Understanding Public Health Systems through Data Integration and Visualization

In today’s age, data is ubiquitous – public health systems certainly express this fact, with datasets such as support assessment information, Medicaid expenditure data, and data from case management systems all providing important information about individuals in service systems.

However, these data often reside in information silos that make meaningful interaction and knowledge translation between them difficult. Data integration is the combination of technical and business processes used to combine data from different sources into meaningful and valuable information. Data integration digitizes workflow, improves collaboration, saves time, reduces errors, and, ultimately, delivers more valuable data that can be analyzed and acted on.

Additionally, visualization tools can organize and display this information in meaningful ways that allow policymakers to make sense of the data and make data-driven decisions. Users of these visual interfaces can scan, recognize, understand, and recall visually-structured representations more rapidly than they can process non-structured representations. The science of visualization draws on many fields, including perceptual psychology, statistics, graphic design, and advances in rapid processing and dynamic displays to design user interfaces that permit powerful interactive visual analysis.

For this presentation, Megan Villwock, Business Systems Analyst with Cambridge-based Human Services Research Institute (HSRI) will demonstrate how data integration and visualization methods have been used in several jurisdictions to help policymakers understand their intellectual and developmental disability (IDD) service systems and make data-driven policy decisions as part of a systems redesign process. Among the data sources used will be Supports Intensity Scale assessment data, Medicaid Waiver expenditure data, demographics derived from case management systems, and historical metadata.

Taken together and measured over time, these integrated data can provide policymakers with a relevant, timely, and comprehensive view of their systems to foster responsive policy decisions that best serve people.

Transcript

1. Understanding Public Health Systems through Data Integration and Visualization Presenter: Megan Villwock ◦ mvillwock@hsri.org ◦ www.hsri.org
2. What is public health? 2 * Winslow, Charles-Edward Amory (1920). "The Untilled Field of Public Health". Modern Medicine. 2: 183–191
3. Patient Delivered Partner Therapy 3 Having the medication to dispense on site at no cost has been very helpful to increase the likelihood that the partner will receive treatment. The provision of the Azithromycin has made it especially important to our teens. Many male partners will not come to the clinic. Patients are so appreciative that both they AND their partners will be treated promptly.
4. Patient Delivered Partner Therapy 4 40 agencies 201 agencies 24,355 CT+ patients 62 sites 33.5% submitted logs
7. The problem isn’t a lack of data (Well, sometimes it is) 7
8. 8 Multiple Sectors
9. 9 Competing Governance Requirements
10. What is a program participant? 10 It depends on which database you’re in Case Management – “a participant is an individual who is currently on someone’s case load” Financial Management – “a participant is an individual who was billed for a paid service through our agency”
11. 11 Data Siloes
12. 12 Back Office Workarounds
13. 13 Lots of data, but scarce information
14. Integrated service delivery and whole- systems views Collaborative cross-program, cross-agency efforts can provide the right mix of services at the right time About Us Improving the systems that improve lives 14
15. 15 Are we allocating available public funds appropriately?
16. What do we need to know to answer the question? Who is being served? What is their assessed level of need? What resources did they have access to? How much money did each person ultimately spend? And, ideally… Did this make a difference? 16
17. What data are available? 17 Case Management System Houses data for case managers to keep track of information about the people they work with. Medicaid Management Information System Claims processing and information retrieval system that controls Medicaid business functions, including management reporting. Assessment Data System Database application that supports administering, scoring, and retrieving assessment data. Different operational uses Different vendors Different formats and layouts Different definitions Different (or no) standards Unorganized Data Relevant data from other sources, kept either formally or informally. Typically not housed in any structured database or file system.
18. • What background information is relevant or essential? • What data are available? • Who is the decision maker? • What would a successful outcome look like? • Data sourcing • Rigorous filtering and transformation process • Properly formatted physical structure • Metadata • Security and Privacy • Usability • Real-time visibility • Always on • Delivered as a service • Business intelligence tools What do we need to do to answer the question? Context 18 Standards Technology The question was: Are we allocating available public funds appropriately?
19. • Age • Region • Living setting • Assessed level of need • Assigned support budget • Services authorized • Services rendered • Dollars spent • Exceptions reviews and outcomes • Amount spent in relation to assigned budget • Quality of life indicators What background information is relevant or essential? People 19 Services Outcomes
20. What data are available? 20 Case Management System Houses data for case managers to keep track of information about the people they work with. Medicaid Management Information System Claims processing and information retrieval system that controls Medicaid business functions, including management reporting. Assessment Data System Database application that supports administering, scoring, and retrieving assessment data. Unorganized Data Relevant data from other sources, kept either formally or informally. Typically not housed in any structured database or file system.
21. Standards + Technology Integrated data capture and reporting tools provide timely, high-quality information for decision-making Integrate Bring data together from multiple systems in a single technology solution Curate Make the data useful for discovery and analysis, and ensure the value of the data as it is maintained over time Communicate Convey ideas and information in forms that can be seen 21
22. Verity Analytics Integrating, curating, and communicating data to drive system transformation Continuous integration with disparate data sources On-demand, 24/7 access to relevant data and reporting Historical tracking across data systems to better understand chains of events that lead to specific positive or negative outcomes 22
23. Data Visualization The modern equivalent of data communication 23
24. 24 The greatest value of a picture is when it forces us to notice what we never expected to see. - John Tukey
25. 25 When communicating results to non-technical types, there is nothing better than a clear visualization to make your point - John Tukey
26. John Snow and the cholera outbreak of 1854 26
27. Demo Are we allocating public funds appropriately? 27
28. How interactive visualization lets you dig deeper Pose multiple questions per visualization Switch axes, add confounding factors, or break down specific variables Focus on detail Zoom in or hover over areas of interest to reveal exact values Summarize the data you need to know Explore your data as much or as little as you need to Tell a story through data Look for variation or trends over time, then take it one step further – interact with your data to explain the pattern you’ve found 28 Static analyses only reveal part of the story
29. References/Recommended Reading 29 Storytelling with Data – A Data Visualization Guide for Business Professionals Cole Nussbaumer Knaflic Information Dashboard Design – Displaying Data for At-a-Glance Monitoring Stephen Few