Stroke Data Challenge

Stroke is the No. 5 cause of death and a leading cause of disability in the United States.  At the American Heart Association / American Stroke Association, we want to see a world free of cardiovascular diseases and stroke.  Our mission is to be a relentless force for a world of longer, healthier lives.  We are hosting a data challenge to encourage cross collaboration among researchers to deepen our understanding of the impact of social and structural determinants of health on stroke.  This data challenge is funded by the Gordon and Betty Moore Foundation.

  • Health disparities include environmental threats, individual and behavioral factors, inadequate access to health care, poverty, and educational inequalities. 
  • Social determinants of health include resources such as food supply, housing, economic and social relationships, education, and health care. 
  • Structural determinants of health include economic, governing and social policies that affect pay, working conditions, housing and education.

Research Question

The Get With The Guidelines-Stroke data challenge seeks proposals to advance the understanding of how social and structural determinants of health influence stroke presentation, care quality, and clinical outcomes. 

All applicants must submit a proposal that  leverages the novel data linkage which includes information about life expectancy, education, income, home ownership, poverty, and employment to better understand how social and structural determinants of health impact stroke presentation, care, and outcomes.  Possible topics may include:

  • Timing and method of presentation to hospitals for stroke
  • Delivery of quality care for patients with acute stroke
  • Timing and utilization of proven interventions for stroke
  • Complications from stroke treatment
  • Outcome from stroke
  • Discharge disposition
  • If you want to explore and analyze more than one topic, you will need to submit a proposal for each topic.

Data

We have created a unique, linked, dataset specifically for this data challenge.  With the support of the Institute for Health Metrics and Evaluation (IHME), they have merged the GWTG-Stroke registry data (patient-level, inpatient data) with county level health metrics. 

  • GWTG-Stroke registry data from January 2010 – December 2019
    • Get With The Guidelines is a proven in-hospital approach to improving patient outcomes across cardiovascular and stroke focus areas. Each Get With The Guidelines program promotes consistent adherence to the latest research-driven guidelines and measurement, while providing data and information to professionals for continual improvement in patient care.
    • GWTG-Stroke data collection form
  • IHME has produced county-level estimates for life expectancy, all-cause mortality, total stroke mortality, and mortality due to the following stroke subtypes: ischemic stroke, intracerebral hemorrhage, and subarachnoid hemorrhage for the years 2010-2019. These results are available by total population, age, sex, and race/ethnicity.

    • These mortality estimates will be accompanied by county-level sociodemographic covariates related to education and income across the same time series. IHME has produced the following covariates, many of which are also available by race/ethnicity:
      • Proportion of population graduated from college
      • Proportion of population graduated from high school
      • Per capita income
      • Poverty
      • Unemployment
      • Median household income
      • Foreign-born status
      • Home ownership
    • Methods overview for data linkage

    • Data Dictionary describing the county level estimates

How to Participate

  1. Register here(link opens in new window)(link opens in new window)(link opens in new window)(link opens in new window)(link opens in new window)(link opens in new window)(link opens in new window)(link opens in new window)(link opens in new window) to indicate your intent to submit a proposal for the data challenge.

  2. Complete the Stroke Data Challenge proposal form.

  3. Confirm your registration and submit your completed proposal(link opens in new window)(link opens in new window)(link opens in new window)(link opens in new window)(link opens in new window)(link opens in new window)(link opens in new window)(link opens in new window)(link opens in new window) by March 31, 2023 at 5 pm CT.    
    1. Guidance for how to prepare your proposal can be found here.
    2. Participants are encouraged to review our Quality Research Publication Database to ensure their research doesn’t overlap with existing publications. After reviewing, if researchers have additional questions about the feasibility of their proposal, please contact [email protected].
    3. Proposals will be reviewed to ensure there is no substantial overlap with projects already underway through the standard and early career proposal pathways.
  4. If your proposal is accepted, you will be contacted with instructions on how to move forward with the analysis and will be required to execute our standard Non-Disclosure & Data Use Agreement to gain access to the linked dataset on the Precision Medicine Platform.  You will be sent a link to sign the NDA-DUA electronically.

