eDrugTrends is an inter-disciplinary project between the Ohio Center of Excellence in Knowledge-enabled Computing (Kno.e.sis) and the Center for Interventions, Treatment and Addictions Research (CITAR) at Wright State University developed to monitor cannabis and synthetic cannabinoid use.
Principal Investigators: Raminta Daniulaityte, Amit P. Sheth
Co-Investigators: Robert Carlson, T.K.Prasad, Ramzi Nahhas, Silvia Martins (Columbia), Edward W. Boyer (UMass)
Graduate Students: Farahnaz Golroo, Sanjaya Wijeratne, Lu Chen, Adarsh Alex
Past Members: Pavan Kapanipathi, Sujan Perera,
Postdoctoral Researchers: Francois R. Lamy
Software Engineers: Gary A. Smith
Aims: The purpose of this paper is to analyze characteristics of marijuana concentrate users, describe patterns and reasons of use, and identify factors associated with daily use of concentrates among U.S.-based cannabis users recruited via a Twitter-based online survey.
Methods: An anonymous web-based survey was conducted in June 2017 with 687 U.S.-based cannabis users recruited via Twitter-based ads. The survey included questions about state of residence, socio-demographic characteristics, and cannabis use, including marijuana concentrates. Multiple logistic regression analyses were conducted to identify characteristics associated with lifetime and daily use of marijuana concentrates.
Results: Almost 60% of respondents were male, 86% were white, and the mean age was 43.0 years. About 48% reported marijuana concentrate use. After adjusting for multiple testing, significant predictors of concentrate use included: living in “recreational” (AOR=2.04; adj. p=0.042) or “medical, less restrictive” (AOR=1.74; adj. p=0.030) states, being younger (AOR=0.97, adj. p=0.008), and daily herbal cannabis use (AOR=2.57, adj. p=0.008). Out of 329 marijuana concentrate users, about 13% (n=44) reported daily/near daily use. Significant predictors of daily concentrate use included: living in recreational states (AOR=3.59, adj. p=0.020) and using concentrates for therapeutic purposes (AOR=4.34, adj. p=0.020).
Conclusions: Living in states with more liberal marijuana policies is associated with greater likelihood of marijuana concentrate use and with more frequent use. Characteristics of daily users, in particular, patterns of therapeutic use warrant further research with community-recruited samples.
The ultimate goal of this proposal is to decrease the burden of psychoactive substance use in the United States. Building on a longstanding multidisciplinary collaboration between researchers at the Center for Interventions, Treatment, and Addictions Research (CITAR) and the Ohio Center of Excellence in Knowledge-enabled Computing (Kno.e.sis) at Wright State University, we propose to develop and deploy an innovative software platform, eDrugTrends, capable of semi-automated processing of social media data to identify emerging trends in cannabis and synthetic cannabinoid use in the U.S.
Cannabis remains one of the most commonly used psychoactive substances in the U.S., and current epidemiological studies indicate broadening acceptability. Over the past several years, synthetic cannabinoids (“synthetics,” such as Spice, K2) have emerged as new designer drugs. Synthetics, after gaining popularity as “legal” alternatives to cannabis, have been associated with adverse health effects such as seizures and changes in mental status requiring ICU admission. In the context of profound changes in cannabis legalization policies that are taking place across the U.S., close epidemiological monitoring of natural and synthetic cannabinoid products is needed to assess the impact of policy changes and identify emerging issues and trends.
Web 2.0 empowered social media platforms such as Twitter and Web forums have opened up new venues for individuals to freely share their drug use experiences and post questions, comments, and opinions about different drugs. Content analysis of such tweets and Web forum posts can provide valuable insights about drug user behaviors and attitudes. Content analysis can be extended to include temporal, geographic, and social network dimensions of social media data to track changes over time, examine regional differences, and to identify opinion leaders who influence attitudes and behaviors associated with cannabis and synthetic cannabinoid use. However, because of the volume and challenges introduced by web-based data, social media sources remain largely under-utilized in drug abuse research.
This multi-PI study will integrate cutting-edge information processing techniques, such as the Semantic Web, Natural Language Processing (NLP), and Machine Learning (ML), to advance the analysis of social media data for drug abuse epidemiology research. Our multidisciplinary framework builds on the successes of our collaborative R21 (Grant No. DA030571-01A1) and strong representation in all key areas—substance abuse epidemiology, statistical and qualitative methods, drug abuse toxicology, and state-of-the-art computational and engineering expertise.
- Develop a comprehensive software platform, eDrugTrends, for semi-automated processing and visualization of thematic, sentiment, spatio-temporal, and social network dimensions of social media data (Twitter and Web forums) on cannabis and synthetic cannabinoid use.
- Deploy eDrugTrends to:
- Identify and compare trends in knowledge, attitudes, and behaviors related to cannabis and synthetic cannabinoid use across U.S. regions with different cannabis legalization policies using Twitter and Web forum data.
- Analyze social network characteristics and identify key influencers (opinions leaders) in cannabis and synthetic cannabinoid-related discussions on Twitter.
eDrugTrends builds on the existing infrastructure developed by our research team:
- Twitris - a semantic web application that facilitates understanding of social perceptions by semantics-based processing of massive amounts of event-centric data.
- PREDOSE - a semantic web platform, developed to detect emerging patterns and trends in prescription drug abuse using social media.
