Discussion of Issues Surrounding the Classification and Diagnosis of Schizophrenia

One issue related to and is reliability. Reliability refers to the consistency of a measuring instrument, such as DSM (the and Statistical Manual) that is used when diagnosing schizophrenia. Reliability can be measured in terms of whether two independent assessors give similar scores (inter-rater reliability). High reliability is indicated by a high positive correlation. Inter-rater reliability has been assessed for diagnoses of schizophrenia and found to be relatively low.

This was especially true for earlier versions of DSM but it was hoped that later revisions of DSM would prove more reliable. However, more recent versions have continued to produce low inter-rater reliability scores. For example, Whaley (2001) found only a small positive correlation of +0. 11 between different raters. Differences in cultural interpretations also pose a threat to the reliability of the diagnosis of schizophrenia. A research study by Copeland et al. (1971) gave a description of a patient showing clinical characteristics associated with schizophrenia to US and UK psychiatrists.

Of the US psychiatrists, 69% diagnosed schizophrenia, whereas only 2% of the UK psychiatrists gave the same diagnosis. This suggests that the diagnostic criteria had quite a different meaning in different cultures and therefore are not reliable when used in different cultural settings. Reliability is also an issue for diagnosis. This was raised by Rosenhan (1973) who claimed that situational factors were more important in determining the ultimate diagnosis of schizophrenia, rather than any specific characteristics of the person. Rosenhan demonstrated this in his well-known study called ‘Sane in insane places’.

He arranged for ‘pseudopatients’ to present themselves to psychiatric hospitals claiming to be hearing voices (a symptom of schizophrenia). All were diagnosed with schizophrenia and admitted, despite the fact they displayed no further symptoms during their hospitalisation. Throughout their stay, none of the staff recognised that they were actually normal. The unreliability of diagnosis was further demonstrated in a follow-up study by Rosenhan. Psychiatrists at several mental hospitals were told to expect pseudopatients over a period of several months.

This resulted in a 21% detection rate by the psychiatrists, even though none were actually sent. This shows that the diagnostic criteria used by psychiatrists could not reliably identify a person with schizophrenia. A second issue is validity which concerns both classification and diagnosis. For example, there is the issue of comorbidity which is related to the validity. Comorbidity refers to the extent that two (or more) conditions co-occur (such as schizophrenia and depression) and therefore the extent to which the condition is ‘real’ and distinct.

One way to avoid the issue of comorbidity is to just use first-rank symptoms of schizophrenia when diagnosing (e. g. delusions or hallucinations). However, Bentall et al. (1988) claim that many of the first-rank symptoms of schizophrenia are also found in other disorders (e. g. depression and bipolar disorder). This makes it difficult to separate schizophrenia as a distinct disorder from other disorders and suggests that schizophrenia is not a distinct condition. It may be more realistic to suggest that there is no such discrete disorder as schizophrenia but instead there is a spectrum of psychotic symptoms.

Allardyce et al. (2001) claim that symptoms used to characterise schizophrenia do not define a specific disorder because its symptoms are also found in other categories of psychosis described in DSM and therefore there should just be a psychotic spectrum. Another aspect of validity is predictive validity. Predictive validity demonstrates the validity of a diagnosis by demonstrating that it can predict scores on some criterion measure. If a disorder has high predictive validity then it should be clear how the disorder would develop and how people would respond to treatment. Research has found low predictive validity for schizophrenia.

Some patients (about 20%) do recover their previous level of functioning but 40% never really recover. This much variation in the prognosis suggests that the original diagnosis lacked predictive validity. It means that diagnosis was not helpful in dealing with the course of schizophrenia. Research has shown that other factors may be more influential on the ultimate outcome of having schizophrenia. For example, it seems more to do with gender (Malmberg et al. , 1998) and psychosocial factors, such as social skills, academic achievement and family tolerance of schizophrenic behaviour (Harrison et al. , 2001).