The Rise of Big Data in Hospitals: Opportunities Behind the Phenomenon

In and around the healthcare industry there has been a lot of talk about "big data" — very large sets of complex data that become difficult to process using database management tools. Big data is emerging in the industry because hospitals and health systems are collecting large amounts of data on patients every single day. The data comes for a variety of settings — clinical, billing, scheduling and so on. Unfortunately, in the past, a lot of that data was not leveraged to make patient care and hospital operations better. Recently though, there has been a shift to change that.

The rise of big data
According to Joel Dudley, MD, director of biomedical informatics for The Mount Sinai Medical Center in New York City, healthcare organizations are realizing that all of their data can be captured and leveraged as a strategic asset.

"Big data is not just about storing huge amounts of data. It's the ability to mine and integrate data, extracting new knowledge from it to inform and change the way providers, even patients, think about healthcare," says Dr. Dudley.

Anil Jain, MD, CMIO of Explorys, a healthcare analytics company, and former senior executive director of information technology at Cleveland Clinic, calls the abundance of data in healthcare a "perfect storm." His description is fitting as a variety of factors have converged in the past few years, increasing the amount of digitized healthcare information:

• The federal push for electronic health records has increased the number of hospitals and providers who use them, subsequently increasing the amount of electronic data generated.
• Newer reimbursement models and accountable care organizations need large amounts of information to be analyzed in order to more accurately understand what occurs with patients.
• New technology in general, including devices, implants and mobile applications on smartphones and tablets, has increased the amount of data available to providers.

In addition, there is a lot of pressure to become evidence-based and predictive with healthcare services — to leverage historical data and to create predictive models, says Mr. Dudley.  

Dr. Jain agrees, saying that without movement to aggregate, manage and analyze big data, the healthcare industry would be in information overload. For this reason, many hospitals and health systems have begun big data initiatives in their information technology and informatics departments.

"[At Mount Sinai], we have a mandate to work together to leverage informatics to leverage the future of our healthcare services. We are invested in [big data]," says Dr. Dudley. "We are putting power behind informatics, big data and predictive modeling because it will be a big part of healthcare going forward. [The administration] is enabling us to start the dialogue and sit down with clinicians."

Mount Sinai is not alone in this sense. Providers, patients and the entire healthcare industry may realize a variety of beneficial implications from the use and analysis of big data.

1. Better point-of-care decisions. Big data will change how physicians take care of patients at an individual level, fostering more personalized support right at a patient's bedside, says Dr. Jain.

"The analysis to deal with big data can produce valid and relevant data that is more current, which gives physicians the means and motivation to make the right decisions at the right time," says Michael Corcoran, senior vice president and chief marketing officer of Information Builders, a business intelligence and software solutions company.

For instance, NorthShore University Health System in Evanston, Ill., has seen the impact at the point-of-care through predictive modeling. As a result of its large data sets, the health system has developed models to identify which patients are likely carriers of a dangerous microorganism, Methicillin-Resistant Staphylococcus Aureus. By implementing the results of that modeling into its EMR, providers within the health system will receive alerts when a patient is admitted that meets the characteristics of being a high-risk carrier of MSRA, as determined by the predictive model.

"We've found that we can use the models to identify about 90 percent of MRSA in our patient population," says Ari Robicsek, MD, vice president of clinical and quality informatics at NorthShore.

NorthShore has also used this modeling to predict which patients are likely to develop Clostridium difficile. "The hope is that with these models we will be able to perform special interventions on those patients at the highest risk of developing infection," says Dr. Robicsek.

2. Reduced readmissions. Big data also provides predictive models for the likelihood of readmission within 30-days, which is another area NorthShore is targeting with its big data and informatics work, according Dr. Robicsek.

"We will put data from the EMR into our enterprise data warehouse, which serves as a computation engine. We compute a patient's risk of being readmitted in 30 days and then feed that data back into the EMR," says Dr. Robicsek. "A user can look at a panel of patients to see which patients are at risk — high, medium or low — of being readmitted in 30 days."

According to Dr. Robicsek, NorthShore uses a similar predictive model to identify patients who were recently discharged from the hospital with a high risk for readmission and then sends messages to primary care practices. The messages alert the patients' primary care providers of their high risk and if they have any follow-up appointments scheduled.

"The practice could use the patient list to say 'here is a high-risk patient with no follow-up scheduled. Let's reach out to that patient and make sure we get them in for an appointment.' We have already noted a substantial reduction in readmission rates," says Dr. Robicsek.

Without large data sets showing trends and patterns in huge groups of patients, this type of accurate predictive modeling would not be possible.

3. Population health management. Big data also informs population health management as findings from predictive models can be shared with providers across the care continuum. According to Dr. Jain, big data offers providers the ability to use information and discover patterns in patient populations that may not have been possible before. The implications for better, coordinated and specialized care are endless.

4. Research advancements. Big data also advances clinical research towards new knowledge discovery quickly and efficiently. According to Dr. Jain, when large amounts of data are available, research does not become less traditional; rather, it becomes more meaningful.

"There was an old adage that it used to take around seven years for something to that was discovered in research to be applied in the care of patients. It's called bench to bedside," says Dr. Jain. "Lately, I've been using the term: bench to bedside to bottom line. The paradigm has shifted. The discoveries we are making with big data are informing decisions at a rate where the potential outcome — the bottom line — is more readily apparent."

For instance, Dr. Jain was part of a study published in the Journal of the American Medical Informatics Association that replicated a study conducted in the Netherlands. Due to big data capabilities, Dr. Jain and his fellow researchers found similar findings with 10 times as many patients in a much shorter time period.

5. Operational improvements. According to Mr. Corcoran, big data can have an operational impact on hospitals as well because it provides hospitals and their staff performance metrics with which to compare operational efficiencies.

"When individuals can see their performance ranked among many others, it gives them more motivation to achieve better results. They have comparable, daily, real-time results, pushing them to deliver on those same levels," says Mr. Corcoran.

In this sense, big data can have an impact from the back office all the way to inpatient care; rather than merely collecting the data, hospitals can analyze and operationalize it to inform decisions at a variety of levels within the organization.

"I've seen a dramatic difference in the time it has taken a nurse to insert an IV into a patient from one floor to another due in part to an awareness of performance metrics. Big data makes performance data even within a hospital more accessible so that efficiency and accuracy can increase," says Mr. Corcoran.

Big data is emerging in the healthcare space, and it is likely that it will continue to magnify over time. According to Mr. Corcoran, healthcare organizations are going to keep collecting massive volumes of data, so aggregating and analyzing that data will be a continual challenge. However, that effort will be worthwhile as we begin to see the implications big data promises.

"For me, big data is about how to sift through the data to convert it to useful information — a more usable format and with the right visualization," says Dr. Jain. "At the end of the day it is about making sure providers can do the right thing for the right patient at the right time. We are trying to improve healthcare for everyone."

As the healthcare industry tries to improve the cost and quality of services for patients throughout the United States, healthcare organizations need to embrace the big data challenge. "Our institutional understanding has been that we need to be in a place where we are using this data to be competitive in the future and to provide the best care we can. Using the data has been an organizational imperative. There has never been the consideration of giving up," says Dr. Robicsek.

More Articles on Big Data:

Should Hospitals Use Automated Software to Handle 3 Big Data Issues?
4 Steps to Leveraging "Big Data" to Reduce Hospital Readmissions
Paul Levy: Get the Data and Then Use it Right

Copyright © 2024 Becker's Healthcare. All Rights Reserved. Privacy Policy. Cookie Policy. Linking and Reprinting Policy.


Featured Whitepapers

Featured Webinars