How community crowdsourcing can help accelerate healthcare analytics advances

Benson Hsu, MD, vice president of data and analytics at Sanford Health, explains how sharing data with 'smart people' from outside the health system has led to some innovative applications for population health management.
By Mike Miliard
04:27 PM
Share

In a healthcare industry where data is a valuable strategic asset – often zealously hoarded and protected – Benson S. Hsu, MD, vice president, enterprise data and analytics at Sioux Falls, South Dakota-based Sanford Health, has lately been taking a much more liberal approach to sharing data sets with other organizations the community.

Sanford, a sprawling integrated delivery system based the upper Midwest (but with clinics as far away as California and Oklahoma), shares certain clinical, claims, financial, and operational data with outside researchers – who then put it to work developing analytic apps that in turn help the health system with quality improvement, cost efficiencies, patient engagement and more.

At first, Hsu's bosses were incredulous of this plan, he said: "You want to give data out, and when you discover something you want to give it to our competitors?"

"I said, 'Yes, but it's the right thing to do. More importantly, it's the right thing to do for the community – and the community is going to recognize that Stanford Health is here for the community," he said.

"Secondly," he added, "it's innovation. Innovtion in our backyard, based on our population, our social determinants, our disparities."

Plenty of smart people work at Sanford, of course. But plenty of other smart people – with skill sets separate from those at the health system – work elsewhere.

[Data governance: Make analytics an 'amazing asset']

By enlisting researchers at South Dakota State University (mathematics), Dakota State University (computer science/Informatics), University of South Dakota (business), University of North Dakota (public health), Hsu said he's been able to harness their brainpower to enable innovations that might not otherwise have been discovered.

"We asked for help," he said.

On the way to enabling that sort of sharing, however, Sanford had to overcome some structural and cultural challenges, he said.

Having grown quickly over recent years with the addition of dozens of clinics in sometimes far-flung locales, Sanford has hamstrung by data silos (multiple customized versions of the Epic EHR, multiple financial accounts, multiple HR systems); analytic silos (with data crunched for everything from clinical decision support to IT reports to HR/health plan analytics); and language barriers (no common data terms, no common analytic calculator, no common benchmarking tools), said Hsu.

But about a year of focused work to reshape those disaparate areas into a robust analytics foundation, Sanford was able to arrive at the oft-elusive "one source of truth" – something that quickly "became our calling card," he said.

"We wanted to achieve three things, foundationally, and we accomplished all three in one year," said Hsu. By building a common team speaking a common language with a common data source, Sanford was well on its way, with help from external partners, to tackling some infamous population health management challenges.

By using that unified and crowdsourced approach to analytics, the health system has been able to make inroads with predictive risk, chronic disease management, diagnostic testing and technology utilization and more.

The key? "Placing researchers at the forefront of healthcare innovation with timely data and an opportunity to inform, innovate and drive the way healthcare is provided to our communities."


  Population health will be among the topics at the HIMSS and Healthcare IT News Big Data & Analytics Forum in Boston, Oct. 24-25. What to expect:

⇒ Charlotte hospitals analyze social determinants of health to cut ER visits
⇒ Big Data: Healthcare must move beyond the hype
⇒ Tips for reading Big Data results correctly
⇒ Small hospital makes minor investment in analytics and reaps big rewards 
 MIT professor's quick primer on two types of machine learning for healthcare
⇒ Must-haves for machine learning to thrive in healthcare


Like Healthcare IT News on Facebook and LinkedIn