Problems with Psychiatric Diagnosis

The Current Classification System

Two classification systems are used worldwide for the majority of psychiatric diagnosis: the DSM, and the International Standard Classification of Diseases and Related Health Problems (ICD; World Health Organization). The most recent editions are the ICD-10, and the DSM-IV-TR. Over time, the two systems have become more and more alike, primarily for the purposes of diagnostic compatibility. This essay will therefore focus solely on the DSM classification system purely for simplicity. The DSM uses a categorical approach to diagnosis; a patient either has a disorder, or they do not. Disorders are conceptualised as separate entities, so that each disorder is assumed to be distinct from other disorders in some sort way. According to this system of classification, patients are diagnosed on the basis of symptoms. This may be the presence or the potency of symptoms, which requires a subjective judgement by the clinician. It is acknowledged outright in the DSM-IV-TR that this classification system is not based on aetiology, but is based on the reliability of classifications. In other words, the DSM is purely a naming system that splits disorders in a pragmatic way for the purposes of reliable diagnosis and research. It is hoped that these categories may point to distinct aetiologies, but this is not what the DSM currently aims to do.

 

Problems with the Current Classification System

Categorical Classification

The primary problem with the categorical classification system is that a decision needs to be made in regard to the criteria by which someone should be classified as having the disorder. If the classification is based on a subjective judgement of symptoms, then it is doubtful that anything valid separates a patient who has the disorder from someone who is sub-threshold. In this case, it would not make sense to claim that one person ’has’ the disorder, whilst another does not have. ‘Having’ a disorder implies that there is a distinct aetiology. For example, someone with depression may have a certain biological deficiency. However, it is not on the basis of a biological deficiency that a diagnosis is made using the DSM. Therefore, it is nonsensical to claim that someone ‘has’ a disorder and that another person does not ‘have’ the disorder when symptoms are used to categorise.

Comorbidity is also a concern of clinicians. That is, patients can often be diagnosed within several diagnostic categories. Comorbidity is generally not so much a theoretical problem as it is a practical problem. It is quite likely that many disorders share common aetiologies, yet manifest in different ways. It is also likely that one disorder may cause another disorder. However, the way a patient is diagnosed can have massive ramifications in terms of the treatment they receive. For example, almost all sufferers of Body Dysmorphic Disorder (BDD) have another diagnosis such as MDD. However if the BDD is causing the MDD, it would be unadvisable to treat the MDD as this would not treat the cause of the problem. Yet this is the sort of situation that symptom-based categories may create.

 

Symptom-based Classification

Basing diagnostic categories on symptoms has one great advantage: clinicians only need to deal with what is known when communicating with their patient, rather than making theoretically-based inferences, which are inevitably only as useful as the theories that they are based on. It also reduces the extent to which subjective judgements need to be made, which increases inter-rater reliability. Basing diagnosis on symptoms also fits with lay conceptualisations, which provides better face-validity and increases the extent to which patients and the lay public will be able to understand and relate to a diagnosis.

The main disadvantage of symptom-based classification is that the categories may cross aetiological boundaries, leading to ineffective treatments in some cases. An analogy may be found in physical illness. Imagine if all patients who had a headache were diagnosed as having ‘headache disorder’, and it was found that in most cases, rest and a paracetamol were effective treatments. Obviously this treatment would not be effective for those patients suffering from a brain tumour. Similarly, it is quite possible that the symptom-based categories of the DSM have many different causes, and that better treatments may be found if the diagnostic categories were based on their respective aetiology.

 

The Pathophysiological Solution

A team of researchers organised by the APA published a document that discusses some of the issues to be considered in the new DSM-V edition (Kupfer, First, & Regier, 2002). The recognition that the current symptom-based categorical classification system is fundamentally flawed is evident in this publication. Charney et al. (2002) suggest that the current symptom clusters will probably not map onto distinct disease states; and that an aetiologically-based classification system is more likely to guide better treatment and more valid diagnoses. Given that all of the authors have a medical background, it is unsurprising that their solution is a classification system based on pathophysiology; reflecting the underlying medical/pharmacotherapeutic approach of both the APA and the document’s authors. This approach generally views mental disorders as being behavioural manifestations of an underlying biological condition, which is largely influenced by genetic propensities. The research presented clearly indicates that mental disorders are influenced by biological factors. According to Charney et al. (2002), future technological and methodological developments will lead to the discovery of genes and biological markers that specifically indicate certain disorders that cut across current diagnostic categories. Therefore, current diagnostic classifications may be stunting research by crossing aetiological boundaries; providing misleading research.

