News Stay informed about the latest enterprise technology news and product updates.

Challenges remain for mining healthcare big data to patient benefits

Healthcare big data could address many of the industry's problems -- if we could only agree on its definition.

Big data is a term often used, but frequently misunderstood, making it difficult for healthcare providers to find meaningful applications of the concept to improve outcomes, make workflows more efficient and cut costs.

As big data teaches us more about what patients want and how they want it, big data will ensure that we will remain relevant to the patients we serve.

Alistair Erskine, Geisinger Health System chief clinical informatics officer

To Beth Israel Deaconess Medical Center Chief Information Officer John Halamka, M.D., healthcare big data is all about bringing data into a platform to answer a specific question. Speaking at Health Care Innovation Day 2014, jointly sponsored by ONC and the Gary and Mary West Health Institute, Halamka added that the "mythology" of big data suggests it will be able to answer every question, instantaneously. However, it is a much more useful concept when applied in a more specific manner.

For example, Halamka said his organization is using big data analytics to support its accountable care ambitions. Beth Israel is using analytics to spot patients who are experiencing gaps in care, which may include patients who never scheduled recommended follow-up visits or who have not received appropriate preventive care. An employee can then intervene.

These sorts of strategies are important in an accountable care world, Halamka said, because just sitting back and waiting for patients to contact their doctor is a financial liability. This type of specific application of healthcare big data can be useful.

Geisinger: Big data can tailor care to a patient

Alistair Erskine, chief clinical informatics officer at Geisinger Health System, said healthcare organizations should use clinical data to personalize patient care; providers who don't risk losing credibility among their patients.

"A fragmented, cottage industry designed for the needs of its staff has a strong potential to be leaped," Erskine said. "As big data teaches us more about what patients want and how they want it, big data will ensure that we will remain relevant to the patients we serve."

Some challenges make it hard for organizations to embrace big data and put it at the center of decision-making processes. Erskine pointed out that providers still have incentives to hoard data for competitive reasons, which keeps data siloed. Moreover, the majority of clinical data that providers generate remains in unstructured, and therefore mostly unusable, forms. And then there's the privacy issue: Providers will have to find ways to analyze data, while at the same time respecting patients' rights.

How does big data fit in healthcare?

Fragmented healthcare industry challenges use of big data

Big data tips: How to draw value from analysis

Step one in a big data plan should be an analytics strategy

Big data can relieve physician workloads

If providers fail to overcome these obstacles, current negative trends could worsen. Gregory Moore, Geisinger's chief emerging technology and informatics officer, said doctors are burned out. He said the average Geisinger family physician has about 2,500 patients. If each doctor were to perform every recommended test, procedure and follow-up consultation for every single patient, doctors would have to work 18 to 20 hour days, he said.

"We're at human limits," Moore said. "We have a culture now where we're ready for technology innovations to come in and help."

Once the use of big data in healthcare becomes the norm, it will make things like defining appropriate treatment bundles for populations and automating recommended treatments easier, Moore said. This will help eliminate persistent care gaps by ensuring every patient receives appropriate treatment, while also alleviating some of the burden on physicians.

"It's our responsibility to take the data and put it into action," Moore said.

Ed Burns is site editor of SearchBusinessAnalytics. Email him at and follow him on Twitter: @EdBurnsTT.

Dig Deeper on Clinical data analytics software and systems

Start the conversation

Send me notifications when other members comment.

Please create a username to comment.