The Power of Geo-Analytics (and Maps) to Improve Predictive Analytics in Healthcare
If a picture is worth a thousand words, then a good map is worth at least twice that…
There is much potential for using maps and geo-analytics as a tool for healthcare predictive analytics. Geography and spatial relationship is a significant tool in terms of outcomes research, comparative clinical effectiveness, and evidence-based medicine.
As early as the 1840s, Dr. John Snow used a map to track cholera deaths in the Soho district of London from a contaminated water source – the now infamous Broad Street pump (see image below). Dr. Snow is widely considered to be the father of modern epidemiology.
Geo-Analytics: Ready for Primetime in Healthcare
Geographic information systems (GIS) and geo-spatial analysis is a well-developed industry – passing its 50th anniversary in Fall 2012; yet, with the possible exception of epidemiology, much of healthcare and medicine has not fully leveraged this powerful analytics technology. Various national and local data sources are captured as map layers and used with geo-analytics to routinely optimize supply chain and logistics, support military deployment, and forecast weather.
A natural extension of these established approaches is to leverage GIS mapping of health care facilities, patient disease burden, and accountable care population health. This robust technology is an indispensable part of many industries. Now is the time for healthcare to fully leverage this same power of geo-analytics.
Geo Data and Medical Informatics
- Tracking water quality in major metropolitan areas (University of Cincinnati).
- Exploring your personal medical ‘Place History’ and exposure to reported chemicals (ESRI).
- Visualizing the risk of heart disease and stroke in the United States (CDC). (See image below.)
Geographic variation in medicine (or geo-medicine) is championed by well-known thought leaders such as Jack Wennberg and Jim Weinstein (Dartmouth), Bill Davenhall (ESRI, Ted Talk), and Jack Lord (University of Miami). Stemming from early work in the 1970s, these studies were first called “small area variation analyses” by Wennberg and Gittelsohn.
Their simple message was geographic variation is explained largely by different practice styles, where the greater the distance apart (spatial variation), the less consensus on the evidence base for any given medical point of view. In other words, the weaker the clinical evidence base, the greater the geographic variation in standard medical practice and protocols.
Real World Geo-Spatial Analytics in Healthcare
So that brings us to the real question at hand – how ‘healthy’ is your healthcare system?
Health Catalyst is developing the methodology to integrate multiple inputs into a visually appealing analysis (think, maps) of geographic care boundaries, population health demographics, and provider locations.
Potential map layer inputs can include sources such as Hospital Service Areas and Healthcare Referral Regions (Dartmouth Atlas of Health Care), CMS service area definition (Hospital Compare), data from the U.S. Census Bureau and Health Benefit Program filings with state health insurance departments, among others (see image below). In addition, we are working to apply Central Place Theory and healthcare facility levels defined by medical specialty composition to medical service area definitions.
The goal of all these analyses is to measure and improve referral tracking and management, patient leakage to outside networks, and objective decision making for population health managers.
We join with other geo-medicine leaders offering to you that indeed “geography is the destiny of medicine.”
What do you think the future holds for geo-analytics and healthcare? Can you think of more example of how it will give us tools to provide better and more effective care?
Would you like to use or share these concepts? Download this presentation highlighting the key main points.