 

Fundamental Problems with Pathophysiologically-based Classification

Diagnosis Must be Based on Symptoms

The pathophysiological model of mental disorders would suggest that the difference between those who have a disorder and those who do not, is genetic or biological. However this does not account for the importance of context in the conceptualisation of disorders. The DSM-IV-TR itself emphasises the importance of social norms in defining what should be considered a ‘disorder’ (Interestingly, the DSM is clearly involved in setting social norms for psychological normality, indirectly making their definition circular). Apart from social norms however, what causes someone distress or significant impairment will differ from one individual to the next due to differences in circumstance and personality. That is, although someone may have a genetic and biological propensity, the way that the individual copes with their propensity, the support they receive from others and the cultural acceptance of their differences may create an environment where these propensities do not manifest into a significant problem.

As suggested by Cooper (2004), due to the process nature of mental disorders, it is likely that only some mental disorders are natural kinds, and that many are likely to be partial kinds. To extend on an example by Cooper, mental disorders may be much like a chemical reaction. Although the reaction requires several elements, it is the way in which these elements combine which produces the effect; and this effect will be influenced by the type, quantity, and reactive properties of the elements being mixed. Therefore, although a chemical reaction may be understood in terms of the elements it is comprised of, it must be categorised by either its effect, or the interactive process which leads to the effect; not the elements themselves. This would be especially important if one were to have no knowledge of the elements that were involved, and to what extent the elements had reacted. Similarly, as mental disorders are the manifestations of an interaction between a variety of biopsychosocial elements, then it must be defined by either the symptoms of this process, or the process itself; not the genetic, biological, psychological, or sociological elements it is comprised of.

There is also a practical problem with basing a diagnosis on genetics or biology: measurement. Currently it is not economically viable to provide genetic and neuroimaging tests to patients. Even if this were possible, due to individual differences in neural structures, neural functions, and biochemistry, it is unlikely that such measures would be able to provide any sort of distinct marker that could be used to diagnose a person into a distinct and meaningful diagnostic category. Furthermore, an understanding of the disorder would still require an understanding of how these pathophysiological factors interacted with psychosocial factors to produce significant distress or dysfunction.

 

Lack of evidence

For many of the disorders in the DSM including anxiety disorders, mood disorders, and schizophrenia, there has been a clear lack of evidence that they should be understood as biological disorders. Charney et al. (2002) indicate that a single gene or neurobiological marker that is useful in diagnosis or predicting treatment response, is yet to be found. The idea that disorders have a genetic element originally stemmed from familial or twin studies, which have shown that disorders run in families and that identical twins are more likely to share a disorder – even when raised separately. However there is a fundamental problem with the conclusion that this infers the disorder is genetic. Shared genetics would undoubtedly lead to a variety of similar behaviours, similar coping mechanisms, similar choices of occupation, and similar responses to stress. Undoubtedly, the relationship between genetic similarities and disorders are at least partially mediated by a number of behavioural and cognitive similarities, which affect the way that the person interacts with their environment. Furthermore, there may be a number of genes, with cumulative influences, which are involved in mental disorders – partially through their effect on many psycho-social factors.

This is not to say that all disorders don’t have their roots in genetics. Genetics provide the building-blocks of all biological and psychological phenomena, thereby being fundamental to the way in which people interact with their environment. The scientific evidence in genetic, neuroscience, and developmental research that psychological dysfunction is related to one’s genetic and biological make-up is overwhelming (See Charney et al., 2002; Pine et al., 2002). However, the extent to which any specific genetic element maps onto a distinct symptom cluster is both unlikely, and lacking in evidence. This is perhaps the motivation for devising a pathophysiologically based diagnostic system. The idea being that if the DSM categories were based on genetics, they would map perfectly onto specific genes or gene combinations. This logic is clearly circular, and makes an a priori assumption that mental disorders can be defined purely in genetic terms – totally disregarding the necessity for mental disorders to be defined in terms of symptoms or processes (as described earlier). Describing a mental disorder as being “genetically based” should therefore be cautioned. It is not the disorder that is genetically based, but the causal components of a disorder. The way these components interact is certainly complex and non-linear, and should perhaps be the primary focus of scientific investigation. That is, to link genetics with manifested symptoms, the dynamic way in which they are related needs to be explored.

