As in other areas of psychiatric research, the reliability and validity of diagnosis has been a central issue in research on personality disorders. Also, as in other areas of psychopathology, there is no ''gold standard'' that can be used to validate diagnoses. However, the problem seems to be greater for personality disorders than for axis I disorders. There are a number of issues that probably contribute to the difficulty in reliably and validly diagnosing personality disorders. One important factor is that axis II symptomatology is often less florid and dramatic than axis I symptomatology.
Skodol et al. (1990) identified a number of issues that contribute to the difficulty in assessing personality disorders. In comparison to axis I disorders, personality disorders are more likely to be ego syntonic. This means that to the individual with a personality disorder, his or her symptomatology is not experienced as alien to his or her usual experience of self. The story is often different for axis I disorders. For example, a panic attack is typically experienced as being distinctively different from normal experience. The nature of personality disorders may make it more difficult for the individual to describe symptoms to an interviewer. The symptoms of personality disorder may be more recognizable and troublesome to individuals in the environment of the person with the disorder than to the person him or herself. Clinicians may be inclined to rely more on their own observations than on the patient's reports when assessing the presence of personality disorder. In the type of assessment of personality disorder typically used in research, the respondent is asked a series of questions about symptoms and answers are generally taken to be veridical. There is usually some provision for the diagnostician to override the subject's self-report if there is contradictory information. Given that personality disorders are defined as long-standing stable characteristics, one time, cross-sectional assessment may not be ideal. If the person has a concurrent axis I disorder, the symptoms of that disorder may influence the report of axis II symptoms.
Zanarini and her colleagues (1987) also identified a number of issues that contribute to the difficulty of making reliable and valid diagnoses of personality disorders. Many of the DSM diagnostic criteria are not clearly operationalized, necessitating clinical interpretation. Some symptoms, such as low self-esteem, have a very high base rate and may require significant clinical judgment as to whether they achieve clinical significance. Many diagnostic criteria (e.g., vanity or criminal behavior) reflect traits that are generally held to be negative, and may be denied on the basis of social desirability. Reporting certain traits may require a level of insight that is absent in some individuals.
Another central issue in diagnosing personality disorders is the occurrence of certain spectrum relationships that exist between personality disorders and axis I disorders, which are thought to represent phenotypic variations of the same underlying pathology. Such relationships have been suggested to exist between borderline personality disorder and depression, depressive personality disorder and depression, schizotypal personality disorder and schizophrenia, avoidant personality disorder and social phobia, cluster B personality disorders and substance use, cluster B and C personality disorders and eating disorders, cluster C personality disorders and anxiety disorders and cluster A and schizophrenia (Tyrer et al., 1997). There is evidence that the co-occurrence of personality disorders with axis I disorders predicts worse outcome than an axis I disorder alone (Reich and Green, 1991) and that personality disorders may impair subsequent axis I treatment response. However, acknowledgement of the relationship in devising treatment options could influence the delivery and type of treatment and therefore possible subsequent success (Oldham, 1994).
Was this article helpful?