First described by Hippocrates, “melancholia” or melancholic depression was considered a specific condition that commonly struck people out of the blue – and put them into the black. In modern times, it came to be described as “endogenous depression” (coming from within) in contrast to depression stemming in response to external stressors.
In 1980, the third edition of the Diagnostic and Statistical Manual (DSM-III), the official classificatory system of the American Psychiatric Association, re-modelled depressive disorders. The new classification operated largely on degrees of severity, comprising “major” depression and several minor depressions.
This is how depression came to be modelled as a single entity, varying only by severity (this is known as the dimensional model). And over the last decade, this model has been extended to include “sub-clinical depressions”, which is basically when someone is sad or down but not diagnosable by formal mental illness criteria.
The changes generated concern about the extension of “clinical depression” to include and “pathologise” sadness. While everyone feels down or sad sometimes, normally these moods pass, with little if any long-term consequences.
The boundary between this everyday kind of feeling down and clinical depression is imprecise. But the latter is associated with a greater severity of symptoms, such as losing sleep or thinking life isn’t worth living, lasts for longer and is much more likely to require treatment.
The dimensional model is intrinsically limited; “major depression” is no more informative a diagnosis than “major breathlessness”. It ignores the differing – biological, psychological and social – causes that may bring about a particular depressive condition and which inform the most appropriate therapeutic approach (be it an antidepressant drug, psychotherapy or social intervention).
Ignoring the cause of depression leads to both under-treatment, such as failure to prescribe an effective medication, and over-treatment, such as prescription of medication that’s unnecessary and may have side effects.
The model also essentially marginalised melancholia as a categorically different type of depression, with progressive DSM manuals according it insignificant status as a major depression “specifier” (an addendum to a diagnosis intended to provide more detail).
As a specifier, and not a disorder in its own right, melancholia is not considered categorically separate to other types of depression. And this matters – much less research and training is devoted to it as a result, and doctors are often unaware of its clinical implications.
A distinct pattern
My research team is trying to establish melancholia’s categorical status and detection, and so improve its management. Here’s what we know – or think we know - about the distinctness of melancholia.
First, it shows a relatively clear pattern of symptoms and signs. The individual experiences profound bleakness and has no desire to socialise, for instance, finding it hard to obtain any pleasure in life or to be cheered up.
Sufferers also experience a lack of energy and have difficulty concentrating, although they generally show “diurnal variation”, reporting improvement in mood and energy as the day goes on. Reflecting changes to their sleep/wake cycle, people with melancholia tend to wake early in the morning.
Episodes commonly emerge “out of the blue”. Even if it follows a stressor, it’s disproportionately more severe than might be expected and lasts longer than the stressor.
Melancholia has a strong genetic contribution, with sufferers likely to report a family history of “depression”, bipolar disorder or suicide. It’s largely biologically underpinned rather than caused by social factors (stressors) or psychological factors, such as personality style.
The illness is also unlikely to respond to placebo, whereas major depression has a placebo response rate in excess of 40%. But melancholia shows greater response to physical treatments, such as antidepressant drugs (especially those that work on a broader number of neurotransmitters), and to ECT (electroconvulsive therapy). ECT is rarely required, however, if appropriate medications are prescribed.
Melancholia shows a lower response to psychotherapy, counselling and psychosocial interventions - these treatments are more salient and effective for non-melancholic depression.
It’s useful to draw an analogy here with diabetes: while Type 1 is more a biological disease state and generally requires drug treatment (insulin), Type II is more likely to reflect other factors, such as obesity. The latter generally benefits most from non-drug strategies, such as exercise and dietary changes.
Melancholia shows similar “treatment specificity”, with medication being the treatment of choice.
Tracing biological origins
Melancholia has long been thought to have primary biological origins, including perturbations in the hypothalamic-pituitary-adrenal (HPA) axis, in sleep architecture and in neural circuits.
Early this year, our research team published a neuroimaging study that suggested a differential key “signature” marker found only in people with melancholic depression (when compared to people with non-melancholic depression and non-depressed controls).
We showed incoming connections to the brain system that control attention (the insula) were halved, while connections from the insula to the brain’s executive control centre were also decreased.
The implications of these findings will require further investigation, but they could mean that a disruption to brain connectivity may explain some of melancholia’s symptoms.
Clearly, melancholia needs to be recognised as a distinct psychiatric condition – not simply as a more severe expression of depression. This recognition could lead to improved clinical and community awareness, which is important because managing melancholia requires a specific treatment approach.