C-PTSD: the understanding we need now

I Imagine a woman who has sought help for difficulties concentrating, restlessness and exhaustion. After an assessment, ADHD is entered in her patient record. Medication may help her stay focused and still leave something crucial untouched. She wakes with her body on high alert. A raised voice makes her freeze. In close relationships, she waits for warmth to turn into threat, just as it did when she was growing up.

Perhaps ADHD is part of the explanation. Perhaps not. Perhaps several processes are present at once. All we know in this composite case is that a coherent trauma history was never taken.

What happens when a diagnosis becomes the answer before the question has had time to be formulated? When healthcare asks what symptoms a person has, but not what she has been through?

Well established: trauma is a risk, not a destiny

A potentially traumatic event is not the same as a persistent trauma response, and a trauma response is not automatically PTSD or C-PTSD. According to the WHO, around 70 per cent of the world’s population experience such an event at some point. Among those exposed, a minority, around 5.6 per cent, develop PTSD during their lifetime. Trauma, then, is common. PTSD is not.

This is an argument against determinism, not against the importance of trauma. People may recover, experience temporary difficulties, develop other problems or have a lasting post-traumatic disorder. Trauma is neither a diagnosis nor a destiny.

ICD-11 is also clear about the boundaries. C-PTSD, code 6B41, first requires the core of PTSD: re-experiencing, avoidance and a persistent sense of threat. In addition, there must be severe and persistent disturbances in self-organisation (DSO): affect regulation, negative self-concept and relationships. The symptoms must cause significant functional impairment. Difficulties with concentration and memory may occur, but are not in themselves the defining criteria for C-PTSD.

Not every difficult childhood is C-PTSD and not every conflict is trauma. If the concept comes to include every discomfort, we lose the precision that those most exposed need.

Well established: the diagnoses are clinical, not biologically verified

There is no clinically validated blood test, genetic test, hormone test or brain finding that can establish PTSD or C-PTSD in an individual. The diagnosis is based on exposure, symptom pattern, course and functioning. The ITQ can structure the assessment but does not replace a clinical interview.

This is not confined to post-traumatic conditions. For most common primary psychiatric syndromes, including depression, anxiety disorders and ADHD, there are no validated biomarkers that can establish the diagnosis in an individual. Tests and neurological investigations can be crucial for differential diagnosis and certain medically defined conditions. But that is different from laboratory confirmation of our most common symptom-based diagnoses.

After fifty years of major investment, biomarker research has found a great many candidates but almost none that are sufficiently reliable and clinically useful. The APA’s work on the future of DSM noted in 2026 that the diagnoses remain symptom-based. With Alzheimer’s disease as a particular exception, biomarkers lack the accuracy required for routine psychiatric diagnosis.

That does not mean biology is unimportant. In 2024, a large genetic study found many risk loci associated with PTSD. But an association at group level is not a diagnostic test, and a test result does not automatically establish the cause in an individual. Biological does not mean innate or unchangeable. Psychological does not mean imagined.

My thesis: biology has been given an interpretive authority it has not earned

Biological psychiatry has not been without results. It has yielded important knowledge about genes, the nervous system and medication. But it has not yet achieved one of its most ambitious aims: to validate our most common psychiatric diagnoses as distinct biological disease entities with diagnostic markers, clear mechanisms and mechanism-guided treatment choices for the individual.

This is not proof that biology does not matter. It is a reason to reject its exclusive claim to explanation. My criticism concerns the reductionist idea that the most real cause lies inside the individual and can preferably be measured in a test. When the threat has been a parent, partner, perpetrator or institution, the whole problem cannot be understood by examining only the person exposed to it.

Here is a complacent consensus to challenge. A diagnosis can be useful without being a proven cause. Yet a descriptive category can slowly turn into an explanation of the very symptoms that gave the category its name. Consensus can be practically necessary and scientifically comfortable at the same time. Real suffering can then be described, medicated and administered without being understood.

This is my thesis, not a motive that has been measured among researchers: what can be coded, delineated, prescribed and followed up may gain institutional advantages over a life history. The psychiatrist Kenneth Kendler has instead argued for empirical pluralism, in which biological, psychological and social causes influence one another.

I am not claiming that trauma is the answer to everything. I am claiming that today’s answers are incomplete and must be treated as testable hypotheses, not endpoints. If biological profiles can distinguish relevant conditions, predict their course and select treatment better than a clinical and contextual assessment, I will reconsider my criticism. If the trauma history and course do not support a trauma hypothesis, it should be rejected. Where the evidence ends, the question begins, not certainty.

DSM made a choice, and choices have consequences

The DSM-5 group considered DESNOS/complex PTSD but did not judge the evidence sufficient for a separate diagnosis. Objections concerned, among other things, overlap, diagnostic independence and clinical usefulness. DSM-5 also broadened the PTSD criteria. ICD-11 later chose another map: a narrower PTSD core and C-PTSD as a separate sibling condition.

Two authoritative systems, then, draw different boundaries in the same landscape of symptoms. This shows that the boundary is a scientific and institutional choice, not a blood test result. No one needs to have intended to conceal C-PTSD for a classification decision to make the condition less visible. When the name is absent, it may affect which questions are asked, which groups are studied and which treatment needs become visible. We do not know how large this effect is. There is no evidence that the DSM decision was intended to give ADHD precedence or protect a medical model.

