Publication Date

2008-01-01

Availability

Open access

Degree Type

Thesis

Degree Name

Master of Science (MS)

Department

Psychology (Arts and Sciences)

Date of Defense

2008-01-10

First Committee Member

Sheri Johnson - Committee Chair

Second Committee Member

Malcolm Kahn - Committee Member

Third Committee Member

Charles Carver - Committee Member

Fourth Committee Member

Eric Youngstrom - Outside Committee Member

Abstract

Background: Bipolar disorders represent a serious mental health problem, but clinicians often fail to detect bipolar diagnoses. The validation of brief and accurate self-report questionnaires may aid in the detection of bipolar disorder, leading to more appropriate treatment and faster recovery. Many such measures exist, but few have been thoroughly tested in undergraduates. Methods: Three self-report questionnaires used to detect bipolar disorder (the Hypomanic Personality Scale[HPS], Mood Disorder Questionnaire [MDQ], and General Behavior Inventory ? 15 item version[GBI-15]) were administered to undergraduate psychology students during the first week of the semester. Participants who were selected based on high and low scores on the self-report screeners completed the Structured Clinical Interview for the DSM-IV, an instrument for diagnosing mental disorders. Participants also completed a battery of self-report measures for constructs previously found to be related to bipolar disorder. Results: Receiver Operating Characteristic (ROC) curves, sensitivity and specificity, and positive and negative predictive values were used to investigate usefulness of the three screeners in predicting SCID diagnoses of bipolar spectrum disorders. The three screeners did not demonstrate very good sensitivity or area under the curve for detecting a bipolar spectrum diagnosis, and they generally demonstrated low to moderate predictive values. Of the three, the GBI-15 performed the most adequately in this sample (positive predictive value of approximately .33). All three screeners demonstrated adequate negative predictive values (between .88 and .92). Discussion: The GBI-15 has some unique features that may help explain its outperformance of the other screeners in undergraduates, but suggestions are provided for the development of better screening tools.

Keywords

Assessment

Share

COinS