The Opioid Hack-A-Thon
Two Days Changing the Future of Opioid Addiction
Two Days Changing the Future of Opioid Addiction
Every day, more than 115 people in the United States die after overdosing on opioids, making the opioid crisis the top current public health problem in the United States. The problem continues to worsen, as opioid overdoses increased 30 percent from July 2016 through September 2017 in 52 areas in 45 states. Specific to California, certain regions of the state have a greater number of opioid overdose deaths than the national average, creating a need for California to implement solutions and prevent a worsening crisis.
Building on insights from the previous US Department of Health and Human Services (HHS) Code-A-Thon in December of 2017, the University of California Institute for Prediction Technology (UCIPT) is hosting a Hack-A-Thon featuring HHS. This Hack-A-Thon is designed to bring interdisciplinary teams (e.g., data science/visualization experts, patients/family members of those affected by the crisis, public health/medical/law enforcement professionals, and legal/ethical/psychological researchers) with new approaches to solve this growing problem. Teams will be provided with various data sources related to opioid outcomes, as well as alternative approaches related to the crisis (e.g., virtual reality and augmented reality data, cannabis data, wearable and social media data, and data on behavior change approaches). Teams will have 24 hours to develop visualizations to address a number of tracks on the topic, with a final panel of judges deciding the winners.
This event will take place in California and have a California focus on issues of the opioid crisis. The goal of the hackathon is for the winning teams to have their solutions implemented within the public health system to help address the opioid crisis. Lessons learned from this hackathon can be applied to other states to help address the national crisis.
This event is part of a study on whether and how hackathons can be used to scale implementations of opioid-related solutions in public health settings. All hackathon participants will become a part of the study. Selected participants will be invited to complete surveys and interviews related to their experiences at the event.
For more information, see:
predictiontechnology.ucla.edu/
https://www.hhs.gov/opioids/about-the-epidemic/index.html https://www.drugabuse.gov/drugs-abuse/opioids/opioid-overdose-crisis#one
Hack-A-Thon
Registered teams will compete across 4 different tracks.
Mentorship
We will have a fellowship program to pick select participants to be fellows. The program will provide mentoring to fellows on their solutions after the event.
Implementation
We will work with the winning teams to implement their solutions into public health settings.
Event Tracks
There will be 4 different event tracks with prize money given to the winners of each track.
Software Application Development
A. Infrastructure/application interface for ethical/secure sharing of opioid-related data among key stakeholders
Opioid-related data typically suffer from reporting lagtimes and lack of detailed information, making it difficult for researchers and public health departments to address the growing crisis. Real-time data can be provided by various stakeholders (e.g., patients, families, industry, public health, law enforcement, etc) through social media, internet search data results, medical records, wearable data, etc. Analysis of these data may help to address this problem by providing real-time novel, personalized information that help public health departments take actionable steps such as deploying interventions and resources. However, there are a number of infrastructural, ethical, and user experience-related questions around collecting and facilitating the sharing of these types of personal data. For example, how could a system or technology like this be created for data to be shared and used in a secure and ethical way so that stakeholders are aware of the risks/benefits of being engaged in this activity?
Winning solutions to this track should include a working prototype of a technology and/or mockup showing the interface for how each of the key stakeholders would use and understand how to use the technology.
​
​
​
​
B. Designing personalized behavior change apps that can be applied by patients, their families, and/or providers/health systems for long-term reductions in overdose-related risk behaviors.
​
Behavior change programs attempting to reduce risk addiction and overdose have had limited long-term success. A number of barriers prevent success in sustainable behavior change, such as physical location distance between patients and recovery centers, especially in rural areas; continued prescribing of opioids and/or co-prescribing of sedatives among providers to patients better suited for different therapies or lower doses; and current technologies that are designed to reduce addiction and/or overdose do not effectively incorporate knowledge and lessons from behavior change science on how to create sustainable behavior change.
Winning solutions to this track should include a working prototype and/or mockup showing the interface/work-flow designs with features for lasting behavior change.
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
​
C. Real-time/prediction models and visualization tools to prevent addiction and overdose
Little data are available on the changing trends in opioid-related medical outcomes needed for public health to intervene and deploy appropriate resources. Through gaining insights into these changing trends, public health official and health systems may be able to immediately act on this information by providing resources and/or interventions to affected regions and individuals. Our advisory board has helped to identify 3 primary areas of need. We are seeking data-based tools/visualizations/models that could help track trends on or help 1) Decrease opioid prescribing among providers, 2) Increase willingness to prescribe (among providers) or use (among patients) medication-assisted therapy, and/or 3) Increase harm-reduction (e.g., increasing availability of naloxone or related overdose prevention kits).
​
​
​
​
​
​
​
​
​
​
​
​
​
​
D. Real-time/prediction models and visualization tools showing new integrative therapeutic benefits and approaches to prevent addiction and overdose
​
Many new types of approaches (either stand-alone and/or integrative with traditional medical/other) are available for combating the opioid crisis (e.g., cannabis, VR/AR, yoga, meditation), however, there have been limited studies looking at whether and how these non-traditional data sources might actually affect or improve opioid-related outcomes.
