• Skip to primary navigation
  • Skip to main content
  • Skip to primary sidebar

NIDA IRP

National Institute on Drug Abuse - Intramural Research Program

  National Institute on Drug Abuse | NIH IRP | Treatment Info | Emergency Contacts
  • Home
  • News
    • Featured Paper of the Month
    • Reviews to Read
    • Hot off the Press
    • IRP News
    • Awards
    • Technology Development Initiative Paper of the Month
    • Seminar Series
    • Addiction Grand Rounds
  • About
    • About NIDA IRP
    • Contact Us
    • Directions and Map
    • Careers at NIDA IRP
    • Emergency Contacts
    • Employee Assistance Resources
  • Organization
    • Faculty
    • Office of the Scientific Director
    • Office of the Clinical Director
    • Office of Education and Career Development
    • Administrative Management Branch
    • Molecular Targets and Medications Discovery Branch
    • Cellular and Neurocomputational Systems Branch
    • Molecular Neuropsychiatry Research Branch
    • Neuroimaging Research Branch
    • Behavioral Neuroscience Research Branch
    • Integrative Neuroscience Research Branch
    • Translational Addiction Medicine Branch
    • Core Facilities
    • Community Outreach Group
  • Training Programs
    • Office of Education and Career Development
    • OECD Awards
    • Summer Internship Program
    • Postbaccalaureate Program
    • Graduate Partnership Program
    • Postdoctoral Program
    • NIDA Speakers Bureau
    • Clinical Electives Program
    • Clinical Mentoring Program
  • Study Volunteers

AI-based analysis of social media language predicts addiction treatment dropout at 90 days

A figure from this study

Hot Off the Press – May 23, 2023

Published in Neuropsychopharmacology by Brenda Curtis and Salvatore Giorgi of the NIDA IRP Technology and Translational Research Unit in collaboration with other researchers.

Summary

Assessing an individual’s risk of substance use treatment dropout traditionally involves comprehensive, structured interviews which are time consuming and costly to administer. More accurate risk assessment would allow clinicians and treatment facilities to better allocate support resources to those in higher need. The current study suggests that digital phenotyping technologies (i.e., digital proxies of behavior and emotional states) could be used in tandem with standard clinical measures to produce more accurate risk measures for those entering treatment for substance use. The study uses state-of-the-art artificial intelligence and natural language processes methods to create digital phenotypes from participants’ past social media language. Clinical data and the language-based digital phenotypes, both collected at treatment intake, are then used to generate risk scores for each participant in order to predict future treatment adherence. Of the participants who were assigned low-risk scores at intake, nearly 80% remained in treatment after 90 days, whereas 20% of the high-risk participants remained. These results suggest that digital phenotypes are well equipped to capture heterogenous experiences which are predictive of substance use treatment dropout. Further, they can be used as a low cost, unobtrusive tool for clinicians to assess future risk of treatment dropout and allocate support resources to those with higher risk.

Publication Information

Curtis, Brenda; Giorgi, Salvatore; Ungar, Lyle; Vu, Huy; Yaden, David; Liu, Tingting; Yadeta, Kenna; Schwartz, H Andrew

AI-based analysis of social media language predicts addiction treatment dropout at 90 days Journal Article

In: Neuropsychopharmacology, 2023, ISSN: 1740-634X.

Abstract | Links

@article{pmid37095253,
title = {AI-based analysis of social media language predicts addiction treatment dropout at 90 days},
author = {Brenda Curtis and Salvatore Giorgi and Lyle Ungar and Huy Vu and David Yaden and Tingting Liu and Kenna Yadeta and H Andrew Schwartz},
url = {https://pubmed.ncbi.nlm.nih.gov/37095253/},
doi = {10.1038/s41386-023-01585-5},
issn = {1740-634X},
year = {2023},
date = {2023-04-01},
urldate = {2023-04-01},
journal = {Neuropsychopharmacology},
abstract = {The reoccurrence of use (relapse) and treatment dropout is frequently observed in substance use disorder (SUD) treatment. In the current paper, we evaluated the predictive capability of an AI-based digital phenotype using the social media language of patients receiving treatment for substance use disorders (N = 269). We found that language phenotypes outperformed a standard intake psychometric assessment scale when predicting patients' 90-day treatment outcomes. We also use a modern deep learning-based AI model, Bidirectional Encoder Representations from Transformers (BERT) to generate risk scores using pre-treatment digital phenotype and intake clinic data to predict dropout probabilities. Nearly all individuals labeled as low-risk remained in treatment while those identified as high-risk dropped out (risk score for dropout AUC = 0.81; p < 0.001). The current study suggests the possibility of utilizing social media digital phenotypes as a new tool for intake risk assessment to identify individuals most at risk of treatment dropout and relapse.},
keywords = {},
pubstate = {published},
tppubtype = {article}
}

