Posted by: Jenny Laurello
Big data, Data analytics, EHR
Although slower than other industries, healthcare organizations have begun to embrace big data strategies in an attempt to become data-driven entities. And it’s no wonder — as time goes on, providers are facing more and more healthcare data.
Electronic health records, accountable care models and new technologies
First, the Medicare and Medicaid Electronic Health Record (EHR) Incentive Programs, funded by the American Reinvestment and Recovery Act in 2009, served as a catalyst for increased implementation of electronic medical records both in the ambulatory and inpatient setting. Recent statistics from the Office of the National Coordinator (ONC) indicate that EHR adoption has doubled over a one-year period, yielding significant data from EHRs.
Second, emerging reimbursement models and payment reforms (e.g. accountable care organizations) encouraged in the Affordable Care Act of 2010 have made it important for healthcare organizations to be able to quickly examine multiple pieces of data. The goal has shifted to maximizing quality and minimizing cost of the care, all the while maintaining the patient experience. Moreover, the alignment of physicians with hospitals and hospitals with other hospitals often requires integrating disparate data in order to get a full picture.
Finally, there has been an explosion of smart medical equipment (e.g. beds, intravenous pumps, telemedicine, implantable devices, patient portals, and imaging) that yield more and more data. As this voluminous data comes to a healthcare system with velocity and variety, traditional storage databases may not be enough to manage it, which is why organizations need a big data strategy.
Meanwhile, studies in the pre-personalized medicine era found that clinicians typically utilize two million pieces of information to manage patients. How can we sustain that, as even more data becomes available?
Making data useful
It seems storing and retrieving data is only the initial necessary component of a big data strategy. The other is the ability to rapidly analyze and present actionable information gleaned from big data. Just ask any clinician how challenging it is to find the actual recommendation from a typical progress/consult note (i.e. data) in an EMR.
Thus, the millions of beams of varying colored light that represent big data need to be strategically condensed to a few laser beam focused prescriptive analytics that tell the healthcare provider what must be done at what time for optimal care. For example, a heart failure readmission prediction model that incorporates hundreds of clinical and administrative pieces of patient-specific data to generate a recommended action at the time of admission or discharge would be more sustainable than a screen full of clinical data. Another example would be to combine data from a variety of sources such as socioeconomic, adherence, compliance and clinical data to intervene only on appropriate patients in a resource sensitive care management program.
A successful big data strategy puts very little data in front of clinicians in exchange for prescriptive analytics backed up by actionable information.
Before coming to Explorys, Dr. Jain spent 16 years at the Cleveland Clinic, leading several health IT innovations, including programs to support research and quality informatics and creating interactive dashboards to monitor the meaningful use of electronic health records.