Winning solutions to this track should be models and/or visualizations that include data from non-traditional public health datasets. For example, instead of relying entirely on models from emergency department/hospital visits or prescription drug use (as expected in Track C above), solutions for this track should include non-traditional data sources such as cannabis data, social media data, internet search data, or yoga data).
Data Science/Artificial Intelligence and Visualization Tracks
TOP TRACK A PRIZE $5,000
​
WINNERS: John Hsu, Sherie Hsieh, Alan Young, Brandon Howard, and Paul Gauvreau developed iPill, an application that regulates opioid dispensing for a patient while offering alternatives to taking the opioid pill.
TOP TRACK B PRIZE $5,000
​
WINNERS: Ellie Gordon, Alex Chen, Brian Nguyen, Cameron Bacciarini, and Andrew Dennis presented Recovery. Recovery incorporates clinical trial data to improve treatment methods.
TOP TRACK C PRIZE $5,000
​
WINNERS: Su Jeong Park, Christine Chen, Karen Lee, and Ah Reum Lee used applications of machine learning to predict opioid overdose deaths in California.
TOP TRACK D PRIZE $5,000
​
WINNERS: Qingpeng Zhang, Jiandong Zhou, Bianca Giusto, and Michael Masterman-Smith used open source data analysis to predict opioid overdose and help reduce opioid usage.
Symposium
October 14th
8:30am - 12:15pm
8:30-8:40 a.m.
Welcome and Opening Remarks
Sean Young, PhD, MS
​
8:40–8:55 a.m.
Opening Remarks from HHS
Latecia Engram, MSPH
​
8:55–9:15 a.m.
Tackling the Opioid Epidemic in California: Opportunities for Data Solutions
Matthew Willis, MD, MPH
​
9:15–9:30 a.m.
Prosecutorial Responses to the Opioid Epidemic
Tony Rackauckas, JD
​
9:35–9:50 a.m.
Opioid Use Disorder: An Equal Opportunity Destroyer
Bharath Chakravarthy, MD, MPH
​
9:50–10:05 a.m.
The HOPE Intervention and Behavior Change Technologies for Opioid Overdose Prevention
Kiran Gill, MPH
​
10:05 a.m. - 10:25 a.m.
Approaches and Challenges for Monitoring Impaired Driving Cases
Captain Helena Williams
​
10:30-10:45 a.m.
Break
​
10:45-11:00 a.m.
Implementation of New Opioid Technology Solutions into the Public Sector
Brian Mittman, PhD
​
11:00-11:15 a.m.
Analysis of Online Health-Related User-Generated Content
Vagelis Hristidis, PhD
​
11:15-11:30 a.m.
Dose of Reality
April Rovero
​
11:30-11:45 a.m.
Intersection of Cannabis and Opioid Use
Jeff Chen, MD, MBA
​
11:45-12:00 p.m.
Applying Lessons from Other Medical Technologies to the Opioid Crisis
Dan Cooper, MD
​
12:00-12:10 p.m.
Thank You and Final Remarks
Sean Young, PhD, MS
Day 1
Hack-A-Thon
October 14th
12:00pm - 12:00am
12:10 pm - 1:00 pm
Participant Check-In
1:00 pm – 1:15 pm
Transition to Auditorium for Opening
Remarks
(Auditorium)
​
1:15 pm – 1:25 pm
Opening Remarks from UCIPT Dir. Sean Young and HHS Chief Data Officer Mona Siddiqui
(Auditorium)
​
1:25 pm – 1:45pm
Design-a-Thon Debrief
(Auditorium)
1:45 pm – 11:45 pm
Team Coding Period
6:15 pm - 6:45 pm
Sponsored Exhibits
6:45 pm – 8:00 pm
Sponsored Dinner
​
Day 2
Hack-A-Thon
October 15th
12:00am - 12:00pm;
3:00pm - 5:00pm:
12:00 am – 7:00 am
Optional Team Coding Period
7:00 am – 8:00 am
Sponsored Breakfast
8:00 am – 2:30 pm
Team Coding Period
11:00 am – 1:00 pm
Sponsored Lunch Available
(Atrium)
​
11:00 am – 1:30 pm
Round 1 Judging and Decisions
(Teams Continue Coding Throughout)
​
1:30-1:45 pm
End of 24-hour Coding Period
Transition to Auditorium for Final Presentations
Round 1 Winners Announced
​
1:45-2:30 pm
Round 1 Winners Finalize Pitches/Demos
(Work Rooms then return to Auditorium for Final Pitches)
​
2:30 pm – 3:00 pm
Challenge A Track: Final Presentations and Judging
​
3:00 pm – 3:30 pm
Challenge B Track: Final Presentations and Judging
​
3:30 pm – 4:00 pm
Challenge C Track: Final Presentations and Judging
​
4:00 pm – 4:30 pm
Challenge D Track: Final Presentations and Judging
​
4:30 pm – 5:00 pm
Final Judges Deliberation and Winners Announced
​
​
​
Agenda
Frequently Asked Questions
Q: WHAT IS A HACKATHON?
