Healthcare Meets Big Data

When healthcare and big data collide, things can get messy. On the one hand, the sheer volume of data being collected is opening up new paths to discovery and holds out the prospect of a smarter, data-driven future. On the other hand, we’re often buried in more information than we know what to do with, and it’s not always the information we need most.

The good news is that we’re now nearing the point of convergence, where healthcare providers can actually gather and harness the right inputs to make more precise, predictive and productive decisions. The data can be used to build models that not only lead to better patient treatments, but help solve payer and provider problems as well. Imagine using the same set of tools to better understand and enhance everything from therapeutics and clinical trials to health insurance processes and medical billing operations.

The Goal: Better Decision Support

The need for those solutions has never been greater. The healthcare industry is under intense pressure from payers and regulators to streamline processes, improve care management and deliver better outcomes at lower cost. Patient expectations have risen, too, driven by the consumerization of healthcare. No longer content to take whatever’s handed them, patients want everything faster, in a way that’s convenient, affordable, transparent and tailored to their needs.

Meeting those demands means making faster, better-informed decisions, and that’s where big data can make a big difference. In the past, healthcare organizations had a very restricted data set to work with. They had access to patient records through EHR (Electronic Health Records) systems, of course, but lacked broader contextual information, ranging from the clinical interactions that occur during care delivery to the external variables known as the “social determinants of health” (SDH). These are the conditions in which people are born, grow up, work and age. They include a range of environmental factors and large-scale social forces such as economic policies, development agendas, social norms and political systems.

Health systems need decision support solutions that factor in these broader inputs, helping clinicians to create personalized health plans based not just on a patient’s immediate condition, but also their history and that of their family, the impacts of social determinants, and data from people with similar conditions and backgrounds. Combining all of that information with the interoperability of EHR systems across different networks, and we’re looking at a new era when broad data sets can be used to help drive healthier outcomes for patients and families.

Several of the pieces are already in place. Every day, systems collect thousands of pieces of health data about patients. New technologies such as telehealth, wearables, bluetooth devices and remote sensors are delivering additional statistics on blood pressure, physical activity, blood glucose levels, weight and more. And then there’s all the data within the health system and payer networks – data on how treatments are prescribed, how medications are used, how patients respond, how billing processes work… the list goes on. This information is then stored away, waiting to be extracted, transformed and analyzed.

At the same time, new analytics technologies have emerged to help us make sense of the data. It’s not just headline-grabbing breakthroughs like IBM Watson; all sorts of less glamorous but no less important machine learning, micro-services and AI innovations are quietly making their way into many of the tools we use, providing new insights and more precise results.

Needed: A Unified Hub

But no matter how smart the technology gets, we’ll never have a complete picture without all the relevant information. And that’s where there’s still a gap. Though applications such as health clouds, EHRs and telehealth capture lots of details around patient care, they’re still missing many of the clinical communications and conversations that take place in separate, disconnected systems. That information is essential to weaving together a coherent view of patients, treatments and outcomes, yet it’s largely inaccessible. Conversations are lost in siloed chat apps and email. Documents are often stored in multiple, difficult-to-find locations, unavailable to the clinicians who need them. Information that could impact a patient’s health is often not shared in a timely, efficient way.

What’s been lacking is a unified system for communication and collaboration: a single integrated hub for a wide range of interactions among clinicians and other staff – even across multiple provider organizations. Such a hub can dramatically simplify life for all involved, enabling them spend more time delivering actual healthcare and less time struggling with disconnects and tracking down information. Just as important, it provides a central source for all those communications and interactions that were missing from or buried in other tools. Those communications bring all-important clinical context to the more conventional types of data (such as patient records), stored in systems such as EHRs. With the advent of clinical collaboration hubs, all that essential contextual information is in one place, searchable and easily retrievable by clinicians and other staff.

The interactions that take place in such a hub aren’t just a resource for clinicians and healthcare administrators. They’re also rich fodder for built-in machine learning and AI, which can be applied by the system itself to improve performance and better assist human decision makers. That automated intelligence can analyze user behavior and relationships to learn people’s preferences, interests and aims – information that in turn can be leveraged to create a more personalized user experience, to fine-tune search results, to provide predictive recommendations and to proactively alert users to people, news and activities of interest.

Moreover, by providing robust analytics that harness the system’s social graph (the web of relationships and interactions that take place in the hub), a collaborative hub can help organizations dig deep into untapped data to discover correlations between healthcare processes, patient outcomes, cost efficiencies and more. Data can be extracted, transformed and loaded into a single dashboard that allows providers, payers, and decision makers to visualize relationships and codependencies that were once unseen. Which departments are interacting most? Is information getting shared to the right people in a timely way? Where are the gaps, and how can communications be improved to increase quality of care, reduce medical error rates and improve organizational performance? Imagine having timely answers to questions like those.

Big Data, Bold Decisions

Is this vision coming true? Yes and no. Collecting and integrating some types of data – such as the social determinants mentioned earlier – is still a challenge. A lack of interoperability in EHR systems and concerns about security, privacy and data ownership are slowing the kind of data convergence needed to generate accurate models. And the models themselves – needed to turn all that data into actionable guidance – are a work in progress.

On the bright side, one key ingredient has arrived: the collaborative hub, providing clinicians with a “one-stop shop” for communication and connection within and across providers. That unified hub doesn’t just unleash new efficiencies and reduce disconnects, redundancies and errors; it also aggregates essential information that was formerly siloed and inaccessible but can now be used to improve decision-making. In fact, a state-of-the-art hub can use its own built-in predictive analytics to assist in such decision-making.

That smarter, data-driven future we talked about earlier isn’t so far off after all. In some ways, it’s already here.