Simplistic Assumptions

The idea that mental disorders can be considered to be biological or genetic entities may grossly underestimate the complexity of the biopsychosocial interaction (For a review, see Luyten, Blatt, Van Houdenhove & Corveleyn, 2006). Biologically, people may differ in the way they interact with the world, the way they perceive the world, and the way they cope with the world. Someone who has a disordered perception of the world yet copes with this perception in an effective manner man not have a disorder. Conversely, someone who has an ineffective coping style yet perceives the world in an adaptive way may not have a problem. However it may be the perpetual interaction between perception and coping strategy that leads to disordered behaviour. Each of these components may be influenced by the individual’s biology, yet the processes cannot be understood at a purely biological level of analysis, such as the pathophysiological solution.

In an attempt to address the biopsychosocial interaction, the diathesis-stress model has been used (Monroe & Simons, 1991). In this model, psychological disorders are viewed as the result of a genetic/biological propensity, plus stress. The problem with this account is that it fails to acknowledge the fact that biology affects the perception of what constitutes stress, and the way in which this perception is influenced by one’s environment, as well as the important psychosocial factors that don’t fit under the stress categorisation. Furthermore, it does not acknowledge the reciprocal relationship between diathesis and stress (Monroe & Simons, 1991). That is, external stressors influence one’s reactivity to stressors at a biological level. In order to improve this conceptualisation, biospychosocial dynamic interactionsm models have begun to address these issues (Luyten et al., 2006). However, these models are in their infancy, and may be greatly confused by the underlying assumptions of the various fields being combined including personality trait and schema theories, which rely on mind-body dualisms. For example, Luyten et al. (2006) suggests that the way genes interact with the environment is likely to be dependent on personality. However, personality is surely a result of genes, the environment, and the way in which they have interacted. Personality, genetics, and biology have historically been separated by psychological, genetic and biological theorists; and stitching them back together is proving difficult.

The Dimensional Solution

Inasmuch as categorical approaches have been used for their pragmatic advantage for clinicians, dimensional approaches have been advocated by some researchers due to their pragmatic solution to arbitrary boundary problems in psychiatric diagnosis. Dimensional approaches stay clear of aetiological problems by disregarding the entire idea of ‘disorder’. The dimensional approach views psychological disorders as being on the extreme end of a continuum, of which normal behaviour may also be positioned (Cuthbert, 2005; Widiger & Samuel, 2005). In this view, a psychiatric diagnosis would therefore not necessarily indicate that someone ‘has’ a disorder, but simply that certain dimensions of their personality are extreme, and are resulting in distress or dysfunction. The DSM-IV-TR acknowledges that the implementation of dimensional approaches may be an important modification to the manual, yet due to the difficulty of implementation, lack of clinical utility, and lack of an agreed-upon set of dimensions, implementation at this stage would be premature.

The strength of the dimensional approach is that it recognises that individuals may differ in a linear way in regard to the extent of their symptoms. This fits with phenomenological reality, but it also fits with biological reality to a large extent. For example, individual differences in neurotransmitter levels, receptor quantity, neural structure and functionality, hemispheric differences, and neurological utility all vary gradually between individuals. In other words, even if individuals were categorised into pathophysiologically distinct groups, it would require a judgement to be made about the level of a given biologically entity or occurrence that should indicate a disordered state.

Like the pathophysiological approaches however, the dimensional approaches do not acknowledge biopsychosocial interactions. Although certain dimensions may summarise a generally co-occurring set of biological markers which influence the way an individual may interact with their environment, it does not acknowledge the importance of stressors, stressor perception, and coping styles. It is also unlikely that a dimensional system that is not based on a clear aetiology would provide useful guidance for therapeutic intervention; whether it is psychotherapeutic or psychopharmacological.

