Very few people with an interest or involvement in mental health can have failed to notice the recent media interest in the publication of the latest update to the American ‘psychiatric Bible,’ the DSM (Diagnostic and Statistical Manual), now in its 5th edition. It has taken 14 years to write, weighs in at over 800 pages (“thick enough to stop a bullet,” as one psychiatrist put it) and costs in the region of £80.00.

But it has been mired in controversy, with suggestions that the pharmaceutical companies have been too influential in the revision process. For example, 18 out of 27 members of the taskforce who revised the manual had direct links to pharmaceutical companies. Furthermore, there have been accusations that the criteria needed to diagnose some mental health conditions have become too wide, or the thresholds too low, with the effect of ‘medicalising’ normal reactions. For example, the diagnosis of major depressive disorder (MDD) applies to people who experience persistent low mood, loss of enjoyment and pleasure, and a disruption to everyday activity – which is a reasonable conclusion. But under previous criteria, people who had recently been bereaved were specifically excluded, as all of these symptoms are a normal part of the grief process and not signs of mental illness. But in the DSM-5 this exclusion has been removed, so for example someone who experiences these normal responses after the death of a spouse could now very easily be diagnosed as mentally ill. As a result, a wide variety of individuals and organisations have accused the updates to the DSM-5 of ‘over-medicalisation,’ with the possibility that some people will be prescribed drugs when all they really need is to progress through the normal stages of grief and adjust to life after loss.

There has been less controversy in terms of changes to the criteria for dissociative disorders but there are some interesting and significant changes. To understand these revisions, it is important to consider how dissociative disorders (DDs) are diagnosed in the UK and therefore what impact the updates to the DSM-5 will have. Of course, on the whole it would probably be true to say that DDs are often not diagnosed at all! Much has been written in the clinical literature about how often DDs are misdiagnosed – as bipolar disorder, schizophrenia, depression, borderline personality disorder to name but a few alternatives. The ISSTD guidelines (2011) explain why: “The difficulties in diagnosing dissociative identity disorder result primarily from lack of education among clinicians about dissociation, dissociative disorders, and the effects of psychological trauma, as well as from clinician bias” (p.117) and “Most clinicians have been taught (or assume) that dissociative identity disorder is a rare disorder with a florid, dramatic presentation” (p.117).

But is this a problem? Many people suffering dissociative distress feel that if they could just get a diagnosis, then everything would be alright. And there are certainly a range of benefits from having a diagnostic label – it can help people make sense in a positive way of what is going on for them, of dissociation as a clever, protective survival mechanism that helped them endure otherwise unbearable trauma. Knowing that they are not ‘mad’ or ‘crazy,’ but that their symptoms make sense and are utterly logical in the light of their experiences can be a massive step on the pathway towards recovery. Knowing that the ‘problem’ is a DD can also help determine the appropriate pathway for treatment and recovery: long-term individual psychotherapy is the treatment of choice recommended in the ISSTD Treatment Guidelines, as opposed to harmful and unhelpful treatment with anti-psychotics or a focus merely on symptoms such as substance abuse or eating disorders.

So getting a diagnosis can be very helpful indeed, although many are disappointed to realise that within the strictures of numerous NHS Primary Care Trusts, it does not in fact lead to appropriate treatment and can for some people actually mean being side-lined as ‘too complex.’ There can be other downsides too such as difficulty in obtaining life insurance, or discrimination at work (illegal though that may be.)

Diagnosis is certainly not the panacea that many people hope it will be, and in many cases it can actually be unnecessary as a wide range of therapists are less concerned with the ‘labels’ that their clients come to the therapy room with than they are with the person sat in front of them, whatever form their distress is taking and whatever an ‘official’ diagnosis or lack of it may say.


