A spectrum of views

There are few absolutes in psychiatry and most models of mental illness are derived from piecemeal observations and limited research evidence. This makes them difficult to prove and equally difficult to defend. In practice, while mood symptoms feature strongly in many varied clinical presentations, capturing the essence of mood disorders, and then delimiting these from others, has proven immensely difficult. This raises questions as to whether mood is the best feature to define depressive and bipolar disorders.

Causal classifiers of mood disorders, such as endogenous and reactive, although clinically meaningful, have been difficult to anchor to neural substrates. This is partly because research has revealed reciprocal interactions between predisposing factors compounding the complexity of their actions. For example, the relationship between genes and environment (Caspi et al., 2003) is far more sophisticated in reality than initially anticipated. In addition, there are differences between aetiology and pathogenesis, highlighted by Ghaemi and Dalley (2014), which further obfuscate understanding of processes that underpin the development of mood disorders.

Another approach has been to partition mood disorders into those that are ‘primary’ or ‘secondary’. This allows for an agnostic classification, whilst still presumably differentiating separate forms of the illness. But descriptive classifications of the spectrum of mood disorders have predictably struggled because of the inherent limitations of sampling and of ‘knowing what to look for’. In this regard, perspective has proved critical. By way of illustration, white light shone through a prism creates a spectrum of diverse colours; but the same white light could be passed through filters to create an entirely different ‘polarised’ perspective. Like the blind wise men of parable, we may perceive the same elephant entirely differently. Thus, although our concepts of mood disorders may well be true, because they are incomplete they are likely to be fallacious. Our perception of the spectrum of mood disorders derives from differing orientations and what we consider the boundary of the spectrum to be, particularly whether it merges with psychosis (with common genes), personality dysfunction (and its boundary with borderline personality) or ‘normal’ variations in mood. Consequently, our present approach is subject to extreme variability and can result in the sudden emergence or loss of a disorder and/or its transformation (Malhi et al., 2014; Malhi, 2013).

Alas, the fundamental problem remains: we have not managed to define mood disorders in a manner that correlates accurately with biological substrates that meaningfully inform treatment and clinical outcome. The clinical picture of established mood disorders is almost always complex and is the consequence of a multitude of factors that are often difficult to appraise and sometimes impossible to disentangle. The desire to observe mood disorders as they take form, so as to understand their origins, is logical but confounded by the fact that presentations early in the course of illness are the most varied and volatile (Das et al., 2013) and, when emerging in adolescence, difficult to disentangle from developmental stresses, so their evolution is unknown. Thus, over the course of mood disorders (longitudinally), research is limited by diagnostic uncertainty at the outset and increasing complexity as the illness progresses. Cross-sectionally, bottom-up approaches attempting to identify the ‘joints of nature’ have thus far failed to find markers that provide clinical clarity. As discussed, top-down, clinical differentiation lacks precision (especially when widely heterogeneous disorders such as depression are considered) and, so far, have not mapped onto neural substrates. In practice, a patient with a mood disorder has a unique experience, which has many facets, each of which they can choose to focus upon when asked to do so (sleep disturbance, mood, motivation, cognition, etc). For the individual, the illness has a sense of continuity – reflecting the dimensionality of symptoms.

Ghaemi and Dalley (2014) are therefore correct to raise concerns: some things are clearly a poor fit. For example, the fact that ‘mixed states’ occur in reality indicates that a taxonomy tethered to mood is probably incorrect. Furthermore, the evolution of mania from bipolar depression and the eruption of mania in response to antidepressant treatments suggest that the complete separation of depressive and bipolar disorders is perhaps not possible. Additionally, genetic studies place bipolar disorders in closer proximity to psychoses, again indicating that our perspective is perhaps flawed.

What is the clinician to make of all of this in relation to identifying bipolarity? We have to acknowledge, and be skeptical of, the reality of a spectrum that presents differently from different perspectives. It is clear that relying on cross-sectional simplistic assessment of our patients is prone to failure and that, instead, we need to undertake comprehensive longitudinal assessments, which carefully probe for genetic, developmental and psychosocial factors. Only then can we hope to determine with some degree of certainty whether an individual lies within the bipolar spectrum. Only when we have a thorough understanding of the individual can treatments be applied with some confidence.

It is no wonder that there are many misconceptions and that, in addition to a spectrum of disorders, we have a spectrum of views.