Connect with us

Science

Could AI help predict the effects of cannabis on mental health?

Machine learning tools could help predict whether cannabis would be beneficial for someone experiencing mental health issues. 

Published

on

Machine learning software was used to develop a predictive model of mental health symptom change with cannabis use.

A new study highlights how machine learning tools could help predict whether cannabis would be beneficial for someone experiencing mental health issues. 

Cannabis is increasingly being reported as a helpful way to reduce symptoms of anxiety, insomnia and depression. However, as everyone responds to products differently, it is still something of a guessing game as to whether it will help ease someone’s symptoms or risk exacerbating them.

In a recent study, researchers explored whether machine learning methods, such as artificial intelligence (AI), could predict symptom change in mental health patients after cannabis use. 

The study used data from over 1,300 patients, aged between 18 and 71 years old, gathered through the app Strainprint, which allows people to track their experience and symptom changes with cannabis. The most common symptoms recorded were anxiety, insomnia and depression (over 50%), but observations also included irritability, stress, PTSD, PMS and intrusive thoughts. 

The app allows consumers to rate their symptom severity before and after treatment with cannabis, as well as inputting the concentration of the product they are using and mode of administration – such as whether it was flower, oil or an edible. 

After collecting the data, the researchers used the machine learning software XGBoost to develop a predictive model of mental health symptom change with cannabis use. This was defined as a ‘subjective difference in symptom severity between pre- and post-symptom ratings’.

Symptom severity, age and gender

Their findings revealed that the severity of symptoms before cannabis use was the ‘strongest predictor’ of symptom change, followed by age, gender, and the CBD/THC ratio consumed.

“Interestingly, these features are often among the least explored in cannabis literature,” the authors say. 

“We found that males and females had contrasting trends in symptom improvement that seemed to be dependent on baseline symptom severity.”

While the study investigated self-reported gender rather than sex, they believe the differences could be attributed to the different effects of cannabis on male and female hormones and on the ‘prevalence of depression, anxiety, and insomnia in women during reproductive years’.

They add: “These findings highlight the importance of analysing sex- and gender-dependent effects on symptom improvement with cannabis use in future studies.”

Predicting changes in symptoms

The researchers also note that while there was some level of improvement reported across all symptoms, the effects on each individual symptom varied.

Those who reported less severe symptoms of depression before cannabis use were predicted to report symptom improvement afterwards, while those who reported higher depression actually reported ‘worsening symptoms’ following treatment.  

When it came to insomnia, however, those reporting less severe symptoms were predicted to report ‘very little to no symptom improvement after cannabis use’, while those with more severe symptoms were predicted to benefit. 

The authors say: “The current study provides additional rationale for future studies investigating specific symptom profiles within diagnosed mental health disorders. Moreover, findings from this study provide valuable insight into current cannabis use patterns.”

Interpreting the findings ‘cautiously’

However, they acknowledge that the results should be ‘interpreted cautiously’ due to the limitations of the study. These include the fact that the symptoms were self-reported and the extent of data which could be collected through the app, including limited information on the cannabinoid content of products. 

Additional factors such as lifestyle, comorbidities, other medications or therapies, and frequency of use may also have influenced symptom change but were not recorded. 

“Self-perceived effectiveness alone cannot obviate the arguments against more widespread use,” the authors state.

“While our study does demonstrate some potential acute therapeutic benefits with cannabis use for mental health management, it is unclear how extended or frequent cannabis use may impact symptomatology in the long run.”

Calls for further research

This aside, the paper provides some key insights into how cannabis is being used and areas worthy of more investigation, as well as highlighting the role that advancements in technology could play in building the evidence base for the use of cannabis in treating mental health conditions.

The authors have called for further research to explore the long-term effects of cannabis use on mental health and address the findings in greater detail. 

“Notwithstanding its limitations, the current study is foundational in providing critical information on how various cannabis use-related factors may contribute to perceived symptom change in individuals managing mental health challenges in the real-world,” the researchers conclude.

“The current study highlights a need for additional research in this area, with findings revealing a central role for user profiles and baseline severity of perceived symptoms on outcomes of acute cannabis use for various mental health concerns.”

Home » Science » Could AI help predict the effects of cannabis on mental health?

Sarah Sinclair is a respected cannabis journalist writing on subjects related to science, medicine, research, health and wellness. She is managing editor of Cannabis Health, the UK’s leading title covering medical cannabis and CBD, and sister title and Psychedelic Health. Sarah has an NCTJ journalism qualification and an MA in Journalism from the University of Sunderland. Sarah has over six years experience working on newspapers, magazines and digital-first titles, the last two of which have been in the cannabis sector. She has also completed training through the Medical Cannabis Clinicians Society securing a certificate in Medical Cannabis Explained. She is a member of PLEA’s (Patient-Led Engagement for Access) advisory board, has hosted several webinars on cannabis and women's health and has moderated at industry events such as Cannabis Europa. Sarah Sinclair is the editor of Cannabis Health. Got a story? Email sarah@prohibitionpartners.com / Follow us on Twitter: @CannabisHNews / Instagram: @cannabishealthmag

Trending

Cannabis Health is a journalist-led news site. Any views expressed by interviewees or commentators do not reflect our own. All content on this site is intended for educational purposes, please seek professional medical advice if you are concerned about any of the issues raised.

Copyright © 2024 PP Intelligence Ltd.