In the UK, there are two principal routes towards a DD diagnosis: privately or via the NHS. The private option includes going to the Pottergate Centre in Norwich, run by Remy Aquarone, as we have discussed in other articles. Achieving a diagnosis via the NHS is more of a postcode lottery and will generally be carried out by either a clinical psychologist or a psychiatrist. But whether privately or through the NHS, diagnosis involves a qualified professional assessing a client against certain defined criteria, and that is where the DSM, and its European-based counterpart the ICD (International Classification of Diseases) come in. Both of these ‘manuals’ provide lists of mental health conditions with associated codes for quick reference, and a list of criteria that have to be met for that condition to be diagnosed. Alongside these essential reference guides, a range of other tools have been developed, from screening instruments such as the DES (Dissociative Experiences Scale) and SDQ-20 (Somatoform Dissociation Questionnaire) right up to the ‘gold standard’ of Marlene Steinberg’s SCID-D (Structured Clinical Interview for DSM-IV Dissociative Disorders).

So the basis of diagnosis of all mental health conditions are these two diagnostic manuals and although there is, as would be expected, a certain amount of overlap between the two, they do have their differences. Firstly, the DSM is published by the American Psychological Association and is predominantly used in the United States. The ICD meanwhile is published by the World Health Organisation and is most often used in the UK and other parts of Europe. This is relevant for two reasons: firstly, because most research that has been conducted into DDs, as well as the screening and other tools such as the DES and SCID-D, are based on the DSM system; and secondly because the two manuals view DDs quite differently. The DSM understands DDs as being chronic, long-term conditions that are developmental in nature – in other words, they develop over a period time in response to trauma and attachment difficulties as a child’s brain grows and matures, and they therefore tend to persist over time too as they are more or less ‘hard-wired’ adaptations. The ICD, however, sees DDs as being on the whole acute (ie short-term), reactive, and transient responses to here-and-now stressors and traumatic experiences. Both concepts of course are part of the truth, an issue acknowledged by the DSM taskforce. But most of what has been written about DDs, especially dissociative identity disorder, is based around the DSM understanding, so even though the DSM is not generally used for diagnosis purposes in the UK, an update to its criteria for DDs does have an impact on many people working in this field and its criteria will inevitably be most often quoted in training and in clinical literature.

So what has changed in this latest update? You can see the diagnostic criteria for both the DSM-IV and DSM-5 in the tables on the following pages, but in brief the DSM-IV criteria stated that dissociative identity disorder involved the “presence of two or more distinct identities or personality states (each with its own relatively enduring pattern of perceiving, relating to, and thinking about the environment and self.)” It further stated that “At least two of these identities or personality states recurrently take control of the person’s behaviour,” and that someone with dissociative identity disorder would also have an “Inability to recall important personal information that is too extensive to be explained by ordinary forgetfulness” (APA, 1994). The criteria further clarified that this was not due to being drunk or the effect of drugs, and that in children it was not to be confused with fantasy play. So the key elements were two or more parts of the personality along with amnesia for important personal information.


Many researchers, clinicians and commentators have been suggesting for a number of years now that these criteria are inadequate, and the update to DSM-5 was the ideal opportunity for their concerns to be addressed. A working group headed up by David Spiegel and including ‘big names’ in the field such as Richard Loewenstein and Paul Dell wrote a paper, Dissociative Disorders in the DSM-5 (see, presenting their recommendations for the new diagnostic criteria. They identified a number of problems, including the following:

Dissociative identity disorder is complex and has a range of symptoms or phenomena: the current criteria do not reflect this very well.

The current criteria emphasise the “presence of two or more personalities” taking control – in other words switching – and yet switching has been found to happen very infrequently during diagnostic interviews, occurring in less than 15% of people with dissociative identity disorder.

Because people don’t ‘switch on demand’ in diagnostic interviews, too many people end up with the diagnosis of Dissociative Disorder Not Otherwise Specified subtype 1A rather than the more accurate category of dissociative identity disorder. As a result, DDNOS, which should be a “residual category” catching a small number of atypical cases, actually represents 40% of all diagnosed DDs.

