Skip to content Skip to navigation

Finding the Value in Big Data

May 14, 2014
| Reprints
Gaining insight by mining data from routine doctor visits

The term “Big Data” is everywhere in healthcare, and is often viewed as the key to improving outcomes while reducing costs. Yet some caution that Big Data I not a magic bullet that will translate to better healthcare. It is possible to be data rich but information poor, according to Richard Averill, senior vice president of 3M Health Information Systems. The real value of data is in providing real-time information—the right information at the right time—in specific applications that enable better decisions, he says.

Bill Phillips, senior vice president and CIO, Information Services, at University Health System in San Antonio, Texas, agrees. In a recent interview that touched on data analytics and its ability to improve outcomes, he described Big Data as a tool for improvement: “If you are not taking Big Data and analyzing it and training it to improve the outcomes for your patients, then you might as well not have Big Data—it’s of no value.”

Phillips said that the role of analytics has been around financial systems for a long time; in the clinical arena, it is just starting to heat up. One area of potential is improving population health. He notes that his geographic area, San Antonio, has a large diabetic population. “What can we do with outcomes and Big Data information to change the wellbeing of the diabetic?” he asked. “Can we trend people with high A1C, drill down into that and find a way to lower their A1Cs? That’s what we need to do with Big Data.”

There was an interesting article in the May 13 Wall Street Journal, about how some medical researchers are analyzing pools of patient information collected from routine check-ups, to help physicians better diagnose their patients. In one example, researchers looked at data from patients with sore throats and came up with a way to determine whether should seek care for a strep throat test or stay home and take aspirin.

As explained in the WSJ article, Kenneth D. Mandl, M.D. of Boston’s Children’s Hospital is one of the authors of a 2013 paper in Annals of Internal Medicine that examined the records of patients who visited a chain of retail health clinics with a sore throat. His research team used data from 71,776 patients over age 15 with sore throats who visited a chain of health clinics over a 15-month period, to create a score to help figure out which patients did not have to come in right away for strep testing. Uniform data was gathered and collected in an electronic health record. According to the study, an estimated 230,000 unnecessary doctor’s visits could be avoided annually for those with a low enough strep throat risk to stay at home.

There are other examples in the article as well, and it’s exciting that Big Data is no longer just a buzzword: researchers are learning to mine the huge volumes of data that are collected during medical visits, and are making a difference in patient care in very real ways.