First things first, I’m not a “Critical Psychiatrist” but I am a critical psychiatrist in the sense that as a psychiatrist I don’t just blindly accept assertions, I try and evaluate the evidence behind statements. For this Power Threat Meaning Framework (PTMF) I am focussing on the evidence for this suggested alternative for diagnosis (the authors seem to be inconsistent saying it is just a suggested conceptual alternative take it or leave it on social media but in the PTMF itself it suggests using diagnosis is unethical). I will look at the evidence for reliability and utility (usefulness) for clinical, research, administrative and social functions for this classification of mental health problems.
When I read the PTMF I only scanned the bits about diagnosis and biology, from the fragments I saw it seemed a highly selective and inaccurate interpretation of the evidence so to save my blood pressure I will move on from them.
Reliability of mental health classification is usually measured as how often clinicians/ researchers agree on the same way of classifying the same person’s problems. The PTMF repeatedly attacks the reliability of diagnosis whilst seeming ignorant that diagnosis is more reliable than psychological formulation-based classifications (but psychodynamic formulation has similar reliability to diagnosis). Disappointingly there is little information in PTMF about its’ reliability and no news about any research on its’ reliability to be published later. The authors have had 5 years which was ample time to have done some research on reliability e.g. providing vignettes (case descriptions) online and getting participants to formulate the cases using the PTMF and then comparing the similarity of responses. Without reliability data the usefulness of PTMF for lots of functions is badly impaired – how can we know that the problems classified in a particular way in the research would be the same way we would classify them in clinical practice.
The next issue is applicability to the problems people seek help with from services. The authors accept that some problems may not be suitable for the PTMF such as direct effects of brain diseases, short term effects of drugs or core features of autistic spectrum conditions. They may be used for other problems that may arise secondary to these problems, for example low mood caused by the effects of brain diseases on people. Unfortunately, the PTMF does not seem always applicable to all the other problems people present with. For example, it notes the links between trauma and psychosis (but minimises the importance of other factors that operate in conjunction with childhood trauma to produce psychosis) but does not mention that childhood trauma does not cause psychosis in 65% of people who present with psychotic symptoms or that not all clinical features of psychosis are associated with childhood trauma.
PTMF does seem useful for some problems that people have especially if power imbalances or trauma cause the problem. Unfortunately, because of the overselling of PTMF there is no information given as to what proportion of problems presenting to services are best dealt with using the PTMF. Apart from those problems listed as being excluded the authors gave me the impression that they felt the PTMF was the best solution for all problems. (People often have more than one problem and the same person may have some problems helped best with using PTMF and other problems dealt with better using other methods e.g. CBT or diagnosis/ medication). They are also so deeply wedded to their ideology that they recommend stopping specialised mental health teams like EIT. EIT is associated with better outcomes for patients like improved experience of care, improved access to therapy, improved social outcomes and reduced suicide. Better models of care are to be sacrificed on the altar of the authors’ ideology.
There is a lot of references and discussion of social factors and mental health. These clearly showed an association between social factors and mental health but they did not show a sufficiently strong relationship to prove the implied hypothesis that viewing problems through the PTMF is a sufficient and complete explanation of all the problems people present with i.e. other psychobiological factors are involved. Some are reasonably proven – such as the social nature of reason for high rates of people in UK Black communities meeting schizophrenia criteria or for childhood trauma and people meeting personality disorder criteria. There’s also an unfortunate suggestion that ADHD should be viewed primarily through the PTMF which is at odds with research evidence and what patients themselves report about their own understanding and experiences.
The feasibility of using the PTMF in short appointments (of 15-30 minutes) or in emergency work including overnight when time / cognitive resources are under pressure and many issues such as risk or excluding medical causes of problems is limited. The constructs used have little predictive information for outcomes or choosing treatments such as medication (though research in the future may change this). For medication the PTMF uses the misleading “disease centred” vs “drug centred” model whereas most psychiatrists in my experience do not think the medication is reversing a disease process but has research evidence of reducing symptoms or other improvement in outcomes. There is a strange order of questions in the PTMF (asking the person’s story last after 4 questions trying to frame the person’s problems in terms of the PTMF when surely asking the person’s story first would make more sense).