The criteria do not mention “experiences of pathological possession,” which the researchers consider to be a very common form of dissociative identity disorder in other cultures.

These concerns have mostly been responded to in the new criteria, which for dissociative identity disorder at least are mostly minor tweaks of the wording rather than a drastic overhaul. Probably the most significant change revolves around the issue of switching, of parts/alters “taking control.” Although an emphasis on this has been retained for diagnostic purposes, the criteria now allow switching between parts either to be observed by the clinician or to be self-reported by the client. It is expected that this will mean that a large swathe of people previously diagnosed with DDNOS Type 1A will now be reclassified as having DID, which will end what many people have seen as a somewhat arbitrary distinction between the two conditions.

The requirement for amnesia in dissociative identity disorder has also been modified slightly. The criteria talk now of “Recurrent gaps in the recall of everyday events, important personal information, and/or traumatic events.” This is quite a dramatic widening of the interpretation of amnesia to account for how varied it can be. Rather than just being unable to recall “important personal information” – which is vague and yet seems somehow quite tightly defined – the new criteria specifically refer to episodes of so-called “time loss” in daily life (usually caused by switching to another part) as well as amnesia for traumatic events. It is my personal feeling that this will make it easier for people to get a full diagnosis of DID rather than DDNOS Type 1B, which accounted for people who did not apparently have an “inability to recall important personal information.” In reality this seemed to correlate to people who had full co-consciousness between their alters/parts, which in my experience at least seems to be far more common than is usually admitted. Therefore people who do not experience ongoing time loss as they switch between parts can still qualify for DID, as I understand it, because of their lack of memory for the traumatic events in early childhood which led to their dissociative disorder in the first place. We will have to wait and see how tools such as the SCID-D are updated and how they interpret this criterion, but it seems to me that it will ensure that far more people will be identified as having dissociative identity disorder rather than being left in limbo with DDNOS-1 over what is to a large extent a technicality.

As recommended by Spiegel et al, the new dissociative identity disorder criteria also includes a reference to “an experience of possession.” It will be interesting to see how this is interpreted, especially in more evangelical Christian circles, and it is worth reading the original text of Spiegel et al’s article to understand their reasons for pushing for this characteristic to be included in the criteria. There are numerous first-hand accounts of dissociative survivors undergoing unhelpful and at times retraumatising ‘deliverance’ sessions within some evangelical circles, where the presence of parts/alters has been mistaken as demonic possession/oppression. The new criteria will hopefully help to engage Christian ministers in better understanding the effects of trauma.

The final thing to note is that there is a new requirement within the dissociative identity disorder criteria for the condition to cause “clinically significant distress or impairment in social, occupational, or other important areas of functioning” (Criterion C). This brings it into line with the other DDs where for example in DSM-IV Depersonalisation Disorder (300.6), Dissociative Amnesia (300.12) and Dissociative Fugue (300.13) all carried such a requirement. I have not read any rationale for this but if it is applied rigorously it will exclude from positive diagnosis people who claim that DID is a natural state (rather than a developmental consequence of trauma) and who claim that they “enjoy” having the condition: for it to be diagnosed, it must cause distress.

Inherent, therefore, in the new criteria for dissociative identity disorder are changes that will mean that what has been known up to this point as DDNOS will be far less frequently diagnosed, and it should therefore become what it was intended as – a residual category for a few atypical cases. Whilst this is a positive move, the APA are bound to cause some degree of confusion by changing the name of DDNOS and splitting it into two sister categories: Other Specified Dissociative Disorder (OSDD) and Unspecified Dissociative Disorder (UDD). But if the majority of DDNOS-1 cases are reclassified as dissociative identity disorder, it may be an astute move as the old category of ‘DDNOS’ is more or less disbanded and the new OSDD/UDD category applied as intended to only a very few atypical cases.


