Reliability and Validity in Schizophrenia Diagnosis

Schizophrenia is diagnosed by the DSM-V. 

Reliability - the extent to which the diagnosis of schizophrenia is consistent. 

A diagnosis is reliable, if, for a particular individual, different doctors give the same diagnosis consistently. 

A diagnosis is also reliable, if, given multiple individuals with the same symptoms, one doctor gives each patient the same diagnosis consistently. 

One strength of the schizophrenia diagnosis is that it has high reliability, as observed in some studies. 

Jakobsen et al tested the reliability of the ICD-10 classifcation system in diagnosing schizophrenia. One hundred Danish patients with a history of psychosis were assessed using operational criteria, and a concordance rate of 98% was obtained. 

This demonstrates the high reliability of the clinical diagnosis of schizophrenia using up-to-date classification. 

However, one limitation of diagnosis of schizophrenia is low reliability, as seen in some studies. 

In Chienaux et al's study, 2 psychiatrists independently diagnosed 100 patients using both DSM and ICD criteria. 

It was found that inter-rater reliability was poor between the psychiatrists. One psychiatrist diagnosed 26 with schizophrenia using DSM and 44 using ICD. The other psychiatrist diagnosed 13 with DSM and 24 with ICD. 

This inconsistency between mental health professionals, and the different classification systems they used shows that the reliability of schizophrenia diagnoses is low as schizophrenia was under or over reported depending on the manual used by the doctor. 

Validity - the extent to which the diagnosis and classification techniques measure what they are designed to measure. 

A diagnosis is valid, if the criteria used for diagnosis allows us to correctly identify individuals who have a particular illness and allows us to avoid incorrectly diagnosing people who do not have the illness. 

One limitation of schizophrenia diagnosis is that it has issues with validity. 

In Rosenhan's study, 8 healthy confederates faked symptoms of schizophrenia (e.g. hallucinations) in order to be admitted into hospitals. Rosenhan tested how long it took hospital doctors to detect that patients were actually mentally healthy.  

Upon arrival into hospital, the confederates acted normally, yet, it took doctors between 7 and 52 days to detect the diagnoses of schizophrenia were wrong, and to discharge them. 

Rosenhan's research demonstrates that psychiatric staff cannot confidently tell thedifference between an individual that is mentally healthy andan individual with a mental illness, calling into question the validity of schizophrenia diagnosis. As 'normal' behaviour shown by the confederates was misinterpreted as 'abnormal' to support the diagnoses of mental illness they arrived with, this study shows that the validity of psychiatric diagnoses of schizophrenia is low. 

When doctors diagnose schizophrenia, they unconsciously use social norms of their own culture to decide whether an individual's behaviour matches any of the symptoms of schizophrenia. 

This reliance on social norms can lead to overdiagnosis of schizophrenia in patients from a different culture to the doctor. 

One limitation of schizophrenia diagnosis is the existence of cultural bias. 

Cochrane conducted a review comparing the number of people diagnosed with schizophrenia in the West Indies and in the UK. 

The incidence of schizophrenia in the West Indies and in Britain was similar, at 1%, but Cochrane found that Afro-Caribbean people were 7 times more likely to be diagnosed with schizophrenia when living in Britain than when living in the West Indies. 

Higher diagnosis rates in the UK may be because some behaviours classed as positive symptoms of schizophrenia are 'normal' in Afro-Caribbean cultures (e.g. hearing voices as a part of ancestor communication). This research indicates that individuals from certain cultural backgrounds are over-diagnosed, due to an ethnocentric bias, thus reducing the validity of schizophrenia diagnosis.

Both types of bias (alpha and beta) can lead to the over-diagnosis of schizophrenia in men or in women. 

If the criteria used to diagnose schizophrenia over or under exaggerates the differences between men and women, then they won't correctly identify schizophrenia in all men and women, reducing validity. 

One limitation of schizophrenia diagnosis is that it may be affected by gender bias. 

Loring and Powell gave male and female psychiatrists identical descriptions of a patient's symptoms but varied the description of the patient's gender. 

They found that when the patient was described as male, 56% of doctors diagnosed the patient with schizophrenia, but when the patient was described as female, only 20% diagnosed the patient with schizophrenia. 

