Archives
Computer test offers hope for diagnosis, treatment of personality disorders
By PATRICIA DONOVAN
Contributing Editor
A new version of a reliable and well-regarded dimensional test for personality disorders developed by a UB researcher and clinician may lead to clearer diagnosis of personality disorders and point toward more precise and specific treatment plans for the more than 31 million Americans affected by them.
Developed by Leonard Simms, assistant professor and researcher in the Department of Psychology, College of Arts and Sciences, the test is a computerized adaptive version of the Schedule for Nonadaptive and Adaptive Personality (SNAP), a 375-item schedule developed in 1993 by Lee Anna Clark of the University of Iowa.
Simms' version, administered adaptively by a computer, is called the SNAP-CAT.
Both versions of SNAP assess 15 personality traits thought to be important descriptors of individuals with personality disorders, such as manipulativeness, impulsivity and an accentuated sense of entitlement.
The traits then are graphed to demonstrate their relationship to one another and to the statistical norm. Using the graph, clinicians, researchers and impaired persons can "see" the extent to which such traits influence their personalities, often producing social and emotional problems for themselves and those with whom they live and work.
"They can be clinically treated," says Simms, "but they are ego-syntonic, meaning they usually are consistent with self-image. Those with such disorders often deny anything is wrong with them and blame others for their problems. As a result, they may not seek treatment.
"When impaired individuals are able to see the extent to which their behavior deviates from the norm, they are more likely to acknowledge their personal contribution to the problems in their lives," Simms says.
"The original SNAP has been used extensively and effectively in research settings and has shown good evidence of reliability and validity," Simms says, "but its clinical use has been restricted by the fact that it takes a long time to administer, score and interpret. Health insurers often refuse to pay for long psychological assessments."
The SNAP-CAT was developed by Simms with the hope that clinicians and researchers could obtain the same information from clients, but with far fewer questionnaire items. Its computerized adaptive format—borrowed from ability tests such as the Graduate Record Exam—will make it possible to give the test in a clinical setting in a relatively short period of time, which makes it much more useful to psychologists, psychiatrists and patients themselves, according to Simms.
He explains that with computerized adaptive testing, items are tailored individually to each person taking the test.
"As the test goes on, the computer continually estimates the test score and identifies new items that are most appropriate, given the current score estimate," says Simms.
"For example, on an adaptive test of aggression, individuals endorsing items such as 'I get into lots of fist fights' would not need to be administered items such as 'I sometimes get angry,' since the response to the second item can reasonably be inferred from an affirmative response to the first item, which is 'higher' on the trait of aggression."
In a recent article in the journal Psychological Assessment (vol. 17), Simms and Clark described the development and validation studies for the new test version. They concluded that the prototype SNAP-CAT is a largely comparable form of the original paper-and-pencil SNAP that yields reasonably comparable information, but with approximately 35 percent fewer items and in 60 percent less time.
The original SNAP is published by the University of Minnesota Press. The SNAP-CAT prototype developed by Simms currently is being updated and eventually may be published for professional use.
A 2004 National Institutes of Health study, reported in the Journal of Clinical Psychiatry, concluded that more than 15 percent of the adult American population—or more than 31 million individuals—are afflicted with personality disorders and that an additional 30 million or more Americans suffer negative consequences of living or working with these disordered personalities.
The most recent edition of the Diagnostic and Statistical Manual of Mental Disorders (DSM-IV), published by the American Psychiatric Association, includes 10 primary personality disorders, including antisocial, borderline, narcissistic, schizoid and avoidant personality disorders.
"A commonly recognized problem with the DSM categories is that they have a high rate of diagnostic comorbidity," Simms points out, "meaning that individuals are commonly diagnosed as having more than one personality disorder at once, which raises concerns about the validity of the current diagnostic system.
"In addition, patients in individual DSM personality-disorder categories tend to demonstrate a high level of trait heterogeneity, meaning symptoms and their severity vary significantly from one patient to the next, rendering the categories themselves imprecise as descriptors and diagnostic tools.
"The DSM categories are so broad as to suggest no consistent plan of treatment," he says, "and suggest an arbitrary boundary between normal and abnormal personality. On the other hand, dimensional testing such as SNAP and SNAP-CAT can help explain diagnostic problems, such as comorbidity and heterogeneity, and facilitate better diagnosis and more precise treatment plans for these difficult disorders."