Public Health From the Cloud
How big data, GIS and interactive maps can elevate public policy
I used to love Sim City, and probably played it 15 hours a week when I was in high school. I can’t think of a better simulator of real world economics at that kind of scale, and the sheer fun of having to balance so much complexity in an open-ended, pick-your-own-victory-conditions kind of way was addicting. While it doesn’t really tackle microeconomics like car ownership or household consumption choices, it does provide a good model for conceptualizing how public planning decisions drive macro behaviors.
Take zoning — which in Sim City is limited to Residential, Commercial, and Industrial. Placement of zones when first building a city has long term consequences on traffic patterns, prevalence of air and water pollution, land value/income demographics, quality of public services such as education, and of course citizen health.
These same concepts apply in real life as well, where the way citizens respond to urban planning and public policy leads to poorer health outcomes. Long commutes by car to jobs that involve sitting for most of the day result in worse health for the commuters, while the pollution caused by emissions from heavy traffic negatively impact the health of citizens throughout the region. Sprawl of the city population out to suburbs that causes the long commutes also increases the expense of providing public services like police, education, and utilities, straining the city’s ability to equitably serve the whole population.
Without even touching issues of food insecurity or food deserts we can see the complicated relationship between public policy and health.
Challenges facing healthcare today
We are locked politically in a debate about the overwhelming cost of public health care. No matter which side you’re on, you are guaranteed to fail if you do not specifically design the healthcare system to focus its energy on preventing chronic diseases like Diabetes, Heart Disease, and Obesity. A full 86% of all healthcare spending is attributable to chronic disease. Think about that for a moment. We spend 15% of our total GDP to care for health issues that are preventable.
Democrats say things will be better if no one is uninsured and everyone has access to care. Republicans say things will be better to be punitive towards healthcare users who drive the most cost. But if you listen carefully, neither side has a means-tested model for actually reducing chronic disease prevalence at scale.
Big Data and GIS to the rescue
In lieu of ready solutions, we can get good direction from the data. The benefit of displaying large datasets geographically is that you take away a layer of mathematical abstraction and see where your data fits in the real world. For example, if you overlay a map of air pollution, road traffic, income by ZIP-code, and rate of asthma, you begin to get a better sense of the variables that might confound a two dimensional graph of asthma rate over time by ZIP-code. You might also get a better sense of what unrelated zoning or commercial development policies are affecting the health conditions of nearby residents. It effectively enables health outcomes driven public policy focus.
On the government service side of things, applying data-driven maps to public health allows us to dig deeper than demographics and into psychographics. This is essential for any strategy that involves changing human behaviors — and that is the bulk of the work that’s involved in chronic disease prevention. Two people who are the same age, race, gender and income level are liable to respond very differently to the same set of stress factors. Research has shown a personalized approach to prevention is much more effective than any other model. Mapping patterns of movement and other behaviors within a community space will provide an incredibly instructive view of where and how to best engage with each person as a patient.
Leveraging this power within cloud-based applications
Software applications have become the most dominant way of interacting with data for those on the frontlines of care since the HITECH Act in 2009 mandated the meaningful adoption of electronic medical records. This presents an incredible opportunity for those in public health (the back back office of healthcare) to collaborate closely with clinicians who are working directly with the population. Much like the infantry of today’s military serve as spotters for heavy artillery or aircraft, public health experts can help direct clinical staff in the approaches to care that will produce the best outcomes for the community.
Mapping platforms like Mapbox enable complex datasets to be displayed interactively and embedded within mobile and web applications. With extensive documentation and modular APIs, it is easy to build a delightful experience for users across various technical backgrounds and clinical specialties. You can easily take shapefiles or GeoJSON objects and display via a range of touch map interactions, even in real-time. Some examples of high-volume data companies that use Mapbox include Foursquare, Metromile, and Instacart. Other platforms with equivalent features include Google Maps API and ArcGIS.
Filling the gaps
The biggest challenge right now in this emerging space of big data + maps in healthcare is the availability and completeness of data. While a good number of local governments in major cities have Open Data projects, and various federal and state government agencies have a range of datasets available, very few of them are consistently updated.
The challenge remains in finding datasets that are complete and to the geographic granularity required. In order to give back to the GIS community that has helped us so much, we are sharing our in-house library of GIS data. We’ve organized it by city, county, state, region, and national level. Our hope is that together as a tech and data community we can provide all of the pieces for the next generation of civic oriented hackers to help their communities get the insights necessary.
We believe it’s possible to address many of the structural issues within our economy by shifting our top value from economic growth to optimal health outcomes. This shift in mindset will encourage a much greater awareness of how one aspect of a policy platform might affect outcomes for another area.
If you are new to GIS here are some handy resources to get you started (assumes familiarity with transforming big datasets as well as basic HTML/JS/CSS):
Shingai is based out of Berkeley, CA. He is the CEO of ProjectVision, applying machine learning to preventive care. Check out their blueprint for better population health — a model for payers, providers and communities to work together to tackle the causes of chronic disease.
