Machine learning models struggled to predict who would respond to cannabis use disorder treatment

Using data from a multi-site clinical trial, machine learning and traditional models achieved only modest accuracy in predicting cannabis use disorder treatment response, suggesting better predictors are needed.

Tomko, Rachel L et al.·Addiction (Abingdon·2023·Moderate Evidencesecondary-analysis
RTHC-04984Secondary AnalysisModerate Evidence2023RETHINKTHC RESEARCH DATABASErethinkthc.com/research

Quick Facts

Study Type
secondary-analysis
Evidence
Moderate Evidence
Sample
Not reported

What This Study Found

Both multivariable logistic regression and machine learning models (random forest, gradient boosting) had limited ability to classify CUD treatment responders versus non-responders. Prediction accuracy was modest, indicating that commonly measured variables do not strongly predict treatment response.

Key Numbers

Multi-site clinical trial data used. Multiple machine learning approaches tested (random forest, gradient boosting, logistic regression). All achieved modest classification accuracy for treatment response.

How They Did This

Secondary analysis of a National Drug Abuse Treatment Clinical Trials Network multi-site outpatient trial. Adult CUD patients were assessed with multivariable logistic regression and machine learning models (random forest, gradient boosting) to predict treatment response.

Why This Research Matters

CUD treatments have limited efficacy overall. If clinicians could identify who will respond to which treatment approach, they could personalize care and improve outcomes. This study shows current predictive tools fall short.

The Bigger Picture

Precision medicine for substance use disorders lags behind other fields. The inability to predict treatment response for CUD suggests either better biomarkers are needed or that treatment response depends on factors not typically measured in clinical trials.

What This Study Doesn't Tell Us

Secondary analysis limited to variables collected in the original trial. Treatment was a specific multi-component protocol that may not generalize. Machine learning models can overfit to training data. Modest sample sizes may limit model performance.

Questions This Raises

  • ?Would biological markers (genetics, neuroimaging) improve prediction accuracy?
  • ?Are there treatment-matching variables that differ from treatment-response variables?

Trust & Context

Key Stat:
All prediction models achieved only modest accuracy for CUD treatment response
Evidence Grade:
Secondary analysis of a well-conducted multi-site trial with appropriate statistical methods. Limited by available variables and inherent prediction difficulty.
Study Age:
Published 2023.
Original Title:
Who responds to a multi-component treatment for cannabis use disorder? Using multivariable and machine learning models to classify treatment responders and non-responders.
Published In:
Addiction (Abingdon, England), 118(10), 1965-1974 (2023)
Database ID:
RTHC-04984

Evidence Hierarchy

Meta-Analysis / Systematic Review
Randomized Controlled Trial
Cohort / Case-Control
Cross-Sectional / ObservationalSnapshot without intervening
This study
Case Report / Animal Study
What do these levels mean? →

Frequently Asked Questions

Can we predict who will benefit from cannabis use disorder treatment?

Not well, according to this study. Even with advanced machine learning techniques, prediction accuracy was modest. This suggests that the variables typically measured in clinical settings (demographics, use history, co-occurring conditions) do not capture the full picture of what makes someone respond to treatment.

Why is predicting treatment response important?

If clinicians could identify likely responders before treatment begins, they could match patients to the most effective interventions, allocate intensive resources to those who need them most, and spare others from ineffective treatments.

Read More on RethinkTHC

Cite This Study

RTHC-04984·https://rethinkthc.com/research/RTHC-04984

APA

Tomko, Rachel L; Wolf, Bethany J; McClure, Erin A; Carpenter, Matthew J; Magruder, Kathryn M; Squeglia, Lindsay M; Gray, Kevin M. (2023). Who responds to a multi-component treatment for cannabis use disorder? Using multivariable and machine learning models to classify treatment responders and non-responders.. Addiction (Abingdon, England), 118(10), 1965-1974. https://doi.org/10.1111/add.16226

MLA

Tomko, Rachel L, et al. "Who responds to a multi-component treatment for cannabis use disorder? Using multivariable and machine learning models to classify treatment responders and non-responders.." Addiction (Abingdon, 2023. https://doi.org/10.1111/add.16226

RethinkTHC

RethinkTHC Research Database. "Who responds to a multi-component treatment for cannabis use..." RTHC-04984. Retrieved from https://rethinkthc.com/research/tomko-2023-who-responds-to-a

Access the Original Study

Study data sourced from PubMed, a service of the U.S. National Library of Medicine, National Institutes of Health.

This study breakdown was produced by the RethinkTHC research team. We analyze and report published research findings without making health recommendations. All interpretations are based solely on the published abstract and study data.