Date of Award

1-1-2014

Document Type

Dissertation

Degree Name

Ph.D.

Organizational Unit

Morgridge College of Education

First Advisor

Kathy Green, Ph.D.

Second Advisor

Jeff Jensen

Third Advisor

Antonio Olmos

Fourth Advisor

Cynthia McRae

Keywords

Alcohol use, Alcohol Use Disorder Identification Test, AUDIT-C, Depression, Impulsivity, Patient health questionnaire, Rasch analysis

Abstract

This study examined the psychometric characteristics of a brief assessment measure for screening for depression and alcohol use on a college campus prior to primary care medical office visits. The measure was adapted from two widely used measures: the Patient Health Questionnaire (PHQ-4) and the Alcohol Use Disorder Identification Test (AUDIT-C). Impulsivity, which has been associated with both depression and problematic alcohol use, was also examined through additional questions. The research study investigated the psychometric properties of the PHQ-4 and the AUDIT-C, and explored if eight impulsivity items from the UPPS-P measure could enhance screening for depression and problematic alcohol consumption.

A 15-item measure was piloted with 491 college-aged individuals. The measure was examined using several analytic techniques. Exploratory factor analysis identified three factors indicating the measure contained depression (PHQ-4), alcohol use (AUDIT-C), and impulsivity factors. Rasch analysis resulted in identifying 15-item measure as multidimensional. Further Rasch analysis showed the PHQ-4, the AUDIT-C, and the impulsivity questions as unidimensional. The PHQ-4 measure showed adequate fit, scale use, and targeting for this population. Rasch analysis resulted in four-items from the eight impulsivity questions that could be treated as a scale. However, the Rasch analysis of AUDIT-C showed poor item fit and significant differential item functioning and was determined to inadequate as a scale, and so, individual items were used in subsequent analyses.

Hierarchical regression revealed a significant contribution of the impulsivity measure in explaining variance for the PHQ-4, but was lacking in explaining additional measure variance when used with the AUDIT-C individual items. Latent class analysis identified three classes, with the most interesting being male, young, and white that frequently binge drinks regularly (22% of the population).

While the 15-item scale was unsuccessful in improving identification of problematic drinking, the impulsivity items could be useful in helping to better identify depression among this population. The results also questioned the effectiveness of the AUDIT-C in screening for excessive alcohol consumption.

Further research should focus on the development of better brief screening tools in primary practice that are psychometrically sound and contain items that are not only diagnostic in nature. Inclusion of items in these instruments that explore related facets, such as impulsivity, should be explored in future development.

Publication Statement

Copyright is held by the author. User is responsible for all copyright compliance.

Rights Holder

Chris Wera

Provenance

Received from ProQuest

File Format

application/pdf

Language

en

File Size

149 p.

Discipline

Mental health, Quantitative psychology and psychometrics



Share

COinS