This research indicates that there may be unconscious alpha bias in the diagnoses of schizophrenia. As the gender of an individual can influence diagnoses, this is a problem that may lower the reliability of schizophrenia diagnosis. 

Co-morbidity - occurence of two illnesses in the same person which confuses diagnosis and treatment. 

Effects of co-morbidity

If a patient has a co-morbid illness, then doctors are more likely to ignore some of the symptoms, so different doctors may give the patient different diagnoses, reducing the reliability of the diagnoses. 

If two illnesses are often co-morbid, this suggests that the criteria for diagnosing the two illnesses, might be incorrect - they might not actually be 2 separate conditions, so the diagnosis might lack validity. 

One limitation of schizophrenia diagnosis is the issue of co-morbidity. 

Buckley et al investigated how many patients had a co-morbid mental disorder. 

He found that around schizophrenic patients also had a diagnosis of depression (50%), substance abuse (47%) and OCD (23%), demonstrating that schizophrenia occurs alongside many other illnesses and is co-morbid.

Where 2 conditions are frequently diagnosed together, validity of the classification of the 2 disorders is called into question, making it more difficuly to confidently diagnose 1 disorder, like schizophrenia. For example, if schizophrenia looks very similar to depression, they may or may not be a single condition. This confusing picture caused by co-morbidity can limit the validity of schizophrenia diagnosis. 

The symptoms of schizophrenia overlap with symptoms of other mental disorders (e.g. bipolar disorder and depression). 

This symptom overlap means that different doctors might diagnose an individual with a different disorder, reducing reliability of diagnoses. 

In addition, individuals might be diagnosed with the wrong disorder, reducing the validity of  diagnoses. 

One limitation of schizophrenia diagnosis is the problem of symptom overlap. 

Ellason and Ross compared the symptoms of patients diagnosed with dissociative identity disorder and those diagnosed with schizophrenia. 

They found that patients diagnosed with dissociative identity disorder displayed more schizophrenic symptoms than the patients diagnosed with schizophrenia themselves. 

This study shows that there is considerable overlap between the symptoms of schizophrenia and other conditions, which may become confusing. This lack of distinction calls into question the validity of both the classification and the diagnosis of schizophrenia as we are cannot be accurate in thinking that schizophrenia exists as a separate condition to other mental illlnesses. 

However, Serper assessed patients with co-morbid schizophrenia and cocaine abuse, cocaine intoxication on its own, and schizophrenia on its own. 

They found that despite there being considerable symptom overlap in patients with schizophrenia and cocaine abuse, it was actually possible to make accurate diagnoses, showing that symptom overlap did not affect the validity of the diagnosis and clinicians can tell the difference between the two illnesses. 

One strength of reliability in schizophrenia diagnosis is the use of inter-rater reliability by researchers and doctors. In 1962, Beck found that diagnoses were 52% similar. In 2005, researchers found that diagnoses were 81% similar. This suggests that with the increasing use of inter-rater reliability, the diagnoses of schizophrenia are become more reliable over time. 

A limitation in the diagnosis of schizophrenia is validity demonstrated by research support. Rosenhan got 8 healthy volunteers to fake the symptoms of schizophrenia in order to be admitted into hospital, and once there, were told to act normally. It took doctors between 7 and 52 days to realise the diagnoses were wrong and that the patients were actually mentally healthy. This field study indicates that the diagnosis of schizophrenia can lack validity, and so is a weakness. 

Another limitation of schizophrenia diagnosis is cultural bias. People from certain ethnic backgrounds are more likely to be misdiagnosed with schizoprenia. Cochrane conducted a review comparing the number of people diagnosed with schizophrenia in the UK and in the Caribbean. Afro-Caribbean people were 7x more likely to be diagnosed with schizophrenia when living in UK than when living in Caribbean. Cochrane concluded that this was because of cultural bias by British doctors, which highlights a lack  of reliability in schizophrenia diagnosis. 

A final limitation of schizophrenia diagnosis is symptom overlap in diagnostic criteria. The symptoms of schizophrenia overalp with other mental disorders such as bipolar disorder and depression. This symptom overlap means that different doctors may diagnose a patient with a different disorder, reducing reliability and with a wrong disorder, reducing validity as well.