I will leverage social media conversations, specifically Reddit, to examine treatment help-seeking at an unprecedented scale, thereby collecting evidence ?in the wild? as a new strategy to monitor and respond to the opioid crisis through the following aims.
Aim 1. Data Driven Characterization of Treatment Help-Seeking Among Self-Identified Opioid Users: I will systematically identify Reddit posts about opioids, identify if the post is help-seeking for treatment, and contextualize the content of help-seeking posts.
Aim 2. Data Driven Characterization of Lay Responses to Help-Seeking from Self-Identified Opioid Users:
I aim to understand how peers organically respond to help-seeking for treatment by building data-driven models to discern what attributes of posts are predictive of receiving a response from a peer and to examine the typical content and reciprocity of responses.
Aim 3. Assess the Quality of Lay Responses to Help-Seeking from Self-Identified Opioid Users:
I aim to identify the types of treatment, treatment services, and treatment referral services that are provided in the responses and assess the quality of lay responses to opioid treatment help-seeking. Currently, little is known about how drug users engage social media to support or combat their addictions. Reddit is the sixth most popular website in the US with many reporters noting that it may be the last lifeline for opioid users. Yet, little has been done to tap this potential source for social media monitoring, especially given the difficulty of finding self-identified opioid users who are already contemplating strategies to treat their addiction. This study adds a new significant perspective to social media monitoring and will be among the first in public health to consider social media conversation threads by assessing unique posts about opioid help-seeking, the responses these posts get, and the quality of responses to addiction help-seeking, thereby creating a knowledge base for public health leaders to respond. Success of these aims will generate a number of future research directions that will support interventions that address the needs of self-identified opioid users seeking help on Reddit. Moreover, within all three aims this study will bring significant methodological advances to the study of social media around substance use with the aims relying heavily on automated approaches and analysis first developed in the computational sciences engendering more data scientists to join public health to create actionable knowledge. https://docs.google.com/document/d/1z4XHd63xhPZT8OFf9bHQ2q6XHpYp_c4AbouAOtAC3SU/edit 1/1

Public Health Relevance

- Google Docs Project Narrative An explosion of surveillance tools fueled by big media data (including social media) is changing public health. These data reveal what a person is thinking/doing and when they are thinking/doing it based on the content and timing of their public posts, providing an assessment of organic help-seeking for addiction and peer-to-peer knowledge sharing that was never before possible. I will extract actionable intelligence from social media data, specifically Reddit, to support more agile, responsive opioid intervention research by studying (a) treatment help-seeking of self-identified opioid users, (b) lay responses to help-seeking of self-identified opioid users, and (c) whether lay peers link self-identified opioid users to evidence-based treatment, treatment services, or treatment referral services. https://docs.google.com/document/d/1RyIXTdDAbFSkCe81q6hThDiZzKGCXkCy8npWgzr5wfM/edit 1/1

Agency
National Institute of Health (NIH)
Institute
National Institute on Drug Abuse (NIDA)
Type
Mentored Quantitative Research Career Development Award (K25)
Project #
1K25DA049944-01
Application #
9872029
Study Section
Special Emphasis Panel (ZRG1)
Program Officer
Zur, Julia Beth
Project Start
2020-04-15
Project End
2025-03-31
Budget Start
2020-04-15
Budget End
2021-03-31
Support Year
1
Fiscal Year
2020
Total Cost
Indirect Cost
Name
University of California, San Diego
Department
Internal Medicine/Medicine
Type
Schools of Medicine
DUNS #
804355790
City
La Jolla
State
CA
Country
United States
Zip Code
92093