'Willingness to Take Again' Better Predicted Cannabis Abuse Potential Than 'Drug Liking' | Research

The takeaway

In a retrospective analysis of 89 daily cannabis users across four studies, subjective ratings of 'Willingness to Take Again' were a better predictor of actual cannabis self-administration than the FDA-recommended 'Drug Liking' measure.

Drug regulators, abuse potential researchers, and pharmaceutical companies developing cannabinoid products.

What the researchers found

Higher ratings of Willingness to Take Again (OR=1.04) were significantly associated with increased odds of self-administering active THC cannabis over a monetary alternative. Higher Stimulated ratings were associated with decreased odds of choosing placebo cannabis (OR=0.94). Drug Liking was not the most predictive measure, contrary to FDA recommendations.

Why it matters

The FDA uses Drug Liking as the primary measure in abuse potential assessments for new drugs. If this measure is not the best predictor of actual drug-taking behavior for cannabinoids, regulatory evaluations of novel cannabinoid products may need updating.

The numbers in context

89 daily users (71 male, 18 female). 4 studies, 2 sites. THC: 5.5-5.9%. Willingness to Take Again: OR 1.04 (significant). Stimulated (for placebo): OR 0.94 (significant). Drug Liking: not the top predictor. Money alternatives: $0.50 or $1.00.

How the study worked

Retrospective analysis combining data from 4 previous studies across 2 research sites. 89 daily cannabis users (71 male, 18 female) completed cannabis self-administration sessions with a money alternative ($0.50 or $1.00) and subjective effects questionnaires. THC concentrations were 5.5-5.9%. Generalized linear models tested which subjective ratings predicted self-administration.

What this study cannot tell us

Retrospective analysis of combined datasets with methodological differences between studies. Small sample with predominantly male participants. Daily cannabis users may not represent occasional users. Low monetary alternatives ($0.50-$1.00) may limit ecological validity. THC concentrations were similar across studies but administration procedures varied.

How to read the evidence

Retrospective pooled analysis with behavioral outcome validation, but combined datasets and predominantly male sample limit generalizability.

When this study was published

Published 2025.

The bigger picture

As the cannabinoid product market expands with novel compounds (delta-8, THCA, HHC), regulators need accurate tools to assess abuse potential. This study suggests the current FDA-recommended approach may miss the mark for cannabinoids, potentially leading to inaccurate risk assessments.

Questions still open

Whether Willingness to Take Again is a better predictor for non-cannabis drug abuse potential as well
How these findings should change FDA guidance for evaluating novel cannabinoid products

Common questions

Why does it matter how abuse potential is measured?

When new cannabinoid products seek FDA approval or regulatory classification, their abuse potential determines scheduling and marketing restrictions. If the primary measurement tool (Drug Liking) is not the best predictor of actual drug-taking, products may be incorrectly classified as more or less risky than they actually are.

What is the difference between Drug Liking and Willingness to Take Again?

Drug Liking captures the immediate pleasurable experience. Willingness to Take Again captures the overall assessment including both positive effects and negative effects. Someone might like a drug's effect but still not want to take it again due to side effects, making it a more comprehensive predictor.

Read the original research

Subjective drug-effect ratings predict cannabis self-administration in people who use cannabis daily.

The Journal of pharmacology and experimental therapeutics, 392(7), 103622

View on PubMed ↗Publisher’s paper ↗

Citation

Shellenberg, Thomas P; Strickland, Justin C; Regnier, Sean D; Tolbert, Preston T; Lake, Stephanie; Wesley, Michael J; Stoops, William W; Cooper, Ziva D; Haney, Margaret; Lile, Joshua A. (2025). Subjective drug-effect ratings predict cannabis self-administration in people who use cannabis daily.. The Journal of pharmacology and experimental therapeutics, 392(7), 103622. https://doi.org/10.1016/j.jpet.2025.103622