The following content is sponsored by Sandlot Solutions.
Digitizing each aspect of the healthcare industry requires time and space for growth, adoption and evolution before truly taking root. While it could be another two or three years before big data really gets its legs, decisions that healthcare providers make now about the collection, storage and analysis of data will affect their ability to harness the power of big data in the long term.
Big data defined
Many people tell me that even after reading a number of articles on the subject of big data, they walk away uncertain of what the term really means, as there seem to be many interpretations and nuances. My definition of big data is fairly straightforward: It's scalable, streaming data in many forms (structured, unstructured, text, multimedia, etc.) that can be accessed for real-time decision making. Big data also refers to the next evolution of how we can capture, absorb and interpret vast amounts of data accurately and efficiently. Historically, hospitals owned the majority of this clinical and financial data. Unfortunately, much of it was siloed and still required manual processes and a team of analysts to meaningfully tie it together. Moreover, even with that, we didn't have a complete picture of the patient, because anything that happened outside the walls of the hospital was often not fully reflected in the data.
As more electronic health record platforms come online (more than half of all physicians in the U.S. now use EHRs), we can tap into patient data from providers outside the walls of the hospital. We need only look at the number of times we seek health services outside of the hospital to understand how much additional, relevant data this will yield. The integration of data from new sources, coupled with the promise of big data, should work together to reduce healthcare costs and provide a more transparent and patient-centered ecosystem.
Population health management
Currently, healthcare resembles a large block of Swiss cheese, with continuity in some places and large holes in others. Patients go from primary care physicians to specialists to the hospital — and those providers may or may not be communicating in real time, if at all. To manage the health of these individuals, you need to connect all providers, including the patient's pharmacy and their local urgent care center. Only then can you begin to meaningfully discuss applications to broader patient populations. Big data's core underpinnings are efficiency and low cost, allowing you to connect with and assimilate quickly all clinical and claims data from primary and secondary care providers, large hospital institutions, payer organizations and others.
As an example, let's consider a group of patients with diabetes. There are a number of established best practices to ensure compliance with evidence-based standards for this condition. The only way to discover who's compliant and who's not is to look at every possible source of patient data for the presence or absence of these practices, such as the A1C test. Ideally, you would see that the test was performed and receive results within the patient's medical record. If you don't see it there, it may be linked to the test results from a national laboratory or within that patient's claims data. Regardless of where you find the information, the key is to have an information source that consistently provides clinical and claims data, in real-time, at the point of care, within the clinician's workflow. Having this information available avoids duplication of services and allows you to move on to the next step in the treatment plan. Done consistently, this type of widespread action across patient populations can improve quality of care while also driving down costs.
This is the ideal, of course — to be able to do this, and do it widely. The question is how long it will take. In the past, providers were like islands, and those islands didn't talk to each other. (And if they did, they talked via facsimile — I know, scary.) With the continued adoption of EHRs and more widespread health information exchange, we're building better bridges between these islands and facilitating more efficient and effective communication. Every piece of data counts. We may never get to 100 percent of every patient's clinical and claims information, but even getting to 50, 60 or 70 percent is a vast improvement over where we were even a decade ago.
Aggregation
One reality check: Clinicians are on data overload. Healthcare IT professionals want to give clinicians the tools they need to be successful rather than larger amounts of data "as is." Gathering the data is only step one. It's about collecting, de-duplicating, parsing and sorting the data, then building systems that can aggregate and extrapolate the most valuable, targeted information needed to provide quality patient care.
These aggregated systems are not likely to be driven by physician offices or even hospital systems. Most likely, they will come from another entity. And that's why you see the push from the Office of the National Coordinator and other entities to get health information exchanges in place. HIEs amass vast amounts of data across a number of provider organizations. With the data aggregated, there's an opportunity to develop algorithms for analyzing the data and providing value-added information to providers — whether they are managing an individual patient at the point of care or analyzing specific patient populations to identify outliers in need of intervention. Decisions you make today on EHRs, HIEs and strategic organizational alignments will have a significant impact on how well positioned you are to take advantage of what lies ahead.
Ultimately, what big data will represent is the ability to efficiently aggregate data, filter it, synthesize it, translate it into a common vocabulary and route it to where it needs to go to accomplish the goals of improved health and lower costs. The more providers do now to secure a solid foundation for data aggregation, the more they will thrive in the era of big data, making a real and measurable impact on patient outcomes and the sustainability of their organizations.