As we have said here before, any company of any size in any industry can benefit from big data analytics, but perhaps it is the healthcare industry that has the most to gain. Hospitals generate a lot of data for patient care, drug administration, operations, insurance billing, regulatory compliance and a host of other requirements, and all that data can be used to improve the quality of care and operations. Big data analytics seems made for healthcare, and there are dozens of use cases that deliver a high ROI for any medical practice.
Consider, for example, the explosion in electronic medical records (EMRs). Before the implementation of the Health Information Technology for Economic and Clinical Health Act (HITECH), less than 30 percent of physicians and healthcare practitioners were using EMRs. Today, with Medicare penalties in place for HITECH noncompliance, the adoption rate is significant and is expected to continue to grow at a rate of 13.4 percent. The boom in EMR adoption is creating a data glut in healthcare, and because caregivers need to maintain historic patient records, the amount of data is only going to increase.
That’s good news for solution providers selling data storage, and it’s even better news for big data analytics. EMRs can reveal trends in patient care, epidemiology, treatment effectiveness, operational effectiveness, vendor costs and much more. It’s all a matter of finding the right use case for the data.
Here are just seven examples of big data use cases that have real value for healthcare providers:
The Precision Medicine Initiative calls for medical practitioners to apply research and centralized data to promote personalized patient care. President Obama has a vision to establish a national patient databank to facilitate personalized treatment and promote genome research. Even without a national databank, hospitals can apply the same principles to provide personalized care using big data analytics to correlate patient data stored in EMRs to national trends and other data sources. Predictive analytics, for example, can promote preventative care for heart disease and obesity.
Cookbook medicine has been the norm, using the same battery of tests to diagnose by ruling out the cause of illness. With evidence-based medicine, doctors correlate symptoms to narrow the diagnoses. Beth Israel Deaconess Medical Center in Boston, for example, is using patient data from two million patients to provide data points for diagnosing via a smartphone app. In order to facilitate analytics, physician notes are being encoded to standardize references; for example, “high blood pressure” and “elevated blood pressure” are coded in the same way to make data searchable.
Better safety practices
Predictive analytics also promotes quality care and patient safety. In the intensive care unit, for example, patients are prone to a sudden downturn due to sepsis or other infections. Sepsis alone has a 40 percent mortality rate and is difficult to detect for early treatment. The University of California, Davis, has used EMR data analytics to create an algorithm to provide an early warning of sepsis infection. In another example, the University of Iowa Hospitals and Clinics has used predictive analytics to reduce postoperative infection following colon surgery by 58 percent.
Population health management
Just as analytics can be used for predictive care for individual patients, the same methods can be used in epidemiology. Duke University, for example, is mapping EMRs to geographic information system data in order to identify healthcare trends in specific geographic areas. By mapping the right data sets, it’s possible to predict specific ailments—such as influenza—that will escalate in specific areas, making it easier to strategize diagnostics and plan for stocking serums and vaccines. The National Institutes of Health has been working on similar strategies for some time to predict disease outbreaks.
Readmission rates are a chronic problem for hospitals, especially with patients who return within 30 days of treatment. Big data analysis of EMRs reveals trends that highlight those patients likely to need additional treatment to prevent readmission. At a Texas hospital, for example, EMR analytics led to a drop in the readmission rate of cardiac patients from 26.2 percent to 21.2 percent by identifying high-risk patients.
Critics worry that patient records are a prime target for cyber thieves, because they yield personal information that is much more valuable than credit-card data, such as Social Security numbers, Medicare information and prescription information. Proponents point out that big data analytics can actually be a valuable resource in securing medical records by identifying changes in network traffic or behavior that indicates a cyber attack.
Medical insurance is complex and plagued with disputed and fraudulent claims. Big data analytics helps streamline the efficiency of medical insurance claims by revealing claim trends and streamlining claims processing. Patients get better returns on their insurance claims, and caregivers receive payment faster.
These are just a few of the ways that big data analytics is having an impact in healthcare. Anywhere there is a data-driven process, big data analytics can be applied to improve patient care and health center operations. And new technology—such as radio frequency identification (RFID) chips—is delivering more data for potential analysis. New data sources provide new opportunities for big data analytics, and it’s up to solution providers to show healthcare professionals how to unlock the power of that data with new analytics.