Psychometric validation of the Leeds Dependence Questionnaire (LDQ) in a young adult clinical sample

• Published in: Addictive behaviors 2010-04, Vol.35 (4), p.331-336
• Main Authors: Kelly, John F, Magill, Molly, Slaymaker, Valerie, Kahler, Christopher
• Format: Article
• Language: English
• Subjects: Addictive behaviors; Adolescent; Adult; Assessment; Behavior, Addictive - diagnosis; Behavior, Addictive - psychology; Discriminant analysis; Female; Humans; Male; Psychiatry; Psychological tests; Psychometric; Psychometrics; Reproducibility of Results; Substance abuse or dependence severity; Substance-Related Disorders - diagnosis; Substance-Related Disorders - psychology; Surveys and Questionnaires - standards; United States; Validation studies; Young Adult; Young adults
• ISSN: 0306-4603; 1873-6327
• DOI: 10.1016/j.addbeh.2009.11.005

Abstract Objective Measures of substance dependence severity that are both clinically efficient and sensitive to change can facilitate assessment of clinical innovation necessary for improving current evidence-based practices. The Leeds Dependence Questionnaire (LDQ) is a 10-item, continuous, self-report measure of dependence that is not specific to any particular substance and has shown promise in preliminary psychometric research. The present study investigates its psychometric properties in a large clinical sample of young adults. Method Young adults ( N = 300) were enrolled in a naturalistic treatment process and outcome study of residential substance dependence treatment (mean age 20.4 [1.6], range 18–25; 27% female; 95% White). Dependence severity by demographic and diagnostic groupings, factor structure and internal consistency, and criterion- and construct-related validity were examined. Results Dependence severity in this cohort of youth overall was high ( M = 18.65 [8.65]). LDQ scores were highest among opiate and stimulant users, and there was a trend for higher scores among women compared to men ( t = 1.869, p = .063). Factor analysis using a robust alpha factoring extraction revealed a single factor accounting for 63% of the variance in reported dependence severity. The internal consistency was also very high (alpha = .93). Concurrent and convergent validity with dependence criteria, substance use frequency, and general symptom severity, respectively, were also acceptable. Conclusions The LDQ shows considerable promise as a brief, psychometrically sound, measure of substance dependence useful across a variety of substances, that has clinical and research utility. This study supports its use among young adults.