite directions.
The recency effect might be doing something different in this population entirely.
That's my read, and I'd hold it loosely. The recency finding comes from people whose trauma is in the past. Applying it to people whose trauma is live is exactly the kind of extrapolation that's missing a trial.
The severity versus category fight cuts both ways here.
It does. If Lehrner and Yehuda are right that the CPTSD split is really severity dressed up as category, then Daniel's bucketing question may not have a categorical answer at all. It may be a matter of degree, plus one binary fact: is the stressor still live.
The axis Daniel identified might be the one categorical thing in the whole picture.
That's a real possibility. The kind of trauma might be a spectrum. Whether the threat has ended might not be.
There's a detail I keep circling. Leshem, this year, describing therapists working under ongoing threat as living in a perpetual waiting room state. That's the clinicians. Imagine the patients.
The waiting room is a good image for it. Nobody's in treatment, everybody's in triage.
That gap between recognizing a thing and doing something about it is where Hilbert has something to add.
Hilbert: The form had a box for it.
Sorry, go on.
Hilbert: Ongoing stressor. Checkbox, about a third of the way down the intake sheet. I did nights as an intake coordinator at a psychiatric urgent-care clinic for a while. Not clinical, I just took the calls and filled in the forms. In training they told me, tick it if the patient is still in contact with the person who's harming them.
What happened after it got ticked?
Hilbert: Nothing I could see. Patient got the same intake, same waitlist, same referral. I ticked it constantly. It was a busy form. Nobody ever asked me about it, nobody ever followed up on it, and nobody ever explained what it was for.
Do you know if it was a DSM field or a local one?
Hilbert: No idea. I was a night-shift intake coordinator, not a psychiatrist. It was a box on a form. I ticked it and moved to the next page.
That's the whole episode in one checkbox, isn't it. The distinction is real enough that somebody put it on a form. It's unformalized enough that ticking it changed nothing.
The field has been quietly collecting the data anyway. Just not doing anything with it.
Hilbert: One of the nurses told me once that it was for the file. I asked what part of the file. She said the part nobody reads.
That's a hell of an image.
Hilbert: I liked the night shift. Quiet, mostly. You had the phone, the form, and a chair that was too low for the desk, which is a problem when you're built like an anteater and the screws on the housing keep catching your claws on the armrest.
The chair being wrong is its own diagnosis.
Hilbert: It was a good chair. Just the wrong height.
The checkbox is where we've ended up. Real enough to be on the form, inert enough that it changes nothing downstream.
It's the perfect miniature of the whole problem. Clinical recognition exists. Clinical action doesn't follow from it. And that's not because anyone's malicious. It's because the box was never wired to anything.
Which is where the research sits too. Maercker and Brewin gesture at continuous traumatic stress reaction as a potential future diagnosis. Potential. The Ukrainian and Israeli work is gaining traction. But the DSM hasn't moved on CPTSD, and there's no sign it's going to move on CTS.
The construct is doing the work in the literature and none of the work in the manuals.
If you take one thing from this, take the two axes. What kind of trauma, and whether the threat has ended. Daniel's question is the second one, and it's the one nobody has drawn on the map.
The second axis may be the only categorical thing in the picture. The kind of trauma might be a spectrum. Whether the door is still open might not be.
That's the episode. Thanks to Hilbert Flumingtop for producing. This has been My Weird Prompts, the human-AI collaboration podcast. If you want to send us a prompt, email us at show at my weird prompts dot com. We'll be back soon.
