The heterogeneity of “major depression”

Abstract

The concept of major depression, in both the ICD and the DSM, has been a flagship for mental health in general medical settings. It is the principal mental disorder emphasized to general practitioners and hospital doctors, and it has been used by governments to raise awareness of mental health issues in the population at large and in the medical community 1,2,3. It has encouraged the development of specialized forms of psychotherapy for depression, it has allowed the formation of community groups to propagate accurate information about depression, and has led to the development of computer programmes to assist self-treatment using the principles of cognitive-behavioural therapy 4,5. These are substantial achievements, but they come at a price. This is the belief that “major depression” is a homogeneous entity, and indeed that it is usually a “major” disorder. The DSM diagnosis of major depression is made when a patient has any 5 out of 9 symptoms, several of which are opposites. Thus, a patient who has psychomotor retardation, hypersomnia and gaining weight is scored as having identical symptoms as another who is agitated, sleeping badly and has weight loss. This causes real problems with research designs: for example, Lux and Kendler 6 had to group these opposite symptoms together. Even so, it was possible for them to distinguish between “cognitive” and “neurovegetative” symptoms, and show that these have different relationships to a larger set of potential validators. They concluded that their results “challenge our understanding of major depression as a homogeneous categorical entity”. Others have been able to separate the various depressive symptoms, and to compare the relative efficiency of each symptom to making the diagnosis 7. Jang et al 8 factor analysed a larger set of depressive symptom scales, and found that they could identify 14 different subscales, which had rather low intercorrelations, and very different heritabilities. Given these findings, to declare that all those satisfying the DSM criteria for the diagnosis of major depression are suffering from the same disorder seems like magical thinking. We know that many milder cases remit without specific treatment, suggesting that they are indeed homeostatic responses to life stress 9. Depression may be a toxic reaction to drugs or may result from endocrine disorders such as myxoedema or Cushing's syndrome. The depressed phase of bipolar illness may be difficult or impossible to distinguish from unipolar depression. Melancholic, atypical and psychotic forms of depression are yet other variants. Agitated depression needs to be distinguished from retarded depression when choosing the most suitable antidepressant. But even with these exclusions, there are five other forms of major depression that require a range of different responses from the clinician. The first form is depression presenting with somatic symptoms 10. Many patients with this condition may be resistant to accepting that they are depressed. They benefit from special additional measures that explain how emotional arousal and depression can cause their somatic symptoms. Such measures have been developed for many years, and have recently been elaborated 11. The second form is depression with panic attacks. While treatment of depression is the first priority in these patients, it is also important to give them advice on what to do during a panic attack, as it may take a little time before improvement in their depression stops further attacks. They need advice about not immediately leaving the environment in which the panic attack is taking place, explanations about catastrophizing thoughts and advice on helpful “self-talk”. They need to remind themselves that they have had such attacks before, and they will pass off if they calm down and remember the reassuring thoughts that run counter to the content of their thoughts during an attack. Such advice makes the attacks easier to deal with, and less likely to become still worse. The third form is depression in people with obsessional traits. People with these traits in their usual personality often develop quite severe obsessional behaviour and depressive ruminations during a depression. These symptoms may be experienced as the leading symptoms, but can be thought of as epiphenomena of their depressive illness. It is helpful to take the patient through thought-stopping techniques, distraction techniques and response prevention. The fourth form is depression accompanying known physical illnesses. These depressions are particularly poorly recognized by generalists, who typically confine themselves to the treatments for the physical illness 12. Diagnosis of these depressions is complicated by the fact that four of the “diagnostic features” of depression (fatigue, poor sleep, poor appetite and weight loss) may well be caused by the physical illness. This may generate confusion, since no clear threshold for the number of symptoms needed for a diagnosis seems to exist if such symptoms are to be discounted. However, if there is a positive reply to either of the usual two screening questions for depression, it is only necessary to ask three additional questions dealing with poor concentration, ideas of worthless and thoughts of death. A total of three or more from this list of five symptoms allows depression to be diagnosed with high sensitivity and specificity, when assessed against the full list of criteria 13,14. Successful treatment of the depression is associated with a lower mortality and better collaboration with the necessary physical treatments. The special task of the physician is to reach agreement with the patient that he/she is indeed depressed, and to explain the effects that this is having on the quality of the patient's life, the severity of any pains that are experienced, and the disability associated with the physical illness. The range of treatments that are effective in depression among the physically healthy are all effective in these patients, and the only special measure required of the clinician is to guard against harmful interactions between antidepressants and drugs used for the physical illness. The fifth form is pseudo-demented depression. In older people, depression may present as an apparent dementia, but the presenting symptoms turn out to be due to inattention and impaired concentration, while symptoms of depression are undoubtedly present and may be elicited by direct enquiry. The special task here is to reassure both patient and carer that the memory problems are not due to cerebral disease, and are likely to improve a great deal with treatment of the depression. Official classifications of mental disorders often deal with the above heterogeneity by invoking the idea that an individual patient has simultaneously developed more than one “co-morbid” disorders. Many depressions are likely to be accompanied by anxious symptoms, so these disorders should more properly be described as “anxious depressions”. The concept of “co-morbid generalized anxiety disorder and major depression” does not describe most cases of anxious depression, since for this concept the anxious symptoms should have lasted 6 months, while the depressive symptoms need only have lasted 2 weeks. It is therefore describing a depressive reaction grafted on to a chronically anxious person, and is thus a more restrictive concept than anxious depression. However, anxious symptoms are by no means the only symptoms that often occur with depression, as most patients presenting to generalists will present with the various combinations described above. The concept of “co-morbidity” has not led to the development of special recommendations for the management of the very different ways in which a depressive illness presents in different people. Instead, while there are recommended treatments for each separate mental disorder, the implication has been that these treatments are just added together for each co-morbid disorder. Symptoms arising as epiphenomena of depression do not necessarily need the same range of interventions as when similar symptoms are occurring in a non-depressed person, but they do need some help. It is also worth remembering that telling a person that he/she has multiple mental disorders is both stigmatizing and somewhat depressing. It may be questioned whether it is worth making these distinctions between the various subtypes of depressive illness, since once a remission has occurred all these ancillary symptoms are likely to have remitted anyway. The justification is partly due to the need to provide different advice for the ancillary symptoms while the episode lasts, as well as the possibility that remission might occur more quickly if additional advice is provided for the patient during the episode. The five subtypes listed above have been chosen because they each have particular features that attract different clinical approaches to the problem posed by the depression. Rather than making multiple diagnoses, it seem preferable to have regard to the principal manifestations of the patient's present problems, and to respond appropriately to them. If we develop depressive symptoms, we may develop other symptoms, dependent upon any vulnerability factors in early life, on our personality structure, and on stressful features in our present social environment. It is profoundly mistaken to assume that the various common symptom patterns described are rigidly demarcated, and that a classification exists in which the various syndromes are “mutually exclusive and jointly exhaustive”. The clinician must aim to give useful advice to the particular patient seen, without foisting an arcane system of multiple diagnostic labels onto him/her. At present major depression has become a monolith, with the assumption that the diagnosis can be made merely on the number of depressive symptoms present, with an associated disability. It may be politically important to utter such simplifications to doctors in general medical setting, but it is a convenient fiction.

Cite this work

No public copy is stored for this record, so only the metadata is cited.

Stored only when both papers are in the corpus — 335 citations exist upstream.

Papers that cite the same references — the relational fallback while embeddings are still backfilling.