Can three Ohio healthcare heavyweights tame big data?

If Cleveland's biomedical powerhouses—Cleveland Clinic, University Hospitals and Case Western Reserve University's medical school—wanted to make meaningful progress with using big data in advancing their research efforts, they knew they had to play nice with each other in the city's sandbox.

As such, the three recently locked arms to launch the Institute of Computational Biology — an effort its founders say will allow them to take advantage of the hulking amount of clinical information they've each collected over the years. The difficult task at hand is taking clinical data housed within the hospitals' individual medical record systems, stripping it of sensitive personal information and converting it into a format that can be digested by researchers at all three institutions.

It's not a cheap undertaking, either. Officials at Case Western Reserve, which is taking the lead on the project, said the three institutions contributed a combined $21.5 million to the institute. The university declined to break down each institution's individual contribution, but officials stressed none could launch an initiative like this on its own.

“Big data is costly,” said Dr. Pamela Davis, dean of Case Western Reserve's School of Medicine. “There's a certain economy of scale of being able to do this together.”

The three institutions hope Dr. Jonathan L. Haines, a rock star of sorts in the genetics and biostatistics arenas recruited as the institute's director, can kick the effort into gear. Since 1997, Haines has been with Vanderbilt University in Nashville, where he helped launch the university's Center for Human Genetics Research and a biological repository linked to Vanderbilt's medical records database.

“For Dr. Haines, it was time for a change and something bigger and grander,” said Paul DiCorleto, chair of the Clinic's Lerner Research Institute. “He wanted to make something happen in Cleveland. When you put together UH, the Cleveland Clinic and their patient numbers, he can have a great impact with his data analysis.”

The Cleveland Clinic identified the ability to manage big data effectively as one of its top 10 medical innovations in 2012. However, despite all the talk in recent years about the concept, healthcare institutions have been slow to make inroads in harnessing big data, according to Haines.

Other industry sectors, such as retail, already have embraced the concept by taking large amounts of consumer data and identifying trends. Haines singled out as an example those supermarkets that track spending and send customers coupons based on their purchasing history.

“We've been slow to appreciate that in the healthcare industry,” Haines said. “We have all of these data out there that we've been collecting, and no one has been looking at that in a big way. That's what we're trying to do.”

Of course, the big data push in the healthcare space is more complex and costly than Giant Eagle's Fuel Perks program. Healthcare researchers are dealing with far larger data sets that, in terms of sheer size, can balloon into exabytes. (An exabyte is the equivalent to more than a billion gigabytes.) Also, bioinformaticians — those charged with the data's heavy lifting — tend to be among the most highly paid in the medical field.

Privacy also has become a concern, particularly in big data ventures that can span across several different hospitals, according to a recent report from McKinsey & Co., a global consulting firm with offices in Cleveland. McKinsey also noted that even sharing data between departments within a single hospital can be tough.

Haines, in a sense, will be the man charged with untying those knots. He envisions building an infrastructure where researchers can “relatively painlessly” come in and answer questions with the data, all in the spirit of improving healthcare.

“Our goal is to mine big data to try to improve healthcare in every possible way,” Haines said. “There's a tremendous amount of data collected every day through the electronic health records that's just sitting there. There's a tremendous amount of information in there that we can pull if we're smart about it.”

Over time, the Institute of Computational Biology could go well beyond UH and the Clinic, as Haines said he expects other hospitals to become involved. The institute already is in discussions with MetroHealth, the hospital subsidized by Cuyahoga County, about using its data for the project.

The convergence of Cleveland's largest hospitals with its lauded medical school was a big reason Haines was lured from Vanderbilt.

“Both institutions felt this was so important that they would get together and support this equally,” Haines said. “They both have investments in this, so they're both expecting success. They're going to help make that happen.”

The new institute isn't the first time the three institutions have joined forces. Case Western Reserve and University Hospitals founded the Case Comprehensive Cancer Center in the late 1980s, and the Cleveland Clinic joined the initiative in 2003. The center has nearly 350 scientists and physicians and is one of just 41 nationwide to hold the National Cancer Institute's designation as a comprehensive cancer center.

That sort of collaboration served as the model of sorts for the new institute.

“If we wanted to get to the point where we could lead the field, we needed to find somebody of Jonathan's stature,” said Dr. Fred Rothstein, president of UH Case Medical Center. “To do that, we had to put in a program that gave him a broad opportunity to be successful.”

Can three Northeast Ohio health care heavyweights tame big data? originally appeared on the Crain's Cleveland Business website.