Geospatial analytics tools bring 'another layer of understanding' to population health

The confluence of new care models and technology are enabling data scientists to pinpoint gaps in access to care, address social determinants of health, and map data that informs tactics to improve outcomes at the patient and population level.
By Mike Miliard
11:22 AM
geospatial analytics population health

We now live in a world awash in data. And with that proliferation of detailed patient information comes new opportunities to map – literally – new ways forward for better health.

More and more healthcare organizations are realizing that medical data at the point of care only tells a small part of the story.

"We're using just one percent of the data on patients," said Adrian Zai, MD, clinical director of population informatics at Massachusetts General Hospital. "For 99 percent of their lives, they're not in our care."

Socioeconomic factors, environmental and behavioral health, all of these are big determinants of health and wellness. The challenge is how to gather and present that data in the way that can have most impact. Increasingly, geospatial analytics are being deployed to help visualize trends and patterns that affect patients where they live.

The tools are nothing new, but a confluence of interest and technology, has made them more prevalent and easier to use, said Virginia Long, predictive analytics scientist at MedeAnalytics. "It's finally reached a tipping point where payers and providers can both use the information in ways they couldn't before."

The key "is how we use the technology, and finding creative ways to do that," she added.


Learn more at the Healthcare IT News Big Data and Healthcare Analytics Forum in San Francisco, June 14-15, 2016. Register here.


Geospatial tools can help health organizations see variations and usage patterns, enabling them to target intervention programs or enact policy adjustments that can effect change.

The impact of environmental and behavioral health is clear, as evidenced by some recent substantial investments by U.S. healthcare giants. In April, Kaiser Permanente launched a database that enables researchers to examine participants' DNA in conjunction with environmental and behavioral health.

The goal is to gather data from 500,000 participants across Kaiser's seven regions to create one of the biggest and most diverse repositories of genetic, environmental, and health data in the world.

Researchers could then study how genetic and environmental factors might influence diabetes and cancer, for instance, or drill down into data on second-hand smoke, neighborhood violence, environmental pollution, access to healthy food and more.

"Kaiser doing this is wonderful, because I think one thing that's been overlooked is that environmental factors are important," said Long. "It gives context for people's behavior. There are cultural things you can derive from a person's location. You can also start to look at where there might be challenges like access to care and access to resources."

One of Long's current focuses is on using geospatial information to find gaps in access to care. For instance she examined childhood obesity in Mississippi, mapping where there were rural health clinics by county, even by ZIP code. And one ZIP code had a tremendously high rate. 

[Also: What is Fast Analytics, and what can your hospital learn from it?]

"You can imagine that if you took that information and added a clinic or state program there it would be super valuable,” Long explained. “You might be able to make some big changes with just a small addition."

Another of Long's recent focuses at MedeAnalytics is the addition of geospatial information to help add "another layer of understanding" to data. "Knowing where a person is gives you a lot more relevant information to help you know what's going on with the patient,” she said. “Not just their health outcomes, but what the drivers are for those health outcomes."

There's been a lot of talk lately about genetics, of course, but "the other side of the coin is epigenetics, which is location, environment and what the environment does to affect genetics," said Long.

Long's research at MedeAnalytics involves looking at clients' data and coming up with innovative use cases for how it can be put to work.

"We can have these datasets – these awesome tables that have tons and tons of information, but there's a lot of things you need to see visually in order to get a sense of patterns." she said.

Datasets and dashboards can offer a glimpse at health trends, but it can be more "poignant to see things on a map – disease clusters that don't show up to you in a table view of just lines and lines of data," said Long. "We map it, and it becomes apparent."

With geospatial data, users can derive a lot of relevant information. Moreover, mapped data helps inform tactics to reach people with information or resources for better health, she says: "The things that have been useful in marketing are going to become useful in healthcare as well. How do we make sure the right information is reaching people in ways that they can use?" 

[Also: Big data: Bold promise? Or hardest part of population health, precision medicine, patient experience?]

Another area Long has been exploring recently is overuse of the emergency department. "There are abusers, and then there are people who don't have proper access to care," she says. "Even with insurance, they use the ED in lieu of regular contact with their physician. I'm trying to understand what the location biases are in that."

On April 11, the New York Times ran a feature story titled, "The Rich Live Longer Everywhere. For the Poor, Geography Matters."

Drawing on de-identified geospatial data, studied over 15 years by researchers from Stanford, MIT, Harvard and others, the research (first published in The Journal of the American Medical Association) found big disparities in life expectancy, depending on how much money people earn – and where they live.

"The poor in some cities – big ones like New York and Los Angeles, and also quite a few smaller ones like Birmingham, Ala. – live nearly as long as their middle-class neighbors or have seen rising life expectancy in the 21st century," according to the Times. "But in some other parts of the country, adults with the lowest incomes die on average as young as people in much poorer nations like Rwanda, and their life spans are getting shorter."

The good news?

"The fact that some places have increased the lifespan of their poorest residents suggests that improving public health doesn’t require first fixing the broader, multi-decade problem of income inequality," according to the article. "Small-scale, local policies to help the poor adopt and maintain healthier habits may succeed in extending their lives, regardless of what happens with trends in income inequality."

MedeAnalytics' Long agreed: "If you can target those areas and change policies, you can change people's lives.”

Twitter: @MikeMiliardHITN
Email the writer: mike.miliard@himssmedia.com


Like Healthcare IT News on Facebook and LinkedIn