Defense Date

2011

Document Type

Dissertation

Degree Name

Doctor of Philosophy

Department

Psychology

First Advisor

Dace Svikis

Abstract

OBJECTIVE: Most drug use screening measures rely on and are validated against self-report. Fear of negative consequences often promotes denial of drug use. For pregnant women, social stigma and fear of legal consequences make underreporting of drug use even more likely. An indirect screener that could effectively identify pregnant women at risk for illicit drug use without reliance on disclosure would be clinically significant. The purpose of the current study was to develop and validate an indirect measure of prenatal drug use by comparing correlates of prenatal drug use to urinalysis results. METHOD: Pregnant women attending an OB appointment at the VCUHS Women’s Health Clinic were recruited and consented to participate in an anonymous, two-phase study. In Phase 1, women completed a 20-minute computerized assessment which included a true/false index of items known to tap behavioral, medical, psychological, experiential and demographic correlates of drug abuse and dependence. In Phase 2, participants were asked to provide a urine sample for drug testing. Women received a $20 gift card after they participated in each phase. RESULTS: Two hundred and thirty-one women completed both Phase 1 and 2 (94% completion rate). Participants were primarily African-American (66%), single (75%) and receiving public assistance (70%). Urinalysis revealed that 16% of the sample tested positive for recent drug use, while only 5% of women self-reported past month drug use. After examining the univariate and multivariate relationships between each indirect item and drug status (i.e., positive or negative urinalysis), six indirect items were chosen to comprise the Wayne Indirect Drug Use Screener-Pregnancy (WIDUS-P). Cross-validation analyses resulted in a sensitivity of .90, specificity of .75, and AUC of .85. In comparison to direct screening approaches, the WIDUS-P was superior in identifying pregnant women who had used drugs recently. CONCLUSIONS: Findings support the use of an indirect screening tool to identify prenatal drug use, especially over currently-used direct methods. Such a measure could easily be implemented into regular clinic practice and result in more cost-effective and better identification of prenatal drug use.

Rights

© The Author

Is Part Of

VCU University Archives

Is Part Of

VCU Theses and Dissertations

Date of Submission

May 2012

Included in

Psychology Commons

Share

COinS