Development of a self-report format of ASSIST with university students

• Self-report ASSIST is as acceptable as interview version in university students.
• Self-report ASSIST has good consistency, sensitivity and specificity for substances.
• University students preferred the interview format of ASSIST.

The Alcohol, Smoking and Substance Involvement Screening Test (ASSIST) is a reliable and valid tool to early detection in the harmful and hazardous drug use in primary care settings when administered by interview in the general population. As the risk of substance related problems in university students is high, it is necessary to have screening instruments that can be used beyond the health care settings. Thus, we compared a self-report adaptation of ASSIST with the validated interview format in a convenience sample of university students. A counter-balance design was chosen with students alternating between the interview and the self-report formats. Both formats were completed by all students (n = 170) over 30 days. The scores for total involvement, tobacco, alcohol, cannabis and cocaine obtained from the two formats demonstrated good intra-class correlation coefficient (ICC > 0.60). The agreement assessed by kappa between questions of the two formats was considered moderate for tobacco (0.76) and cannabis (0.69) and discrete for alcohol (0.47). The consistency of the self-report questionnaire was also good to moderate (Cronbach's alpha of 0.90 for tobacco, 0.71 for alcohol, 0.86 for cannabis and 0.89 for cocaine) and showed acceptable sensitivity (66.7–100%) and specificity (83.5–97.1%) for tobacco, alcohol, cannabis and cocaine when compared to the ASSIST interview format (gold standard). The findings suggest that self-report version is as acceptable as the interview and that the scores on the two formats are comparable. However, the participants reported more motivation for change behavior and more concern about substance use when they were interviewed.