In a previous post I described how current methods of categorising mental health problems are not particularly reliable or valid. As Kendell and Jablensky (2003) point out, the lack of a clear dividing line between different categories of mental illness is a serious theoretical problem.
Nevertheless, the categories, flawed as they are, do have practical utility. Indeed, the DSM-IV states that no assumption should be made that there are entirely separate categories of mental illness. Similarly it states no assumption should be made that there is a clear dividing line between the sane and the insane. Despite these disclaimers, what is the practical effect of categorisation?
The three most important uses of a classification system are in aetiology, treatment and prognosis; or where the illness has come from, what can be done about it and, what might happen to you. Perhaps the most important of these for the patient is prognosis: “What will happen to me?”.
Bentall points out the difficulty of assessing the outcomes of people with a mental illness because they are often variable across different domains. Clinical outcome attempt to measure the persistence of symptoms, occupational outcomes measure the ability to keep a steady job, while social outcomes measure a patient’s network of relationships.
Despite this, Bentall reports 3 general findings from the evidence on the outcomes of those diagnosed as psychotic.
- The course of psychosis is unpredictable. The only fairly solid finding is that the occurrence of psychotic episodes seem to increase the probability that they will occur again in the future. This process is known as kindling.
- There is little predictive power for individual patients from a diagnosis. This conclusion comes from the evidence that there is enormous variability in the outcomes of people with the same diagnosis.
- Patients with a diagnosis of manic depression have a better outcome, on average, than those with a diagnosis of schizophrenia. Despite this, evidence has shown there is no sharp delineation between the two diagnoses. Kendell & Brockington’s (1980) study showed the difference manifested itself in a sliding scale. The more a person’s symptoms were typical of schizophrenia rather than an affective disorder, the poorer was the outcome.
Moving away from the patient and thinking of the utility of these categories for the academic, though, how do the fuzzy categories affect research? At first sight, it seems that having a number of categories of mental illness is a serious advantage in research. Before the standardisation introduced in the 70s by categorisation, researchers in different places used different definitions so it was difficult to compare results and theories. Categories provide a common language – and that can provide synergy through a common understanding.
The problem is, over time, categories become embedded within the collective consciousness. It becomes accepted, whether explicitly or implicitly, that research can only be conducted with reference to, and with the use of, the official categories. This clearly limits the possibilities for progress. Categories that have, theoretically, been found wanting, persist simply because the effort required to change that many minds is too great.
A related and more sociological idea is the effect of categorisation on the individual. There is an element by which once people receive a label, they then define themselves, and are defined by others, in relation to that label. A person suffering from manic depression becomes less a person than ‘a manic depressive’. If this category is theoretically flawed, then what does this mean about how those suffering mental illness are being defined by others and defining themselves?
More to come about aetiology and treatment in further posts.