So what is this new OSDD/UDD diagnosis all about? Quite simply, it is designed to be applied to people who do not meet the full criteria for one of the other DDs. In both cases the DSM-5 says: “This category applies to presentations in which symptoms characteristic of aDisordercause clinically significant distress or impairment in social, occupational, or other important areas of functioning predominate, but do not meet the full criteria for any of the disorders in theDisorder.” The distinction between the two categories then comes down to whether the clinician wants to expound on why the person has not met the criteria for a specific DD. If they do, for example “Dissociative Trance,” then the diagnosis is “Other Specified Dissociative Disorder,” with the reason added. But when there is not enough information to make a more specific diagnosis, for example in A&E, then the (presumably temporary) diagnosis will be given as “Unspecified Dissociative Disorder.”

Finally, there is one other significant change to the criteria for DDs although in many ways it is a technicality and of much less relevance to people with dissociative identity disorder. This is the fact that dissociative fugue has been dropped as a separate disorder and has been conflated in with dissociative amnesia (DA). DA can now be diagnosed either on its own or “with Dissociative Fugue.” It was felt that there was not sufficient “clinical utility” to keep the two categories separate as it affected so few people.

It could be argued that, as the ICD is favoured over the DSM in the UK, these updates in DSM-5 will not really affect many of us on a practical level. But I do think that there will be a few direct impacts. Firstly, we await the necessary changes that will have to be made to the diagnostic tool, the SCID-D and see how that will affect diagnosis. These updates are not expected until 2014. Secondly, it will be interesting to see how epidemiological research is impacted and whether new prevalence studies show dissociative identity disorder as even higher than the 1-3% of the general population that the ISSTD accepts. With the migration of many people currently assessed as having DDNOS-1 into the more inclusive set of criteria for DID, it could be that rates do jump significantly – we will have to wait and see. Thirdly, the transition of language from DDNOS to OSDD and UDD is bound to cause some confusion for probably a number of years. Hopefully this will be mitigated at least in part by people who were previously labelled as DDNOS now having the more realistic diagnosis of dissociative identity disorder.

In terms of the criteria for dissociative identity disorder, I am sure that there are many people who will remain dissatisfied that they seem to focus so centrally on the phenomenon of switching between parts, even though this can now be self-reported, when there are such a range of experiences significant in living life with DID. Spiegel et al in their recommendations refer to a core of eight symptoms common to DID: amnesia, self-alteration, First Rank Symptoms (a range of experiences first developed in the diagnosis of schizophrenia but actually found in many cases to be more prevalent in dissociative identity disorder than psychotic disorders), voices, trance, somatoform symptoms, depersonalisation and derealisation. The criteria mention only two of these, but Spiegel et al’s article does explain the reasons for this: that identity alteration and amnesia are the only symptoms necessary to diagnose DID. The argument centred around whether the criteria for DID should remain ‘monothetic’ or become ‘polythetic’. Simply put, monothetic criteria require for all criteria to be met. A polythetic approach, on the other hand, requires for some criteria to be met, for example “three out of the following eight symptoms,” such as is found in Post Traumatic Stress Disorder (PTSD) and Borderline Personality Disorder (BPD). The argument is an interesting one but research seems to be clear that a monothetic approach is better – it is easier for clinicians to use and doesn’t introduce grey areas, and in the case of dissociative identity disorder it was felt that it was unnecessary to introduce other symptoms when at core everyone with DID has the two ‘prime’ symptoms of identity alteration and amnesia.

Implicit within this approach, however, must be the understanding that the criteria are there to diagnose but they are not there to describe. They help to distinguish one condition from another, but they do not describe what it is like to live with that condition. Keeping that distinction clearly in mind helps us to understand why switching between parts is so central to the criteria when it is not always evident or visible in a dissociative identity disorder person in daily life. Certainly the new criteria keep things simple, and that can only be a good thing.

For details of the diagnostic criteria in the DSM-5 for dissociative disorders, please click here.