The PTMF avoids drawing thresholds as part of its philosophy of trying to avoid putting people in boxes and separating problems as different from everyday experiences and responses. Unfortunately, this is a problem as one of functions of classification is to establish the problem as sufficiently severe to justify something- such as access to a service, or exemption from sanction or eligibility for benefits. Now this is often combined with other dimensional measures such as for example levels of impairment but being classified as for example having a psychiatric diagnosis makes it more likely to other people that you will meet criteria for these advantages. A diagnosis of agoraphobia makes it more likely to not be able to leave the house unaccompanied and subsequent social impairment.
The suggestions for using PTMF for administration, statistics, legal purposes or accessing benefits are ill-thought out and often do not use PTMF itself to fulfil the roles. Third-party payment health systems such as the NHS (or insurance-based models) require administrative data from healthcare providers to provide a justification for funding them. PTMF suggests using broad activity and severity data (not part of PTMF) but administrative data also requires information on nature of presenting problem that PTMF doesn’t provide. Diagnostic codes may not always provide a lot of differential information (but if somebody’s only diagnosis was mixed anxiety and depression it seems unlikely third-parties would pay for 2 years of expensive inpatient secure rehabilitation for instance). Diagnostic codes are already used by the rest of the NHS for the rest of health and diagnostic codes would still be needed in mental health for conditions PTMF admits it’s not suitable for e.g. Alzheimer’s disease or drug intoxication. We could try and have two different systems in mental health but this seems an unnecessary transaction cost and added complication for most mental health services. Similarly, for statistics why have separate systems for collection of data -and PTMF is unable to provide such data anyway relying on alternatives.
As for legal purposes, courts are likely to prefer diagnostic constructs with in-built thresholds with evidence of association with outcomes they are interested in as well as some dimensional assessments. Schizophrenia is more likely to be associated with fitness to plead problems than mild depression. Psychopathy / antisocial personality disorder is more likely to be associated with re-offending and being hard to predict offending than panic disorder. Of course, other assessments help but the legal system is geared towards making categorical judgements which fit well with diagnostic constructs.
The most challenging section for PMF is access to welfare benefits. People with mental health conditions often suffer long-lasting or temporary but recurrent impairments that make it hard for them to work. Benefits/ welfare agencies are under pressure not to pay out benefits. For claimants to overcome this resistance they need to prove their functional impairment. Now of course as PTMF says there needs to be statements of degrees of impairment even with a diagnosis. But a diagnosis has research evidence of association with degrees of impairment – see the example of agoraphobia mentioned earlier – thus helping to validate professional’s statements of degrees of impairment. Diagnosis also fits in with how the benefits agency assesses impairments in the rest of health. A diagnosis from a professional plus statement of degrees of impairment help the benefits applicant prove to the sceptical benefits officer of entitlement to claim. The PTMF provides no adequate replacement and risks people being denied benefits they are entitled to and suffering severe financial hardship.
There are some frankly awful suggestions in PTMF about disagreeing with diagnosis on benefits applications forms or saying people have recovered when applying for benefits. The suggestions on changing the benefits system ignores applicants need to deal with the system as it is now. As for Universal Benefit Income, this is unlikely to be provided at a level that is necessary for people with health conditions who often need more money than the average person. For example, people may need to pay for taxis rather than walk or take the bus because they are too anxious to do so or need to pay for cleaning as they lack the energy or motivation to do so because of their health condition.
In short, whilst PTMF offers a useful technique for helping some problems that present to mental health services it is not useful as a classification for administration, statistics, research (but can be researched as a technique), legal purposes or access to benefits. It is not a practical system for short appointments or emergency work. It is not a replacement for diagnosis for many of its’ functions.
This is covered by a Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 International (CC BY-NC-ND 4.0) License 
Recovery in the Bin 