Schizoaffective disorder

Summary

The term ‘schizoaffective disorder’ has meant different things to different people over the 80 years since Kasanin described a series of cases with an admixture of schizophrenia-like and mood-like symptoms. This article provides a brief overview of the history of the concept, and suggests a parsimonious approach for clinicians.

The very name ‘schizoaffective disorder’ suggests some sort of hybrid – betwixt and between, a nosological inconvenience. Kraepelin, of course, ruled supreme for many years, with his famous ‘dichotomy’ between dementia praecox and manic depressive insanity, but even he was aware of ‘interform illnesses’, in which there is a ‘mingling of morbid symptoms of both psychoses’ (Kraepelin 1921). It was Kasanin, though, who in 1933 set the proverbial cat amongst the pigeons, publishing a report of nine patients who had both ‘schizophrenia-like’ and ‘mood-like’ symptoms that declared themselves abruptly and resolved within weeks to months but had a tendency to recur: the term ‘schizoaffective’ was born (Kasanin 1933).

Of horses, donkeys and points of rarity

Many of the finest minds in modern psychiatry have attempted to bring clarity to this troublesome area, perhaps most elegantly the late Bob Kendell. He suggested that, if we consider schizophrenia and affective psychosis to be represented by ‘related but distinct’ animals (horses and donkeys, respectively), the group of patients subsumed under the schizoaffective label could be horses (schizophrenia with some affective symptoms), donkeys (vice versa), donkeys and horses harnessed in pairs (having both illnesses at the same time), zebras (an unrelated illness), or mules (a genuine interform illness) (Kendell 1988). There have been numerous attempts to seek a ‘point of rarity’ between the horses and the donkeys – most concertedly, perhaps, by Kendell himself, who eventually conceded defeat (e.g. Kendell 1980). Others have simply felt that the inability to determine different groupings of animals is indicative of there actually being only one animal: a unitary psychosis. Still others, notably Karl Leonhard, delineated many different types of animal (subtypes of psychotic disorder), some of which could be mapped onto the ‘schizophrenia’ end of the continuum, some onto the ‘affective’ end, and some in the middle. Some of the last group, such as the cycloid psychoses, could be seen as types of schizoaffective disorder. Needless to say, the boundaries of the zoo itself, let alone the way in which the animals therein should be separately penned, have still not been resolved.

The lack of a phenomenological point of rarity led researchers to assess whether there are other factors that could shore up the schizoaffective concept. People with schizoaffective disorder tend to have a later onset of illness than those with schizophrenia (the late 20s is the mean), are more likely to be female and tend to have episodic course and good intermorbid functioning (better than in schizophrenia, worse than in bipolar disorder) (Sampson 1988; Jager 2004). Of course, none of these factors in itself delineates the disorder, and many may confound or interact with each other (e.g. having a later onset and better outcome being associated with being female). Family studies are somewhat helpful, as they point to an elevated familial risk for both schizophrenia and mood disorders in probands of those with schizoaffective disorder: mostly schizoaffective disorder does not ‘breed true’, challenging both the ‘continuum’ and ‘discrete disorder’ viewpoints (Bertelsen 1995).

DSM and ICD classification

The authors of the US Diagnostic and Statistical Manual of Mental Disorders (DSM) have struggled with the schizoaffective notion over ensuing editions of their ever-larger tome. Indeed, DSM-III (American Psychiatric Association 1980) simply gave up and left schizoaffective disorder as an orphan without even a set of operationalised criteria. In the revision of 1987, DSM-III-R, the DSM committee adopted an about-face, and applied very restrictive criteria for schizoaffective disorder, including full affective and full criterion A schizophrenia symptoms, which had to be present for at least 2 weeks in the absence of mood symptoms. This general approach has carried through to subsequent editions of DSM. The 10th edition of the International Classification of Disorders (ICD-10; World Health Organization, 1992) is more lenient, emphasising psychotic symptoms at the same time as mood symptoms.

Does the concept have utility?

Ultimately, the true utility of categorisation lies in whether there are implications for treatment. In this regard, some authors have suggested that two subtypes of schizoaffective disorder can be delineated. Thus, Levitt & Tsuang (1988) argued that patients with ‘schizomania’ respond better to mood stabilisers, whereas those with ‘schizodepression’ require antipsychotics and antidepressants, albeit they concede that many patients end up being treated with combinations of agents. This subtypology gains some support from the Roscommon Family Study, in which a latent class analysis categorised patients into one or other of these two classes (Kendler 1995). Having said this, there are interforms even between these two putative subtypes, and genetic studies do not show clear familial aggregation either.

So maybe, as in many areas of psychiatry, we just need to accept the uncertainty associated with the term schizoaffective. The DSM approach of making criteria so limited as to fit very few real-life patients seems unhelpful, and merely tries to impose yet more boundaries around concepts that have little validity. Clinicians probably already know that the best approach for their patients is to weigh up the primacy of the different sets of symptoms and to treat accordingly. In this context, the term schizoaffective does have utility, as it is arguably less stigmatising than schizophrenia as a label and it also keeps clinicians alive to the fact that at least two symptomatic dimensions are operating and requiring of treatment.