Why Big Data in Healthcare is Still Moving So Slowly | Healthcare Informatics Magazine | Health IT | Information Technology Skip to content Skip to navigation

Why Big Data in Healthcare is Still Moving So Slowly

October 28, 2014
by Rajiv Leventhal
| Reprints
Industry consultant weighs in on the cultural elements that are holding back the power of data and analytics

With the rollout of the pay-for-performance model in U.S. healthcare, healthcare leaders are now realizing that having data at their fingertips to make decisions of all kind will be necessary. From both a business and clinical standpoint, big data is a tremendous asset; specifically at the point of care, when patient care organizations apply analytics, it can offer invaluable insights to improve chronic disease management and track at-risk patient populations.

Undoubtedly, the billions of dollars that are being spent on the adoption and implementation of health IT tools means that the amount of data in the healthcare landscape will only grow. However, it should come as no surprise that healthcare is still behind most other industries when it comes to really using big data and analytics. According to Munzoor Shaikh, a senior manager in the consulting firm West Monroe Partner’s  healthcare practice, although some of the leading patient care organizations are using big data at the point of care, the industry has a ways to go. Shaikh recently spoke with Healthcare Informatics about the potential of big data to help in the clinical setting—including in the wake of the recent hospital readmissions news—and how a change in culture could end up paying dividends.

Are you seeing healthcare organizations using big data at the point of care?

I don’t think most places are doing it very well. A few are starting to, and most of that has been in hospital settings around reducing readmissions. But that is such low-hanging fruit to me and so overdue, I’d hardly call that a victory from an industry standpoint.

Taking a step back, I think about analytics as a value chain; there’s a value chain to the data that exists, and I like to break it down into four “As.” The first A is access—you need to have access to data, and this isn’t trivial. Hospitals have lots of data, but sometimes there is only access to one type—sometimes they don’t have claims data, sometimes they don’t have demographics data. I was talking to a wellness company recently that did 35 different biometric markers. Hospitals hardly have that data, so access is a critical thing. You need to know what you have access to, and thereby what type of value might be derived from it.

The second A is aggregation, and this is where most of industry’s efforts are going, though few know how to aggregate data in a meaningful way. Those that do are getting the benefit of meaningful analytics from  it. Aggregation is a big deal now, and a lot of the health plans suffer from this because they can aggregate their claims data, but can’t take lab data, claims data and clinical data, and aggregate it together very well.

Analytics, the third A, only makes sense when you have done the first two. In healthcare, things are moving so slowly in the big data and analytics world. Hospital readmissions are a cutting edge thing, but this is something big data should have solved a long time ago. And the fourth A is application, which involves the educational and engagement processes that are built around behavioral change. These four “As” help build the value and help build on each other. That’s the narrow perspective—from the broader perspective, very little is being done at the point of care, however.

Munzoor Shaikh

What needs to happen for that to change?

Well it’s all just happening very slowly, since this industry moves in slow motion in terms of data and analytics. Hopefully in the next five years readmissions will be solved, and other meaningful things will have emerged. For instance, what about chronic disease? Why should the “point of care” be the hospital? Why isn’t the point of care when I come to my office and go to the clinic? Why isn’t when I wake up in morning and track something on my Fitbit? We need to redefine the point of care and think about it as a “point of living.”

What you’re talking about would require a pretty big cultural change, right?

We have this problem in our healthcare system where we wait until we get sick and need care, and then at the point of care we need to do something. At that point, many things have gone wrong. What about all the other points of living? That’s what the big data vision really is, and you see wellness companies going in that direction. To me, that  has a quicker future, but not necessarily as a high as a return as readmissions, which people are paying more attention to because they’re so expensive.   

Another point of view is what I call a positive deviance perspective. We always talk about using analytics to figure out preventing when you’re going to be sick, but we never think about using analytics to say “you are going to be healthy.” I would like to take whatever it is you do to stay healthy—you may be a positive deviant in our system, meaning you’re not likely to go to hospital or get a chronic disease—and replicate your lifestyle to the rest of my population.  Other countries do this well, but in the U.S. we are more focused on the sick than the healthy, and we want to learn from the sick rather than learn from the healthy.

Is patient-generated health data (PGHD) a step in that direction?

I think we’re starting to scratch the surface on PGHD, and maybe even a little beyond that. Wellness companies are now providing biometric screening data with the ability to say that “If the trend goes like this then you are probably going to become a cardiovascular patient.” So you can identify those possible cardiovascular patients beforehand. But it’s still early on and it might take several numbers of years to mature. Of the four “As” I mentioned, the real challenge is with application, because while the first three As were mostly science, application is a science as well as an art. That’s where we will have an adoption challenge in our population.

Going back to readmissions, can analytics help bring down the staggering numbers that were recently reported?

Everyone has some analytic model and some sort of science going, though it’s not perfect. Where they are failing is the clinical workflow and pathways on the application part. So many fines are happening, and with the carrot-and-stick method, it seems like the stick approach is what we’re resorting to in order to get people to do something about it. That’s how we respond—more to the sick than to the healthy. That positive deviance mentality doesn’t exist within our system—“nothing is wrong with me until it hurts.” It’s part of the culture we live in. I do think that with the approach we use in healthcare, you will see some reductions as a result. The motivation wasn’t quite there before, so now that should change.

Once you identify high-risk populations, it’s about education. You need to know to go home and take those medications or else you will come back and continue suffering. I sit on the board of non-profit and we had a case of someone who went to the ER 42 times in one year. When we looked at it, it was because he didn’t have any housing. We happen to work with the homeless/Medicaid population, so the solution was to give him a home, a shelter. So we provided that through an alliance we had, and the next year he went to the ER twice. That’s an anecdotal readmissions solution problem. The cost of providing free home for a year is fraction of cost compared to 42 ER trips.

Does big data have the ability to replace expensive clinical trials?

I think it has the possibility to do so. This is a $70 billion industry with 30 percent waste. No other industry is like that, having such a high percentage of waste. There are lots of opportunities, but those are probably a little further in the future when it comes to doing predictive analytics on biomolecular systems. In the oncology world, I actually have heard some stories about this, regarding chemo patients. So there is promise there.


The Health IT Summits gather 250+ healthcare leaders in cities across the U.S. to present important new insights, collaborate on ideas, and to have a little fun - Find a Summit Near You!


/article/why-big-data-healthcare-still-moving-so-slowly

See more on

betebettipobetngsbahis