UPDATED 16:27 EDT / DECEMBER 04 2013

Seattle Children’s Hospital turns to Brightlight, IBM to liberate its data

Hospital IT shops live in a world of their own, dominated by systems that are often the opposite of open and where combining data from the patient care and financial management systems is a major challenge. That is exactly what the Seattle Children’s Hospital, one of the top 10 children’s hospitals in the country according to US News and World Report, discovered when it tried to create a data warehouse combining the data from 10 central operational systems including its Lawson hospital financial system and its Cerner and Epic patient management systems. All the hospital’s management wanted to do initially was ask basic questions of the data such as “what is the average length of stay for patients with medical condition X, what is the readmission rate, and what is the average cost of treatment?”

In most industries today this would not sound like a major challenge. But, says David Overcash, CTO and cofounder of Brightlight, a big data analytics consulting company that Seattle Children’s turned to after two major project failures, hospitals are a different world. Cerner, Epic and Lawson in particular all use proprietary data schemas that make extracting the data and converting it into a format that will support standard data analysis tools very difficult.

A long quest

 .

Brightlight did create the solution using IBM PureData for Analytics, and today the hospital has a revolutionary tool for increasing quality of care, supporting medical research, and reducing cost. But getting there was an agonizing journey. The hospital initially decided to install Microsoft’s Amalga, a product that has since been withdrawn from the market. Theoretically Amalga was a data warehouse that would support sharing of HL7 healthcare data among systems and analysis on the combined data. In fact, it was never able to convert the proprietary data formats and tables to support analytics. “It was real easy to get the data in, but very difficult to get data out, but of course they sold it as the silver bullet for everything,” Overcash said.

After two years Seattle Children’s “threw that system out”, but it was left with a set of underlying SQL hardware and software environments. So it decided to build a Microsoft-based data warehouse with SSIS and SQLserver reporting services. The problem with this system was that it kept growing until it was running on more than 30 servers.

“They had a maintenance nightmare moving data and code between all those servers,” Overcash said. “All they were doing was maintaining this infrastructure and not building any new capabilities.” And the entire database was only about five terabytes.

A Brightlight shines

 .

When Brightlight was called in, Overcash said, “we cost justified the system we built for them just on the basis of getting rid of those 30 servers. Then we said, ‘What if we take your nightly ETL process and instead of having it finish at 10 a.m. we can finish at 6 a.m.?'”

image via IBM

Brightlight picked IBM PureData for several reasons. First, medical providers value proven dependability and vendor stability. Lives can literally depend on these systems, and the hospital needs to know they will work 24/7. “When we really looked under the covers of systems like Vertica or Greenplum, they couldn’t give us a real, named client that was running mission-critical analytics 100% on their systems,” Overcash said. “We started to get down to a big vendor tier system pretty quick.”

The hospital also wanted simplicity. It wanted to liberate its staff from the demands of running a complex system to focus on development and supporting research. “There was less overhead required and greater operational simplicity with PureSystems than any of the other environments,” Overcash said. Today Brightlight runs the PureData system remotely for the hospital for about $4,000 per month, leaving the internal staff free to support increasing amounts of data analysis.

Third, “the ability to do ETL processing and data integration of all these disparate systems in healthcare was real important to them, and the PureData environment provides the best in-database processing. We can keep 100% of that processing in the database. That gave them some short-term benefits and will certainly give them some long-term benefits as well.”

Finally, capital is always in short supply in a hospital, and the ability to start small, size the system to the immediate need, and grow it over time and needs expand is a major attraction. Also, says Overcash, while “they didn’t go out and buy a bunch of other IBM products right away, I think from an executive level knowing they were getting into a platform with IBM that is thinking of a smarter planet and other initiatives, as a long-run play they would be in alignment if they want to do more than research data.”

  • Stage 2

Today the hospital is doing increasing amounts of data analysis including medical research with its new PureData data warehouse. But, Overcash says, this is only the first step. The next is to bring data down to the operational level to provide the best treatment customized for each individual patient at the bedside, the equivalent of the data-driven business. While evidence-based medicine has been a generally accepted concept in treatment at the general guideline level — for instance a step-by-step plan for optimal treatment of each disease based on medical data — hospitals in general have yet to grasp the vision of putting data analysis into the hands of each medical professional and business manager.

“We’ve spent a couple of years getting them through this point…. Now we’re saying, ‘You need to put some of these analysts out in the lines of business. Get them closer to the action,'” Overcash says.

Meanwhile, the hospital is happy with what really is a solution to its basic problem. “Working with Brightlight and using the IBM Big Data solution, we are able to understand the hospital’s ‘Big Data’ — the thousands of data points associated with each child — immediately, as needed,” said Wendy Soethe, manager of the enterprise data warehouse at Seattle Children’s Hospital. “As we get deeper into data we’re actually able to see new commonalities in treatments so we can better identify new protocols, make improvements in care, and ultimately make a difference in the type of care children are receiving.”


A message from John Furrier, co-founder of SiliconANGLE:

Your vote of support is important to us and it helps us keep the content FREE.

One click below supports our mission to provide free, deep, and relevant content.  

Join our community on YouTube

Join the community that includes more than 15,000 #CubeAlumni experts, including Amazon.com CEO Andy Jassy, Dell Technologies founder and CEO Michael Dell, Intel CEO Pat Gelsinger, and many more luminaries and experts.

“TheCUBE is an important partner to the industry. You guys really are a part of our events and we really appreciate you coming and I know people appreciate the content you create as well” – Andy Jassy

THANK YOU