Investigation of the Alcohol, Smoking, and Substance Involvement Screening Test (the ASSIST) Version 3.0 in Pregnancy

Abstract

Objectives: To investigate the utility of the World Health Organization’s Alcohol, Smoking and Substance Involvement Screening Test (ASSIST) Version (V) 3.0 for identifying substance use disorders in pregnancy. Methods: ASSIST V3.0’s performance was assessed, focussed on nicotine, alcohol, and cannabis, using a 2-pronged harm categorization: risk to the woman as an individual and risk to the fetus. For the former, risk levels concordant with general population cut-points were utilized. A total of 104 substance users in an Australian public maternity hospital were identified by case-note audit. The ASSIST V3.0 scores were appraised against the scores for established tools—tobacco: the Revised Fagerstrom Tolerance Questionnaire, alcohol: the tolerance, annoyed, cut down, eye-opener, and cannabis: the Timeline FollowBack. Results: Kappa analyses of Specific Substance Involvement Scores for ASSIST V3.0 did not support changing cut-points for the woman as an individual; however, receiver operating characteristics curves delineated an Specific Substance Involvement Scores of 4 as indicative of fetal risk for both alcohol and cannabis. All nicotine users (98 of the 104 participants) were “at high risk”; hence a cut-point indicative of fetal risk for nicotine could not be determined. Conclusions: The role of the ASSIST V3.0 is uncertain for this population. Given the predominance of nicotine use, nicotine use screening could be the primary focus, with follow-up screening for alcohol and other substances if tobacco use were identified; there may be a place for a restructured ASSIST in that context. Positive screening results should be followed by ongoing counseling support throughout pregnancy, with the intensity dictated by the severity of use.