Big Data or Big Hype?

August 23rd, 2012 / By Kristine Daynes

“Big data” is a collection of data so large that common database tools cannot easily manage it. Imagine a wilderness of datasets, endless rows and columns of data points as yet unexplored and untamed.

It sounds adventurous. Google tells me that big data can help me drive outcomes and spark innovation. With the help of advanced analytics, I can harness the digital universe and unlock big data’s hidden value. The rush of metaphors makes me dizzy.

Is it big data or big hype?

I asked Jason Mark, the master black belt for the Lean Six Sigma program at 3M Health Information Systems. “There is a lot of hype,” he said. “Big data won’t solve your problems any more than cloud computing or an EHR. But it can make you better informed and give you more information to address and improve performance.”

Big data describes the increasing volume, velocity, and variety of data gathered by information technology. This data is an asset to organizations that can use it to improve their productivity and competitiveness.

In the healthcare industry, big data could be used to drive cost-effective care and patient safety. A 2011 McKinsey report estimates the value of big data for health care at $300 billion every year, “two-thirds of which would be in the form of reducing national health care expenditures by about 8 percent.”

The challenge is that big data is so, well, big. Organizations can now have cloud-based solutions for centralized data fed by an EHR, RFID and mobile devices, and data exchanges. Big data has implications for how data is captured, stored, searched, and shared. It can be mind-boggling to those who try to query the data and interpret the reports.

When it comes to using big data, Mark suggests a disciplined approach, especially for HIM, case and quality managers who are now presented with all this information. “Be methodical,” he says. “Have control. Know what you want to change. There are already elements of that in change management and quality control methodologies.”

Mark explains that the real meaning in the data isn’t always your first impression when you see a report. Big data takes time to digest. Rather than make massive changes at once, Mark suggests that you make “informed adjustments” as you come to better understand the implications of the data.

“Change one or two things at a time and see what happens,” Mark says. “Pick top priorities. Choose metrics that are right for your facility. Dig in and understand those. Make iterations of improvement.”

Some hospitals are using big data for performance improvement projects such as comparing readmission rates by physician or improving treatment protocols. But big data is a benefit to health care in less obvious ways, too. Most HIT vendors, 3M included, are using big data to improve the performance of their software and systems.

Big data, combined with machine learning, enables software to refine its algorithms automatically as it is used. “Big data could mean dashboards and reports,” says Mark. “It could also mean that as a medical records specialist is coding one record, 3M is giving more intelligence in the software … because we have more data on the back end. As our software becomes informed by big data, end users become consumers of it.”

That makes sense to me. As a Google user, I’m already a consumer of big data. Google uses machine learning to refine its search algorithms to present more relevant search results. That’s why, when I entered “big data” as a search term, Google returned so many results related to health care. It has learned from my earlier searches and clicks. Now big data will do the same for healthcare software.

Kristine Daynes is a Senior Marketing Communications Specialist for 3M Health Information Systems.