Despite these limitations, any system of psychiatric diagnosis would benefit from including some sort of dimensionality in its approach, as it would acknowledge that diagnostic boundaries must acknowledge individual differences in biopsychosocial interactions. In doing so, the primary problem that remains is this: How can symptom-based diagnoses be said to have distinct aetiologies when there is such a complex range of biopsychosocial interactions between causal factors and phenomenological manifestations? The answer to this question may perhaps be guided by a reconsideration of the nature of psychological dysfunction.

References

Alicke, M. D., Klotz, M. L., Breitenbecher, D. L., Yurak, T. J., & Vredenburg, D. S. (1995). Personal contact, individuation, and the better-then-average effect. Journal of Personality and Social Psychology, 68, 804-825.

American Psychiatric Association. (1952). Diagnostic and statistical manual of mental disorders. Washington, D.C.: American Psychiatric Association.

American Psychiatric Association. (2000). Diagnostic and statistical manual of mental disorders (4th ed., revised ed.). Washington, D.C.: American Psychiatric Association.

Charney, D.S., Barlow, D.H., Botteron, K.N., Cohen, J.D., Goldman, D., Gur, R.E., Lin, K.-M., Lopez, J.F., Meador-Woodruff, J.H., Moldin, S.O., Nestler, E.J., Watson, S.J., Zalcman, S.J., 2002. Neuroscience research agenda to guide development of a patho-physiologically based classification system. In D.J. Kupfer, M.B. First, & D.A. Regier. (Eds.). A Research Agenda for DSM-V. Washington D.C.: American Psychiatric Association, pp. 31–83.

Chawla, N. & Ostafin, B. (2007). Experiential avoidance as a functional dimensional approach to psychopathology: an empirical review. Journal of Clinical Psychology, 63, 871-890.

Cooper, R. (2004). What is wrong with the DSM? History of Psychiatry, 15, 5-25.

Cuthbert, B.N. (2005). Dimensional models of psychopathology: research agenda and clinical utility. Journal of Abnormal Psychology, 114, 565-569.

Eifert, G.H. & Forsyth, J.P. (2005). Acceptance and commitment therapy for anxiety disorders: a practitioner’s treatment guide to using mindfulness, acceptance, and values-based behavior change strategies. Oakland, CA, US: New Harbinger.

Gross, J.J. & Thompson, R.A. (2007). Emotion regulation: conceptual foundations. In J.J. Gross (Ed.). Handbook of emotion regulation, pp.3-24. New York: Guilford Press.

Hayes, S. C., Wilson, K. G., Gifford, E. V., Follette, V. M., & Strosahl, K. (1996). Experiential avoidance and behavioral disorders: a functional dimensional approach to diagnosis and treatment. Journal of Consulting and Clinical Psychology, 64, 1152-1168.

Kupfer, D.J., First, M.B., & Regier, D.E. (Eds.). (2002). A research agenda for DSM-V. Washington, D.C.: American Psychiatric Association.

Luyten, P., Blatt, S.J., Van Houdenhove, B., & Corveleyn, J. (2006). Depression research and treatment: are we skating to where the puck is going to be? Clinical Psychology Review, 26, 985-999.

Monroe, S.M. & Simons. (1991). Diathesis-stress theories in the context of life stress research: implications for the depressive disorders. Psychological Bulletin, 110, 406-425.

Pine, D.S., Alegria, M., Cook Jr., E.H., Costello, E.J., Dahl, R.E., Koretz, D., Merikangas, K.R., Reiss, A.L., Vitiello, B. (2002). Advances in Developmental Science and DSM-V. In D.J. Kupfer, M.B. First, & D.A. Regier. (Eds.) Research Agenda for DSM-V. Washington, D.C.: American Psychiatric Association.

Widiger, T.A. & Samuel, D.B. (2005). Diagnostic categories or dimensions? A question for the diagnostic and statistical manual of mental disorders – fifth edition. Journal of Abnormal Psychology, 114, 494-504.

World Health Organization. (1992). International statistical classification of diseases and related health problems (10^th ed.). Geneva: World Health Organization.