Rethinking cannabis risk screening for medical users

Examining the Alcohol, Smoking and Substance Involvement Screening Test (ASSIST) Moderate Risk Cannabis Use Threshold Among a Sample of U.S. Military Veterans with Chronic Pain Who Report Medical Cannabis Use.

Substance use & misuse β€’ β€’ Highly Relevant
πŸ€–

AI Summary

This study examined whether the standard ASSIST screening threshold accurately identifies moderate-risk cannabis use in 501 U.S. military veterans with chronic pain who used cannabis for medical reasons while receiving long-term opioid therapy. Researchers compared ASSIST scores with the Cannabis Use Disorder Identification Test–Revised (CUDIT-R), a cannabis-focused screening measure.

The currently recommended ASSIST cutoff of 4 performed poorly in this group. A cutoff of 7 produced the best overall discrimination, while a cutoff of 6 identified more people who may be at risk. The authors suggest using 7 for ongoing monitoring and 6 for initial screening, but emphasize that the ASSIST still has important limitations for detecting cannabis-related harms in medical cannabis users. Population-specific screening tools are needed.

πŸ’‘ Key Findings

1
The standard ASSIST cannabis-risk cutoff of 4 showed poor performance among veterans using cannabis medically for chronic pain.
High
85%
2
An ASSIST cutoff of 7 maximized overall performance for identifying moderate cannabis risk, but it had low sensitivity.
High
80%
3
A cutoff of 6 substantially improved sensitivity and may be more suitable for initial screening, while 7 may be better for ongoing monitoring.
High
80%
4
The study concludes that the ASSIST has significant limitations in this population and that population-specific cannabis screening tools are needed.
High
90%

πŸ“„ Original Abstract

The Alcohol, Smoking and Substance Involvement Screening Test (ASSIST) is widely used for substance use screening. However, its established risk cutoffs may not be appropriate for people who use cannabis for medical reasons. This study sought to determine an optimal moderate-risk cannabis use cutoff for this population. A sample of 501 U.S. military veterans who reported using cannabis for medical reasons and were prescribed long-term opioid therapy were administered the ASSIST and Cannabis Use Disorder Identification Test - Revised (CUDIT-R). Receiver operating characteristic (ROC) curves compared the performance of the ASSIST to the CUDIT-R. The currently recommended cutoff of 4 demonstrated poor performance (Area Under the Curve [AUC] = 52.8). A cutoff of 7 maximized AUC (66.2). However, this cutoff resulted in low sensitivity (59.9). A cutoff of 6 improved sensitivity (94.7), with a reduced AUC of 59.8. Among this population, raising the ASSIST cutoff to 7 to detect moderate cannabis risk may be appropriate for ongoing patient monitoring, while a cutoff of 6 may be preferable for screening. However, there remain significant problems with the ASSIST as a measure for detecting risk for cannabis-related harms among this population. Development of population-specific screening tools is needed.

Explore More Research

Stay informed about the latest cannabis science.

Your stash, decoded.