Unknown: how many people are missed in Sweden

Sweden still uses ICD-10-SE in its health data registers. The Swedish National Board of Health and Welfare aims to introduce ICD-11 from 2028. C-PTSD can be assessed but still lacks its own Swedish code and therefore separate national statistics. The registers have also lacked diagnoses from primary care. National data collection from publicly funded primary care is proposed to begin in 2028 at the earliest.

This shows a concrete gap in follow-up, not a proven number of unrecognised cases. There is no representative Swedish review of clinical records that can show how many people are missed. The strongest criticism is therefore this: the system cannot yet answer the question.

My thesis: ADHD can become an answer before the question is finished

ADHD is also a clinical diagnosis in which different combinations of symptoms can be enough for the same label. Problems with attention, restlessness, sleep difficulties and emotional instability are not specific to ADHD either. They can occur in post-traumatic and other conditions, or when several conditions are present together.

A full assessment should consider onset, developmental course, functioning in several settings and alternative explanations. The national ADHD programme states that questions about violence and trauma should be asked. We do not know how well this is followed. In a small Swedish study in child and adolescent mental health services, BUP, extensive trauma exposure was common among children referred for ADHD or autism assessment. The study did not show that their diagnoses were wrong. It showed why a trauma history is not a side issue.

Available Swedish and international studies do not provide a robust estimate of the proportion of ADHD diagnoses that, on closer assessment, would be better explained by PTSD or C-PTSD. I therefore cannot show that a considerable proportion are misdiagnosed. My hypothesis is that, for some people, the ADHD label captures surface symptoms while threat, avoidance, shame, self-concept and relationships remain unexplored. If that is so, trauma assessment and effective treatment may be delayed. We do not know how often this happens or what it costs.

Stimulants can reduce the core symptoms of ADHD in the short term, but they are not trauma treatment. Relief of a symptom does not prove its origin. A study of just over 1.2 million new users shows that longer treatment interruptions are common, particularly among young adults, but also that restarting is common. The registers do not show why the interruptions occur. It is therefore not established that most people stop because of side effects or because the effect wears off. A review from 2026 found little or no evidence that therapeutic tolerance is common, but most included studies were of low quality. Reliable long-term randomised evidence is, at the same time, considerably thinner than the short-term evidence (Smith and colleagues, 2026; Gosling and colleagues, 2025).

The same evidential discipline is needed with regard to emotionally unstable personality disorder, EIPS. C-PTSD and EIPS can overlap and coexist, but the proportion misdiagnosed in Sweden is not known. The question is not which diagnosis we should defend. The question is what the diagnosis describes, what may explain the pattern and what the treatment still leaves untreated.

Contested: which treatment, and in what order?

The treatment evidence specifically for fully diagnosed ICD-11 C-PTSD is still limited. The Swedish health technology assessment agency SBU found three relevant randomised studies following sexual violence; all had a high risk of bias. A Norwegian randomised study of 92 people did not find phase-based STAIR Narrative Therapy superior to prolonged exposure or skills training.

That does not mean C-PTSD cannot be treated. It means we should not replace one dogma with another. Trauma-focused CBT and EMDR have support for PTSD, and a meta-analysis suggests improvements also in some symptoms included in C-PTSD, although participants were not required to have fully diagnosed ICD-11 C-PTSD (Karatzias and colleagues, 2019). Stabilisation and skills training may be needed for some. A direct focus on trauma may be possible for others. Treatment needs to start from safety, comorbidity, functioning, the patient’s goals and what actually maintains the problems.

Five places where the next question must be asked

In primary care: when depression, anxiety, sleep problems, pain or repeated sick leave do not quite add up, it needs to be possible to ask: Has something happened that still makes itself felt? A questionnaire can support assessment but is not a diagnosis. Someone has to own the path onward to clinical assessment, safety, treatment and follow-up.

In psychotherapy: take an actual trauma history when symptoms and history warrant it. Assess both the PTSD core and the DSO domains. But do not make trauma the default answer to every problem, and do not let a new label replace an individual formulation.

In child and adolescent mental health services and schools: see changed behaviour as information, not proof of a particular diagnosis. Ask about violence, neglect and fear in a safe, age-appropriate way, ensure protection and refer onward. The teacher should not become a diagnostician, but the school must not turn a blind eye to the fact that concentration can be disrupted by ongoing threat.

In occupational health: do not ask only why the employee is performing less well. Ask what has changed, whether the workplace contains threats, degrading treatment or reminders of earlier threats, and what support is needed. Swedish research shows an association between PTSD and sick leave, but not what misdiagnosis costs. Do not sell a cost figure the research does not provide.

In geriatric care: let new sleep problems, fear, avoidance or strong reactions prompt a gentle question about life history. C-PTSD also occurs in older people. But never replace a medical or cognitive investigation with a trauma explanation; the research does not show that C-PTSD is routinely mistaken for dementia.

I do not offer trauma as an explanation for everything. I offer it as a hypothesis healthcare sometimes does not test. If it does not fit, it should be rejected. But it must first be allowed to be posed. Trauma is common. PTSD and C-PTSD are not. That is precisely why assessment is needed instead of assumptions. Our patients cannot afford to have incomplete models treated as finished truths.