Jun 19 : Preparing the healthcare industry to capture the full potential of big data

Hospitals and healthcare professionals across the industry are increasingly holding the
same debate: what to do with the rapidly growing volumes of all forms of data we face. We want to know how to gain actionable knowledge from it, how to use it to reign in costs, how to aggregate it, how to optimize it, how to share it, how to secure it, how to accelerate the value it can have on quality of patient care, and better yet, how to turn “small” data into big data.

At the same time, the industry is moving from volume-based reimbursement to a value-based model and is struggling, for the most part, to improve overall healthcare delivery: higher quality, lower costs, better patient experience—the de facto “Triple Aim.” And as we try to figure out the definition of success in this new era of healthcare, let alone become successful, we’ve already seen organizations start to restructure their care delivery systems, form accountable care organizations (ACOs) and start to prepare for episode-based payment (payment bundling).

The reason? Well, simply put, we know we will now require, or even be required to have, sophisticated tools, mainly by way of analytics, to ensure we are providing high-quality care while staying within budget. Clinical analytics will empower population health management and allow us to maximize health and minimize costs of defined groups by improving quality and reducing hospital readmissions. And financial analytics will allow us to manage our risk effectively by measuring the cost of delivery, thereby giving us the means to utilize our resources more prudently. More importantly, the big data and data-driven healthcare solutions that are just starting to become available in our industry can be leveraged to answer questions and deliver treatment that has not been fully imagined before now.

The caveat, though, is that in order to do so, we will need access to a comprehensive view of the patient and the care they receive across the continuum to put all the pieces of the puzzle together. Which is exactly why I posed the question recently as to whether or not healthcare is even prepared to utilize the full potential of big data, or if roadblocks will impede its use, since this essential detail seems to be missing from a majority of conversations currently being held across the industry.

Doing “big” data in healthcare
We know that big data will transform healthcare delivery, enable clinical efficiencies and promote greater accountability, or so that theory can be argued. Better yet, as a McKinsey report estimated in 2011, leveraging patient and clinical data could allow us to create more than $300 billion in value every year and reduce U.S. healthcare expenditure by roughly 8 percent.

But recently the industry has focused more on asking whether or not we should even use big data to create a 360-degree view of the patient and our business, as opposed to seeking the answer to a more prudent question: why other industries have been successful in obtaining value from their data sources and leveraging data to improve efficiencies and make more informed decisions—albeit also make more money. If we did, we would quickly recognize two points (among many more): first, what these other industries have figured out is that data is only transformative when disparate systems and data repositories can be linked together to form unified records at an individual level; and second, the industry as a whole is unprepared to address this challenge and thus capture the full potential and influence big data can have on healthcare because much of our data is siloed and fragmented.

We would also better understand data, how innovative its use could be and how it could enable us to reduce costs, identify and improve gaps in care delivery, improve quality, proactively identify at-risk patient populations and make care more accessible. To get to this point, though, we need to create a solid infrastructure—one that leverages data management, analysis, processing, optimization and storage—to empower us with the foundation to make more informed, evidence-based business and clinical decisions.

That’s all well and good but for one simple fact—where do we start?

Unlocking the power of patient data
Hospitals and healthcare systems are full of data—patient records, clinical and administrative documents, voice recordings, medical images, you name it—that are just waiting to be analyzed. And the analysis that can be done on this information holds considerable promise for an industry that is frantically seeking methods to cut costs, improve efficiency and deliver better patient care.

But for most organizations, a majority of clinical and administrative data remains unstructured, proprietary and siloed. In fact, as much as 80 percent of healthcare data is unstructured, according to a recent Institute for Health Technology Transformation (iHT2) report. Matters are made worse when you couple in the fact that 50 percent or more of a patient’s health information typically not captured and available for view in an electronic format.

Between these two points, the result is a majority of content comprising the patient record residing outside of the EMR system. Accessing data from these disparate—and often proprietary—repositories is a tedious process that, for most healthcare organizations, borders on impossible. This abundance of unstructured data makes analysis difficult, and better yet, it makes the ultimate goal of improving patient care problematic.

Before we as an industry can generate any real insights that not only benefit the health system, but the patient themselves, we must establish strategies and utilize standards for managing all the different types of data that we have. The challenge of doing this effectively, however, is identifying what all the potential sources of clinical and administrative data are and determining the value of linking them together since connecting data only adds value when it helps fill in the gaps. To do so, there are three areas we must focus on:

• Aggregate: Aggregating and gaining control of data is undoubtedly the essential first step to transforming care delivery and turning lots of fragmented data into a critical big data asset, but it is often overlooked by many healthcare organizations. Structured and unstructured data needs to be aggregated into a consolidated view of the patient, while adhering to healthcare standards such as HL7. This content also needs to be maintained in its natural form to facilitate search, analysis, data mining, and business policy enforcement.

• Transform: Almost equally as important is that we have to optimize the quality of patient data we store, and once it is in a usable form, we can begin to scrutinize it, understand it and use it. But new and existing applications also need to be transformed with automated processes and workflows that support the relationships between healthcare organizations. This will allow us to generate more meaningful knowledge that can then be used to positively impact patient outcomes and business operations. Analytics capabilities will also be strengthened and reports and dashboards will become enriched.

• Share: But what we do after we have gathered patient and clinical data, and decoupled it from siloed applications and systems? Not only do we need to be able to securely exchange healthcare information with every clinician across a given healthcare organization, but we also need to be able to do the same across the continuum of care, regardless of their facility location. Through the support of Integrating the Healthcare Enterprise (IHE) standard for Cross-Enterprise Document Sharing (XDS), Cross-Community Access (XCA) and basic patient consent, clinicians will gain a comprehensive, longitudinal view of each patient they care for, enabling them to make the best possible decision regarding their care when and where it matters most—the point of care.

The key is to go beyond aggregate data and the capture all structured and unstructured content types, to directly linking information to the individual patient within the electronic medical record (EMR).

This type of integration will provide clinicians and administrators with a complete patient-centric view of all health information—regardless of source, location or format—allowing for more focus to be placed on improving outcomes and reducing costs, rather than on mundane processes and procedures.

Source