Developing Data Tools for Natural Disasters: Implementing Best Practices for Electricity-dependent Medicaid Enrollees (2019-2020)
The Medicaid program provides basic health care services for 2 million low-income and/or disabled people in North Carolina, representing 2 in 5 people in the state. Medicaid beneficiaries are among the hardest hit by hurricanes as low-income individuals are more likely to live in flood-prone areas and have primary dwellings that are less structurally sound, and they are less likely to have flood insurance.
Aside from the direct risks of flooding and winds, many Medicaid beneficiaries depend on medicines that require refrigeration or special medical devices that require electricity. When the power is out for prolonged periods after hurricanes and other natural disasters, these individuals face immediate and severe challenges that can directly threaten health and life. To help address this issue, we need to build new data tools that first responders and emergency personnel can use in real-time to help these individuals during future storms.
This project will create emergency preparedness tools and algorithms using Medicaid claims data to identify individuals with electricity-dependent medical equipment or important refrigeration-sensitive medications. Once developed and tested, these algorithms can be run in the days or weeks before a hurricane strikes to identify high-risk patients. The resulting list of high-risk patients can then be securely disclosed to emergency responders and local health departments to prioritize recovery efforts. This work will be conducted in partnership with the North Carolina Department of Health and Human Services.
These methods and the accompanying documentation and training materials developed by this project will be made available to state leaders so they can be deployed in advance of the next storm. This will enable on-the-ground first responders, public health departments and policy officials to prioritize rescue operations and relief efforts for subsets of especially vulnerable Medicaid beneficiaries in affected areas.
In addition to analyzing quantitative data, the team will interview policy officials, public health officials and emergency response officials in states frequently affected by major hurricane events (NC, FL, NJ) to gather best practices from recent storms to inform development of the analytical model.
Statistical codes in SAS or R that can be run on Medicaid claims data; operational plan for state agencies to use the statistical code and analysis results; report detailing preliminary results and visualizations; manuscript describing the technical artifacts, best practices and results of the project
Spring 2019 – Fall 2019
- Spring 2019: Assign sub-teams; begin Medicaid orientation, claims data orientation and literature review; request institutional approval (IRB process); begin interview subject identification and scheduling; develop interview guide; submit NCDHHS approval; begin on-site interviews, develop analysis plan/table shell drafting, begin initial planning for de-identified assessment and visualization; develop methods and analytical plan; test statistical analysis methods and algorithm; hold interim review with NCDHHS
- Summer 2019: Begin manuscript development; develop training documentation and methods companion guide (there will be select opportunities for student work over the summer; not required for all team members)
- Fall 2019: Conduct final interviews; produce interview summary and findings; iterate on analysis plan/table shells; iterate on visualization plan; run final analyses, troubleshoot and use results to create draft visualizations; create outline operationalization plan using interview findings and team analysis experience; interim review with NCDHHS; continue iteration on technical analysis and operationalization plan; review all written documents by full team; produce final deliverables; present to NCDHHS
This Team in the News
- Hilary Campbell, Margolis Center for Health Policy
- Aaron McKethan, School of Medicine-Population Health Sciences
/graduate Team Members
Rachel Salzberg, Public Policy Studies-AM
Allison Young, Interdisciplinary Data Science - Masters
Xin Zhang, Master of Environmental Management, Energy and Environment
/undergraduate Team Members
Saba Ali, Int Comparative Studies (AB)
Suman Bajgain, Electrical & Computer Egr(BSE), Computer Science (BSE2)
Sarah Blau, Biomedical Engineering (BSE)
Rahul Krishnaswamy, Political Science (AB)
Amelia Martin, Biology (BS), Global Health (AB2)
Mishek Thapa, Statistical Science (BS)
/yfaculty/staff Team Members
Bradley Hammill, School of Medicine-Population Health Sciences
Azalea Kim, School of Medicine-Medicine: General Internal Medicine
Andrew Olson, School of Medicine-Duke Clinical Research Institute
Donald Taylor, Sanford School of Public Policy
/zcommunity Team Members
North Carolina Department of Health and Human Services
Shreya Shah, Undergraduate Student, UNC-Chapel Hill