  5. Each challenge participant/team will be provided a workspace on the AHA’s Precision Medicine Platform (PMP) at no cost to conduct all analyses. Use of the workspace is mandatory for participation in the challenge.  

  6. Submit your research analysis by the deadline of September 29, 2023 by 5 pm CT.

Challenge Timeline

  1. March 31, 2023 by 5 pm CT: Registration and Proposals due.

  2. May 16, 2023: Selected proposals notified.  PMP workspace open for analysis. 
    1. Optional ISC Abstract Submission
      1. August 1, 2023 by 5 pm CT: Deadline to submit abstract for review prior to ISC 2024 abstract submission deadline (for participants that want to submit their abstract to ISC 2024).  Participant will be notified of any feedback and approval to submit to ISC.
      2. August 2023 (specific date TBD): ISC 2024 Abstract deadline.
  3. September 29, 2023 by 5 pm CT: Data Challenge Submissions due.

  4. October 30, 2023: Finalists selected and notified.

  5. February 2024: Finalists present their research to an expert panel at the International Stroke Conference and a winner is chosen.

Prizes & Evaluation

 Prizes

Three prizes will be awarded to the top submissions. Recipients of the grand, second or third place prize will be required to publish a manuscript detailing the winning solutions and will be subject to the standard review process.  

  • Grand prize:  $25,000
  • Second place:  $15,000
  • Third place:  $5,000

Evaluation

Submissions will be reviewed by a peer review committee that includes experts and volunteers from the American Heart Association. Reviewers will evaluate the novel information learned from the analyses that address the focus of the research question.  Specifically, the committee will assess the:

  • Findings or results of the data analyses
  • Data and data analyses (methods that support the findings)
  • Novel information learned
  • Overall impact of the findings and analyses on the question’s focus

Data Challenge Submission

Submission Requirements 

The Precision Medicine Platform provides an easy-to-use, web-based user interface that allows participants to access a secure, cloud-based environment. It contains a variety of standard software and packages such as Python and R. The platform workspace leverages Jupyter Notebooks and RStudio to allow users to create notebooks to document and display results.

Each participant must submit a Jupyter notebook from their individual workspace in the Precision Medicine Platform that provides the following:

  • Full name of corresponding applicant and team members 
  • Institution (if applicable)
  • Email address of corresponding applicant
  • Phone number of corresponding applicant
  • A 250-word description of the findings and how they can be applied
  • Description of data analysis and datasets used for establishing, testing, and validating model
  • Figures or tables (if applicable)
  • Summary

Participants will be sent these specific instructions prior to the submission deadline. 

Submitting Results/Notebook

Once analyses are complete and the notebook is ready for submission, per the criteria outlined above, follow the steps below to submit the notebook for review. All submissions must be complete by September 29, 2023 at 5 pm CT.

  1. The submitted notebook should be in HTML format. In Jupyter, follow these steps:
    1. In the File Menu, selecting Download as > HTML (.html). The HTML file will download to a temporary downloads folder.
    2. Re-upload the HTML file to the workspace by clicking the upload icon in JupyterLab or clicking the Upload button on the Jupyter Home tab.
  2. In the workspace, save the notebook in the /mnt/workspace/Export_Files directory as this will sync the notebook to the workspace page in the portal. 

  3. The notebook will be listed under the Export Files tab on the workspace page in the portal. To the right of the notebook, click the box with the arrow that shows “Share for Grant Application” when you hover over it.

  4. Sharing the notebook creates a static link that will be used by the Data Challenge peer review team.

    Note: Only information in the notebook will be visible to the peer review team. Therefore, it is imperative that you include everything necessary before sharing. If you make changes after sharing, you will have to re-share.

Precision Medicine Platform workspaces will remain available to participants if you are working towards publication of a manuscript.  If your data challenge submission is not selected as a finalist and you do not want to publish your research, we will decommission your workspace shortly after the conclusion of the data challenge.

Contact Information

Please email [email protected] with any questions regarding the data challenge.

 

 

Last Reviewed: Feb 3, 2023