For several years, the discussions about the deliberations of the various task forces involved in constructing the latest revision of the Diagnostic and Statistical Manual of the American Psychiatric Association, the DSM-V, have been shrouded in secrecy. The lack of transparency of the discussions generated highly publicized criticisms from such luminaries as Robert Spitzer, the major developer of the revolutionary DSM-III, and Allen Frances, chair of the DSM-IV task force. This situation radically changed in February with the release of the revisions of the changes proposed by the 13 work groups charged with revising the DSM.
Most of the public response to the proposed changes has centered on several alterations to particular diagnostic categories. The suggestions include using a new diagnosis of “temper dysregulation with dysphoria” for children instead of bipolar disorder, incorporating Asperger’s disorder, autistic disorder, and several other conditions into a single “autism spectrum disorders” category, and recognizing some new types of eating disorders. The focus on these specific changes, most of which are sensible, has deflected attention away from other suggestions that have much greater potential import.
Three changes, in particular, could lead to an enormous pathologization of non-disordered conditions. The first is the suggested revision of the criteria for Major Depressive Episode to remove the bereavement exclusion from this diagnosis. At present, the criteria for major depression require five or more out of nine symptoms including sadness or lack of interest or pleasure that least for at least two weeks. However, the criteria exclude people who experience these symptoms in response to bereavement: “The symptoms are not better accounted for by Bereavement” That is, people who develop enough symptoms to meet the criteria after the death of an intimate are nevertheless not defined as disordered but instead as suffering from a natural, nondisordered response to loss.
The reason for the bereavement exclusion is obvious. Voluminous evidence indicates that bereavement after the loss of an intimate is a natural reaction. The earliest literary portrayals of human experience such as Gilgamesh and The Iliad indicate that grief is a basic human emotion. Likewise, while different cultures have vastly different expressions of grief, sadness and accompanying psychological and somatic symptoms after the loss of a loved one is a universal experience. Even many primates show demonstrable signs of depression-like symptoms after the death of a close relation. In the vast majority of cases, the universal symptoms of grief dissipate with the passage of time and only a minority of the bereaved remains highly symptomatic after several months pass.
If the suggested revision is implemented anyone who is sad, fails to derive pleasure from usual activities, finds it difficult to concentrate, and has sleep and appetite difficulties for a mere two weeks could be diagnosed with Major Depressive Disorder. Because virtually the entire population will be bereaved at some point in their lives and because such a high proportion of the bereaved would meet diagnostic criteria that require two week duration, this proposal could pathologize an enormous number of people.
A second proposal that has the promise of massively medicalizing natural emotions is to adapt dimensional assessments for the existing categorical diagnoses. On the surface, this proposal sounds sensible and desirable. Major Depression, for example, requires the presence of five symptoms but there is no natural cut-off point between four and five symptoms, or at any other particular point for this diagnosis. Depression, as well as the other major conditions in the DSM, seems to naturally be a continuous rather than a categorical condition.
The problem in dimensionalizing common conditions such as depression and anxiety is that a small number of “subthreshold” symptoms typically indicate a non-disordered condition, not a milder form of disorder. The only way to accurately use a dimensional system is to initially use criteria for disorder that separates natural from disordered conditions, regardless of how many symptoms are present. If adequate conceptions of disorder first distinguish contextually appropriate symptoms that are commonly transitory responses to stressors from mental disorders, then dimensional measurement could represent a distinct improvement in the DSM. As the discussion of bereavement indicates, however, the separation of disorders from non-disorders in the DSM-V seems to be getting worse rather than better. The current proposal to dimensionalize measures of frequently occurring disorders threatens to pathologize even mildly distressing conditions. While potentially valuable, it needs reconsideration and reformulation.
A final worrisome proposal lies in the creation of “at-risk” categories for mental disorder. At present, this possible category is limited to psychotic conditions; people who have just one symptom from among delusions, hallucinations, and disorganized speech who have never met the criteria for a psychotic disorder could receive the “at-risk” diagnosis. The diagnosis is well-intentioned and aimed at identifying people who might be at an early stage of a psychotic condition but who don’t yet meet the full criteria. Such people might benefit from early identification and treatment.
The problem with the “at-risk” category is it’s potential as a Trojan horse that would diagnose nearly a-symptomatic people as being in the early stages of a disorder. Yet, at present, there is no way of knowing which people with a single or a small number of symptoms will go on to meet the full diagnostic criteria and which will not. The latter group will typically outnumber the former group so that the potential for false positive diagnoses is enormous.
The current suggested revision only applies to psychotic conditions where it might not create too much damage. If it were applied to widely occurring conditions such as depression and anxiety, however, the result could be a massive amount of new pathology. For example, one of the best known studies of depression, the Dunedin Study, ties the presence of the small allele of the 5-HT gene to this condition. Yet, nearly 20% of people have two copies of the small allele and over half have one copy, so over two-thirds of the population could be viewed as “at-risk” for developing depression . Once a gene is identified as a risk factor for depression, anyone who has the gene may be a candidate for intervention, even if they don’t actually have a depressive condition. Genetic tests could identify “at risk” individuals, who could then be placed on long-term regimes of drug therapies. In the case of the 5-HT gene, a majority of people would be at risk for depression. While the DSM-V working groups have not (yet) proposed an at–risk category for depression or any other commonly occurring condition, this danger might be lurking in the future.
Overall, it appears that the original promise of the DSM-III in 1980 – the creation of a clear, precise, and reliable diagnostic system that would eventually lead to more accurate knowledge about the causes, prognoses, and treatments of mental disorders has not been fulfilled. Indeed, it is difficult to think of a single breakthrough that has resulted from psychiatry’s classificatory system. The major proposals in the DSM-V do not seem as if they will change this situation and could wind up making psychiatry’s central problem of distinguishing pathology from normality even more difficult to resolve.
image©Vincent W. Hevern