A hackathon (also known as a hack day, hackfest or codefest) is an event in which computer programmers, and others involved in software development and hardware development, including graphic designers, interface designers and project managers, collaborate intensively on software projects. —Wikipedia
​
​
Q: WILL I LOSE IP RIGHTS TO THE SOLUTIONS I CREATE AT THE HACKATHON?
No, you will not give up any intellectual property (IP) to what you create at the Hackathon. In addition, we strongly encourage you continue to develop and implement your ideas post-Hackathon.
​
​
Q: WHAT SHOULD I BUILD?
You will be provided with datasets which you can use to build data science models and visualizations. The goal will be to build the models and visualizations that have the best chance of being implemented in public health settings.
​
​
Q: CAN I START MY PROJECT BEFORE THE HACKATHON?
We welcome teams to brainstorm and make tentative plans in advance, but all content, code, analysis, and visualizations must be created during the Hackathon.
​
​
Q: CAN I USE MY OWN DATA AND SOFTWARE TOOLS?
Yes, while we will provide you with datasets, some visualization tools, and software on Socrata (closer to the start date), you are welcome to collect your own data and use your own visualization tools. However, any additional tools or data used must be shared with the judges in order to verify your analyses. We encourage teams to incorporate many different types of datasets including social media data, internet search data, wearable data, or virtual reality data.
​
Q: WHAT CRITERIA WILL YOU BE JUDGED ON?
You will be judged on your creativity, the potential impact of your solution, and feasibility in which the solutions can be implemented in public health settings in California.
​
​
Q: HOW MANY HACKERS CAN MAKE UP A TEAM?
Teams must consist of 3 to 5 participants. We welcome industry-based, academic-based, and mixed industry/academic teams to apply.
​
​
Q: WHAT DO I BRING?
Bring any technology you want to use during the Hackathon. Please bring your laptop, and feel free to bring any other devices or hardware you may want to incorporate into your solutions (ie VR or AR headsets, activity trackers, etc). All food and drinks are provided for free.
​
Q: WHAT DOES IT MEAN THAT THIS IS A RESEARCH STUDY?
This event is part of an NIH-sponsored research study. By participating in this event, you are agreeing to be contacted with surveys and interviews about your experience during and after the event. Winning teas will be expected to continue attempting to work on and implement their solutions after the event, and will be provided with mentorship and travel support to assist them. We will seek to study whether and how winning teams continue working together and the results of this event.
​
Q: QUESTIONS ABOUT SPONSORSHIP?
​
Contact Us
sponsorship.opiodplus@gmail.com
​
Join Our Community
Invitation-only, private channel for Sponsors: Opioid West Hack-A-Thon+ 2018
​
Case Study:
Using Google search data to predict opioids
As an example of a type of analysis/visualization for "Hacking the Opioid Crisis," our research team collected publicly available data from Google search terms. We collected publicly-available data on Google internet searches for both prescription opioids and illicit opioids in 9 large US cities over time, as well as data on emergency department admissions for heroin in those same cities. We found that, within any year, cities that had a greater number of internet searches for opioids, also had a greater number of emergency department visits for heroin the next year. In other words, internet searches for opioids might be used to predict future emergency department visits for heroin.
​
Overall the study creates a call for more work on this topic. We are hoping that participants in this Hack-A-Thon will similarly find new ways to use data to address the opioid crisis.
Planning Committee
Sean Young, PhD
Executive Director of the University of California Institute for Prediction Technology and Professor at the UCLA School of Medicine
​
Mona Siddiqui, MD
Chief Data Officer at Department of Health and Human Services
Alpesh Shah, MBA, MS
Sr Director, Global Business Strategy & Intelligence at IEEE SA
​
Edward Bukstel
CEO at Clinical Blockchain
Lloyd Green
Director, Engagement Marketing & Creative Community Services at IEEE SA
Larry Chu, MD
Professor of Anesthesiology, Perioperative and Pain Medicine, Executive Director of Stanford Medicine X, and Director of the Stanford Anesthesia Informatics and Media (AIM) Lab
​
Dan Cooper, MD
Professor of Pediatrics at the Pediatric Pulmonology Division, the Founding Director of the Institute for Clinical Translational Science, and the Program Director of the UC Irvine Clinical Research Center
​
Luis Cendejas, MPH
Healthy Campus Project Manager, Institute for Clinical and Translational Science
​
Claire Houlihan
Research Associate at UCLA Computer Science
​
Stephanie Soliz
Project Coordinator at UCLA Family Medicine Department
​
Arielle Radin
Doctoral Student, University of California, Los Angeles
​
Advisory Committee
We have assembled an advisory board of public health experts, researchers, patients, and families affected by the opioid crisis.