Identifying Predictors of Benzodiazepine Discontinuation in Medical Cannabis Patients with Post-traumatic Stress Disorder Using a Machine Learning Approach

Read the full article See related articles

Listed in

This article is not in any list yet, why not save it to one of your lists.
Log in to save this article

Abstract

Introduction

Post-Traumatic Stress Disorder (PTSD) is a debilitating mental health condition commonly treated with medications like benzodiazepines (BZDs), despite their potential for negative long-term side effects. Medical cannabis has emerged as a possible adjunctive therapy for PTSD. However, the relationship between medical cannabis use, relief from PTSD symptoms, and the use of BZDs remains unclear. Thus, we sought to identify predictors of changes to BZD usage among medical cannabis patients with PTSD.

Methods

This study utilized survey data from PTSD patients in the Leafwell patient database, collected in the fall of 2023. To assess the relationship between medical cannabis use, PTSD symptom relief, and the discontinuation of BZDs, we employed a multi-step analysis approach. First, we developed a decision tree model to identify key predictors of BZD discontinuation, including prior cannabis use and reported PTSD relief post-medical cannabis treatment initiation. The tree was pruned using the optimal complexity parameter to improve model interpretability. Following this, a secondary logistic regression analysis was performed to confirm the significance of key predictors identified by the decision tree.

Results

In the pruned decision tree, not currently receiving psychiatric care for their PTSD was the strongest predictor of BZD discontinuation, followed by self-reported efficacy of medical cannabis in relieving PTSD symptoms, prior cannabis use, and history of traumatic brain injury (TBI) among medical cannabis patients. Age was also a significant factor, with younger individuals more likely to discontinue. Logistic regression analysis supported these findings, with receiving care, TBI, and cannabis use remaining key predictors. Interaction models suggest prior cannabis use moderates the relationship between those not receiving psychiatric care and BZD discontinuation odds.

Discussion

These findings suggest that medical cannabis may offer a promising route for BZD discontinuation for long-term users with PTSD symptoms. The association between cannabis use and BZD discontinuation highlights the need for further exploration of cannabis as an adjunctive therapy in PTSD care. More research is necessary to confirm the long-term safety and effectiveness of medical cannabis in this context, ensuring that it can be integrated into care without unintended negative consequences. Individualized care approaches remain crucial given varying patient factors.

Article activity feed