Key elements of Twitris and PREDOSE will be adapted and enhanced to develop eDrugTrends, a comprehensive and highly scalable software platform, equipped to process real-time social media data, semi-automate information extraction about knowledge, attitudes and behaviors related to cannabis and synthetic cannabinoid use, and to produce iterative maps and graphical images of drug trends and social (or cyber) networks .
Our study is highly significant in two senses:
- First, the development of eDrugTrends will advance the field’s technological and methodological capabilities to harness social media sources for drug abuse surveillance research.
- Second, our deployment of the platform will inform the field on new trends regarding the use of cannabis and synthetic cannabinoids. The key innovation of our approach is the creative adaptation of the state-of-the art technological advancements in computer science and engineering to meet the unique needs and challenges of drug abuse research. eDrugTrends will have high public health impact by providing a tool that can be used to inform more timely interventions and policy responses to changes in cannabis and synthetic cannabinoid use and associated harms.
- Soon Jye Kho, Swati Padhee, Goonmeet Bajaj, Krishnaprasad Thirunarayan, Amit Sheth. "Domain-specific Use Cases for Knowledge-enabled Social Media Analysis". In Emerging Research Challenges and Opportunities in Computational Social Network Analysis and Mining, pp. 107-121. Springer, Dayton, 2018.
- Ugur Kursuncu, Manas Gaur,Usha Lokala,Krishnaprasad Thirunarayan,Amit Sheth and I. Budak Arpinar. "Predictive Analysis on Twitter: Techniques and Applications". Book Chapter in "Emerging Research Challenges and Opportunities in Computational Social Network Analysis and Mining", Editor: Nitin Agarwal, Springer, 2018.
- R. Daniulaityte, R. Carlson, F. Golroo, S. Wijeratne, E. Boyer, S. Martins, R. Nahhas, A. Sheth. 'Time for dabs': Analyzing Twitter data on butane hash oil use. The College on Problems of Drug Dependence CPDD 2015, Phoenix, Arizona, June 13-18, 2015.
- R. Daniulaityte, Ramzi W. Nahhas, S. Wijeratne, R. Carlson, F. Lamy, S. Martins, E. Boyer, G. A. Smith, Amit Sheth, "Time for dabs": Analyzing Twitter data on marijuana concentrates across the U.S., Drug and Alcohol Dependence, Volume 155, 1 October 2015, Pages 307-311, ISSN 0376-8716, http://dx.doi.org/10.1016/j.drugalcdep.2015.07.1199.
- S. Wijeratne, L. Balasuriya, A. Sheth, Derek Doran, EmojiNet: Building a Machine Readable Sense Inventory for Emoji, In 8th International Conference on Social Informatics (SocInfo 2016) Bellevue, WA, USA, 2016.
- S. Wijeratne, L. Balasuriya, A. Sheth, D. Doran, EmojiNet: A Machine Readable Emoji Sense Inventory, Wright Brother's Day, Wright State University. Dayton, Ohio, USA, 2016.
- R. Daniulaityte, L. Chen, F. Lamy, R. Carlson, K. Thirunarayan, A. Sheth. “When ‘Bad’ is ‘Good’”: Identifying Personal Communication and Sentiment in Drug-Related Tweets. JMIR Public Health Surveillance 2016;2(2):e162. DOI: 10.2196/publichealth.6327
- F. Lamy, R. Daniulaityte, A. Sheth, R. Nahhas, S. Martins, E. Boyer. “Those edibles hit hard”: exploration of Twitter data on cannabis edibles in the U.S. Drug Alcohol Dependence. Drug Alcohol Dependence. 2016;164:64-70. DOI: 10.1016/j.drugalcdep.2016.04.029.
- Sanjaya Wijeratne, Lakshika Balasuriya, Amit Sheth, Derek Doran, A Semantics-Based Measure of Emoji Similarity, In 2017 IEEE/WIC/ACM International Conference on Web Intelligence (WI). Leipzig, Germany; 2017. Demo
- Sanjaya Wijeratne, Lakshika Balasuriya, Amit Sheth, Derek Doran, EmojiNet: An Open Service and API for Emoji Sense Discovery, In 11th International AAAI Conference on Web and Social Media (ICWSM 2017). Montreal, Canada; 2017. Demo | BibTeX
- Innovative NIDA National Early Warning Sysetm Network (iN3)
- Modeling Social Behavior for Healthcare Utilization in Depression
- Context-Aware Harassment Detection on Social Media
- Hazards SEES: Social and Physical Sensing Enabled Decision Support
- Market Driven Innovations and Scaling up of Twitris
- Project Safe Neighborhood: Westwood Partnership to Prevent Juvenile Repeat Offenders
- Market Driven Innovations and Scaling up of Twitris
- kHealth: Semantic Multisensory Mobile Approach to Personalized Asthma Care
- DAO: Drug Abuse Ontology
- PREDOSE: PREscription Drug abuse Online Surveillance and Epidemiology
- SoCS: Social Media Enhanced Organizational Sensemaking in Emergency Response
- Twitris: a System for Collective Social Intelligence
This project is sponsored by the National Institute on Drug Abuse (NIDA) Grant No. 5R01DA039454-02 to the Ohio Center of Excellence in Knowledge-enabled Computing (Kno.e.sis) and the Center for Interventions, Treatment and Addictions Research (CITAR) titled: Trending: Social media analysis to monitor cannabis and synthetic cannabinoid use. Any opinions, findings, conclusions or recommendations expressed in this material are those of the investigator(s) and do not necessarily reflect the views of the National Institutes of Health.
Follow us on Twitter "edrugtrends"
Contact: Farahnaz Golroo