Issues in the Diagnosis & Classification of Schizophrenia (AQA A Level Psychology)

Reliability & validity in diagnosis

Reliability

• For a diagnosis of schizophrenia to be reliable it must show consistency and agreement across diagnosing clinicians i.e. the same set of symptoms must be given the same diagnosis regardless of who is doing the diagnosing 

• Inter-rater reliability refers to the above point: if more than one clinician is diagnosing the same patient then they should both/all agree as to the diagnosis 

• Issues with reliability occur when (as does happen) clinicians disagree as to the diagnosis

• Unreliable diagnosis may happen if clinicians do not use the same diagnostic tools e.g. one clinician uses the DSM and other uses the ICD, regardless of year of publication of both of these manuals

Validity

• Validity of diagnosis refers to the ‘realness’ or 'trueness' of what is being measured i.e. is the classification system being used to diagnose the patient actually set up to diagnose schizophrenia or does it fall short of this criterion? 

• An invalid diagnosis may occur if a clinician, having reviewed the patient’s symptoms, gives a diagnosis of an illness that does not actually fit those symptoms e.g. Emma believes that she has multiple personalities and her doctor diagnoses schizophrenia (the correct diagnosis should be dissociative identity disorder)

Research which investigates reliability & validity in diagnosis

• Santelmann et al. (2016) conducted a meta-analysis of 25 studies with a total sample of 7912 patients diagnosed by different raters and found that reliability of schizophrenia diagnosis had consistently lower inter-rater reliability than the diagnosis of major depressive disorder and bi-polar disorder

  • Strengths of this research include reliability due to the large sample size and robust quantitative data

  • Limitations of this research include potential selection bias regarding the studies chosen for the meta-analysis

• Rosenhan (1973) tested the validity of schizophrenia diagnosis in a field experiment in which he and eight confederates reported false symptoms and were all (but one) admitted to mental hospitals with a diagnosis of schizophrenia 

  • Strengths of this study include high ecological validity due to the naturalistic setting and naive participants plus the insight the findings gave into the 'sticky' label of diagnosis

  • Limitations include lack of ethical validity due to the deception involved plus the lack of control inherent in field experiments

Do all patients receive a diagnosis that is both reliable and valid?

Co-morbidity & symptom overlap

• Co-morbidity is when one patient is diagnosed with two or more mental illness

  • E.g. patients with schizophrenia are at an increased risk for the development of depression as the two illnesses may share a common aetiology and/or genetic basis

  • One of the difficulties in diagnosing a co-morbid patient is trying to ascertain the extent of one illness over another particularly when there is symptom overlap (e.g. Is this depression or is it part of patient X’s schizophrenia?)

• Symptom overlap is when two or more illnesses share some of the same symptoms

  • e.g. avolition (a negative symptom of schizophrenia) overlaps with symptoms of depression - lethargy, lack of motivation, neglecting  personal hygiene etc

  • One of the major issues for clinicians when faced with symptom overlap is that the DSM and the ICD use different criteria to classify the same symptoms

    • e.g. the DSM might produce a diagnosis of schizophrenia whereas the ICD diagnosis might be bi-polar disorder for the same set of symptoms

Research which investigates co-morbidity & symptom overlap

• Newson et al. (2021) - a meta-analysis of 107,349 that DSM-5 diagnostic criteria do not sufficiently distinguish between schizophrenic symptoms and those of co-morbid/overlapping disorders such as depression

  • Strengths of this study include the use of secondary data which means that the research process is swift and cost-effective

  • Limitations include a lack of insight into why symptom overlap and co-morbidity occur i.e. no explanatory power

• Buckley et al. (2009) found that schizophrenia has the following co-morbidities: 50% for depression; 47% for substance abuse disorder; 29% for PTSD; 23% for OCD; 15% for panic disorder

  • Strengths of this study include the support it provides for co-morbidity as a diagnosis issue

  • Limitations include the lack of clarity as to why depression has higher rates of co-morbidity than other disorders

Gender & culture bias in diagnosis

Gender bias

• Gender bias in diagnosis refers to any instances of a person being diagnosed according to their gender, rather than their symptoms

  • E.g. a female is not diagnosed with schizophrenia even when she presents with symptoms as the clinician may view her as a ‘hysterical female’ rather than taking her symptoms seriously (an example of alpha bias)

• Gender bias may also occur when a clinician does not take a patient’s gender into account when making a diagnosis e.g. male models of health are used to diagnose a woman

• Clinicians may also not pay enough attention to the fact that the risk factors for developing schizophrenia are different for male and females so there should be no ‘one size fits all’ approach (an example of beta bias)

Culture bias

• Culture bias in diagnosis refers to any instances of a person being diagnosed according to their culture, rather than their symptoms

  • E.g. an African patient is not diagnosed with schizophrenia by a British clinician, as the clinician may misunderstand or disregard their culture-specific symptoms

    • e.g. ukuthwasa - this has some symptom overlap with schizophrenia, but the clinician does not know what to diagnose and so may end up ignoring all of the patient’s symptoms

• Culture bias may also occur when a clinician does not take a patient’s culture into account when making a diagnosis

  • E.g. if the Western biomedical model is used to diagnose people from non-Western cultures it may (and does) result in African patients being over-diagnosed with schizophrenia due to a lack of understanding of culture-bound syndromes e.g. belief in witchcraft

Research which investigates gender & culture bias in diagnosis

• Hambrecht et al. (1993) - found that males and females are equally at risk of developing schizophrenia but schizophrenia is under-diagnosed in women 

  • Strengths of this study include the representative nature of the sample, addressing issues such as alpha bias and beta bias in diagnosis

  • Limitations include explanations as to why schizophrenia is under-diagnosed in women and what can be done to address this

• Schwartz et al. (2019) found evidence that African Americans are 2.4 times more likely to be diagnosed with schizophrenia compared with non-African American individuals, which points to culture bias in the diagnostic process and cultural insensitivity on the part of clinicians

  • Strengths of this study include good application i.e. the findings of this research could be used to inform diagnosis going forward

  • Limitations include the use of rating scales to assess symptoms i.e. the fine detail of the symptoms is lost when having to apply a standardised measure to them

Link to Issues & Debates:

Mental illness is a socially sensitive topic so researchers must be particularly mindful of how their research will be interpreted and commented on once it is published. Gender and culture are ‘hot topics’ in the media generally so the ways in which research is conducted and presented must not be seen to perpetuate stereotypes or to encourage prejudice and discrimination of people based on their gender, sexuality, culture or ethnicity.