The Alcohol, Smoking and Substance Involvement Screening Test (ASSIST) is an instrument to screen substance -use-related health risks. However, little is known whether the ASSIST could be further shortened while remaining psychometrically sound across different countries, languages, gender identities, and sexual-orientation-based groups.
The study aimed to validate a shortened 11-item ASSIST (ASSIST-11). Using the International Sex Sur-vey data, 82,243 participants (M age = 32.39 years) across 42 countries and 26 languages completed questions from the ASSIST-11 regarding gender identity, sexual orientation, and other information.
Confirmatory factor analysis (CFA) and multigroup CFA (MGCFA) evaluated the ASSIST-11's structure and tested measurement invariance across groups. Cronbach's & alpha; and McDonald's & omega; were used to examine the internal consistency.
Cohen's d and independent t-tests were used to examine known-group validity. The ASSIST-11 was unidimen-sional across countries, languages, age groups, gender identities (i.e., men, women, and gender-diverse in-dividuals), and sexual orientations (i.e., heterosexual and sexual minority individuals).
Cronbach's & alpha; was 0.63 and McDonald's & omega; was 0.68 for the ASSIST-11. Known-group validity was supported by Cohen's d (range be-tween 0.23 and 0.40) with significant differences (p-values<0.001).
The ASSIST-11 is a modified instrument with a unidimensional factor structure across different languages, age groups, countries, gender identities, and sexual orientations. The low internal consistency of the ASSIST-11 might be acceptable as it assesses a broad concept (i. e., use of several different substances).
Healthcare providers and researchers may use the ASSIST-11 to quickly assess substance-use information from general populations and evaluate the need to follow up with more detailed questions about substance use.