Close

The reoccurrence of use (relapse) and treatment dropout is frequently observed in substance use disorder (SUD) treatment. In the current paper, we evaluated the predictive capability of an AI-based digital phenotype using the social media language of patients receiving treatment for substance use disorders (N = 269). We found that language phenotypes outperformed a standard intake psychometric assessment scale when predicting patients' 90-day treatment outcomes. We also use a modern deep learning-based AI model, Bidirectional Encoder Representations from Transformers (BERT) to generate risk scores using pre-treatment digital phenotype and intake clinic data to predict dropout probabilities. Nearly all individuals labeled as low-risk remained in treatment while those identified as high-risk dropped out (risk score for dropout AUC = 0.81; p < 0.001). The current study suggests the possibility of utilizing social media digital phenotypes as a new tool for intake risk assessment to identify individuals most at risk of treatment dropout and relapse.

Close

  • https://pubmed.ncbi.nlm.nih.gov/37095253/
  • doi:10.1038/s41386-023-01585-5

Close

Primary Sidebar

News

  • All News and Featured Publications
  • Featured Paper of the Month
  • Hot off the Press
  • Reviews to Read
  • IRP News
  • Awards
  • Technology Development Initiative Paper of the Month
  • Seminar Series
Home / News Main / Hot off the Press / AI-based analysis of social media language predicts addiction treatment dropout at 90 days
  • National Institute on Drug Abuse
  • NIH Intramural Research Program
  • National Institutes of Health
  • Health and Human Services
  • USA.GOV
  • Emergency Contacts
  • Employee Assistance
  • Treatment Information
  • Contact Us
  • Careers at NIDA IRP
  • Accessibility
  • Privacy
  • HHS Vulnerability Disclosure
  • Freedom of Information Act
  • Document Viewing Tools
  • Offsite Links
  • National Institute on Drug Abuse
  • NIH Intramural Research Program
  • National Institutes of Health
  • Health and Human Services
  • USA.GOV
  • Emergency Contacts
  • Employee Assistance
  • Treatment Information
  • Contact Us
  • Careers at NIDA IRP
  • Accessibility
  • Privacy
  • HHS Vulnerability Disclosure
  • Freedom of Information Act
  • Document Viewing Tools
  • Offsite Links

  • Home
  • News
    ▼
    • Featured Paper of the Month
    • Reviews to Read
    • Hot off the Press
    • IRP News
    • Awards
    • Technology Development Initiative Paper of the Month
    • Seminar Series
    • Addiction Grand Rounds
  • About
    ▼
    • About NIDA IRP
    • Contact Us
    • Directions and Map
    • Careers at NIDA IRP
    • Emergency Contacts
    • Employee Assistance Resources
  • Organization
    ▼
    • Faculty
    • Office of the Scientific Director
    • Office of the Clinical Director
    • Office of Education and Career Development
    • Administrative Management Branch
    • Molecular Targets and Medications Discovery Branch
    • Cellular and Neurocomputational Systems Branch
    • Molecular Neuropsychiatry Research Branch
    • Neuroimaging Research Branch
    • Behavioral Neuroscience Research Branch
    • Integrative Neuroscience Research Branch
    • Translational Addiction Medicine Branch
    • Core Facilities
    • Community Outreach Group
  • Training Programs
    ▼
    • Office of Education and Career Development
    • OECD Awards
    • Summer Internship Program
    • Postbaccalaureate Program
    • Graduate Partnership Program
    • Postdoctoral Program
    • NIDA Speakers Bureau
    • Clinical Electives Program
    • Clinical Mentoring Program
  • Study Volunteers