Discuss issues of bias in diagnostic systems

schizophrenia. This indicates the problem of test–retest reliability with schizophrenia diagnoses. Other research such as Rosenhan (1973) study with pseudo patients suggests that diagnostic systems can be reliable. There were eight pseudo patients admitted, and all of them had a consistent and reliable diagnosis. Although, the diagnosis was reliable and consistent, realistically the diagnosis was wrong as the pseudo patients were not mentally ill, raising issues of the validity of diagnosis systems. The key concern for diagnostic systems is whether they correctly diagnose people who really have particular disorders and do not give a diagnosis to people who do not. It is difficult to establish whether a person truly has a disorder without using a diagnostic system. This means that the only people we can be fairly sure about are those who have already been diagnosed, although many will argue this is also insufficient. This problem is clearly highlighted in Rosenhan’s study with pseudo patients. The pseudo patients were completely ‘normal’ people, but they were diagnosed as abnormal after saying they were experiencing one symptom. This is evident that diagnostic systems can be invalid, and type 1 and type 2 errors can be made. In study 1 doctors had a strong bias towards type 2 errors as they were more inclined to call a healthy person sick. However, in study 2 they made more type 1 errors as they were trying to avoid type 2 errors, by naming sick people as healthy. Although this study was conducted in 1973, it provides some important evidence for a general inability to tell the difference between normal and abnormal behaviour. Thomas Szasz also suggested that it is wrong to use a mental illness metaphor to describe behaviour that does not conform to our expectations. It is clear that there are biological correlates of behaviour but it is reductionist to assume that conditions like depression and schizophrenia are diseases like any other, especially as biological causes have not yet been found for most psychological disorders. In particular, it is important to note that the terms ‘depression’ and ‘schizophrenia’ are essentially labels given to a set of behaviours, emotions, or thoughts.  It is somehow more attractive for many of us to imagine that an underlying condition called depression is the cause of a person’s severe unhappiness, but there is no underlying condition tested for. The depression ‘is’ the unhappiness rather than the cause of it. The fact that diagnoses can be unreliable and invalid suggests they should not be used. However, the same is at least partly true of medical diagnosis generally, yet we wouldn’t suggest abandoning that. The DSM and other methods such as the ICD are continually updated so reliability and validity are always improving. A major issue with psychiatric diagnosis is that they result in labelling. For instance, a person becomes a ‘schizophrenic’ rather than a person with schizophrenia, a label that tends to stick even when the disorder has disappeared.  An invalid psychiatric diagnosis has serious and lifelong implications. For this reason, many critics prefer to avoid the use of such labels. An alternative is to use a more ideographic approach that doesn’t require classification but emphasis analysing each patient’s problems individually. Different parts of the world and different ethnic groups have different ways of explaining their behaviour. It must be asked whether the DSM criteria are valid beyond the culture they were created in. Cooper (1994) has suggested that social consequences should be included when defining features of a disorder, since the social environment of individuals varies so widely between cultures. This is because the same symptoms and behaviour that are tolerated in one culture may cause severe social problems in another culture, and it is unreliable for diagnostic decisions to be determined by cultural and social definitions. Different societies have different views about the causes of mental disorders, and therefore, are likely to have different views about treatment. Certain cultural and subcultural groups are treated differently. For example, Cochrane and Sashidharan (1995) found that black Afro-Caribbean immigrants in the UK are up to seven times more likely to be diagnosed with schizophrenia then white people. In terms of subcultural bias, Johnstone (1989) found that lower-class patients were more likely to spend longer in hospital, be prescribed physical rather than psychological treatments, and have a poorer prognosis. It is possible that the over-diagnosis of mental disorders in certain cultural groups is due to genetic differences rather than cultural bias, for instance, it might be that black Afro-Caribbean’s have a greater genetic vulnerability for schizophrenia. However, rates of diagnosis for Afro-Caribbean’s are not as high elsewhere in the world as they are in the UK. Another alternative explanation for the high diagnosis rates in the UK may be that members of minority ethnic groups in Britain have more stressful lives and this makes them more likely to develop such illnesses. Therefore, the difference would be due to social/environmental factors rather than biased diagnosis.  There can also be gender issues of bias in diagnostic systems. Kaplan (1983) argued that the DSM system emphasises problems in women, making them appear more prone to mental illness and instability than men, although others say this is not the case. The main argument centres round the fact that for some mental illnesses more women are diagnosed then men. However, this may not mean that there is a sex bias in the categorisation system so much as that there are biological or genetic reasons for the gender differences in diagnosis. For example, men are more likely to be diagnosed as having anti-social personality disorder but this may be due to biological factors, such as an excess of testosterone. Broverman et al (1970) were among the first to highlight gender bias as an issue in the diagnostic system of mental disorders. In their study they found that clinicians were more likely to characterise males with traits of healthy adults than they were to attribute such traits to females. Other research also suggests gender bias; Cohen (1983) implied that more women are more likely to be subject to labelling in ‘Borderline Personality Disorder’. This is an over diagnosis of women with BPD as the same traits may be seen as more tolerable and acceptable in men than women. Men be more ‘accepted’ with traits such as being short tempered, impatient etc. whereas with women it is more likely to be seen as ‘abnormal’ if they share such traits. It is possible that at least some cases of differential diagnosis are due to real differences rather than bias. For example, gender differences in social roles and in life experiences e.g. pregnancy and child rearing, may explain why more women are more likely to experience depression than men. The difference explanation is supported by the fact that, although rates of depression are twice as high in women as men, they are similar for bipolar disorder. If there was a systematic gender bias in diagnosis, one would expect to be reflected in bipolar disorder as well because it relies on some of the same symptoms. In conclusion, is it really fair to say that there is bias in diagnostic systems? In terms of the systems being valid, each time the DSM-IV-TR is revised, new categories are added and old categories are dropped, presumably because they are not sufficiently useful. This may be a good thing, however, if they are removing old categories, it suggests that people who were diagnosed with the old categories, may have perhaps had the wrong diagnosis. Also, DSM-IV-TR does not classify clinical problems into syndromes in the simplest and most beneficial way. In diagnostic systems, there may be apparent cultural bias. Systems such as the DSM may not consider that in some cultures what is considered normal may be considered abnormal in others. Therefore, If an individual from one culture were living in a society that had a different culture, they may be labelled as abnormal and start to think they actually have an illness. Can the DSM generalise across the world? It raises issues that there are a lot of wrong diagnoses due to the systems only relating to one type of culture, if perhaps the systems could generalise across the world and take into consideration ethnicity, race, religion etc. then there may be less ‘cultural bias’. In terms of gender bias, it seems that in mental illnesses such as depression, a lot more women seem to be diagnosed. A common explanation for this would be that perhaps women suffer more from the condition due to having children. However, there is no evidence of a difference between men and women psychologically, e.g. men being stronger psychologically than women, this again may just be labelling. It should be taken into consideration that perhaps the reason for the high rates of diagnosis with women having depression, may just be down the fact they are open to speak to people about the problem, and men may be more likely to ‘hide’ away from the problem. Realistically, there may be just as many men suffering with the same conditions as women. Systems need to consider all factors, and not just diagnose on a set of symptoms or just assume that someone is depressed because they are ‘unhappy’. Charlotte Barrow