install.packages("public.ctn0094data")
install.packages("public.ctn0094extra")Competition
How can we optimize medication treatments for opioid use disorder?
Which individuals are more at risk of dropping out of a treatment program?
Which individuals are more at risk of relapsing?
The R/Medicine planning committee welcomes your participation in the virtual R/Medicine data challenge. Participants will have from now until May 17th, 2024 to work on submissions for the challenge. There will 1st, 2nd, and 3rd place prizes for overall quality of analyses/Quarto docs, as well as prizes for 2 honorable mention categories: best table and best visualization.
It is well known that opioid use disorder is currently a major public health concern that has significant implications for the entire healthcare system. Understanding more about how we can best develop effective treatment programs is needed now more than ever as we continue to combat the opioid epidemic.
The R/Medicine Data Challenge
In this data challenge, you will analyze various factors that may contribute to either treatment success or treatment failure for opioid use disorder using harmonized data collected from three multi-site clinical trials from the CTN. We encourage you to read more about how this data was collected here.
We ask that all analyses be done and submitted as Quarto documents. Participants can submit as an individual or as part of a team (but not both). The maximum team size is 3. Your goal will be to transform these datasets into actionable indicators that illustrate which factors are important to understanding why some patients achieve success with treatment for opioid use disorder and why others drop out or relapse.
The data for this challenge is available on CRAN.The following 2 code chunks demonstrate how to install and load the necessary packages. You can read more about the packages on CRAN here and here.
library(public.ctn0094data)
library(public.ctn0094extra)In addition to the provided data, you may request to use additional publicly available data in your submission by submitting a request by May 1st, 2024 at 5:00PM EST. If approved, the data will be made available to all participants.
You will be evaluated by judges based on the relevance, completeness, and quality of your submission. Remember that tables and visualizations will also be judged and ranked separately for the prizes of best table and best visualization. Judges include:
Raymond Balise, PhD, University of Miami
Peter Higgins, MD, PhD, MSc, University of Michigan
Bryan P. Mayfield, PharmD, MS, Precision Analytical
Example Topics
Which demographic characteristics are most associated with drop out/relapse rates?
Are there any drugs/substances that seem to be more associated with higher or lower drop out/relapse rates?
Guidelines
When: Now - May 17th, 2024
Where: The R/Medicine data challenge is entirely virtual, with information and rules of the challenge available on this website. Updates will be posted on this website.
Who you are: We welcome any participants over age 18, especially undergraduate students, graduate students, and early professional data scientists and biostatisticians. Analysts based in academic institutions, government statistical offices, think tanks, policy labs, and community organizations are encouraged to participate.
What topics you can explore: Participants can analyze the included opioid use disorder data in tandem with with any other related publicly available data submitted for approval before May 1st, 2024.
Submissions and evaluation: Participants will conduct their analyses and submit a short project narrative that describes the research question, analytic approach, and key findings. We encourage participants to find creative ways to incorporate tables, visualizations, and other aspects of data storytelling to create a compelling narrative. In addition to their completed analysis with the indicators they used, participants will be asked to submit documentation describing each step of their process. The documentation should be detailed enough as to make the project fully replicable. The narrative, methods, organization, and documentation of each participant’s project will be evaluated for relevance, completeness, and quality. More information on evaluation will be added to the FAQs section at the bottom of this page.
Prize Information
| 1st Place | 2nd Place | 3rd Place | |
|---|---|---|---|
| Best Overall Analysis/Quarto Doc | Free Posit Conf 2024 Registration | $300 off Posit Conf 2024 Registration | $100 off Posit Conf 2024 |
Best Table - $500 off Posit Conf 2024 Registration
Best Visualization - $500 off Posit Conf 2024 Registration
Finalists will be required to present their project at the R/Medicine conference (can be prerecorded).
FAQs
Question: Are IDs of the care team or individual case workers associated with site/clinic visits available?
Answer: No. The granular details on treatment site and care providers are masked to assure the anonymity of the patients. However, the information on the treatment site/clinic can be found in the table named site_masked.
Question: Is there a way to identify when participants no-show to the clinic?
Answer: Unfortunately, there is no easy way to determine this. In theory, participants should have received weekly screenings, but this was likely not the case for all participants.