Data Delivery

Using Cloud (and AWS) in a Reluctant Industry

Cloud isn't a cure-all for IT departments in the health care industry, but it definitely needs to be in the conversation.

We are in an interesting yet challenging environment in health care data in 2015.

I am currently the director of informatics at a translational research institute for a large health care system just outside of Washington, D.C. My team has been tasked with creating a one-of-a-kind data and IT architecture to support the translational research initiatives of our health care system. We are gathering the combination of molecular, clinical and environmental data, and leveraging it to precisely treat our patients as individuals and to give them the best care at the best time.

This is an extremely challenging task at many levels. In this column, I am going to concentrate on the IT technology part of the equation, and leave the scientific and clinical challenges to the true superstars of our initiatives.

Five years ago, when I started working with life sciences and health care data, many of the technologies available today were not on the radar for health care systems. Cloud computing, Big Data, Hadoop and personal wearables were technologies used only in the larger and more progressive research environments. Back then, these were seen as fringe technologies, and were not part of many hospital IT roadmaps.

Fast forward to today, and cloud technologies are, at the very least, part of the conversation.

I was in the audience of a demo recently by a company that was pitching a data service solution to hospital leaders. About 10 minutes into the sales pitch, the presenter began to quickly tense her face and the rate of her speech quickened. Even the tone in her voice began to get defensive. It was just seconds before she was supposed to talk about the infrastructure portion of their solution, which was cloud-based. She was expecting resistance.

This is not unexpected, and I have seen this in multiple presentations to our team. I imagine this is one of the major hurdles that cloud-based companies have when selling to traditional health care systems. Luckily for her, we have been a cloud services user for over four years, and we quickly got the information we needed to make an informed decision. Our team evaluates cloud-based systems on their capability and security, just like any of our on-premises systems.

Today, my team and I are using cloud services in various forms. However, we make extensive use of Amazon Web Services (AWS). In late 2014, we signed our HIPAA business associate agreement (BAA) with AWS (a BAA is a contract between a HIPAA-covered entity and a HIPAA business associate that is used to protect personal health information in accordance with HIPAA guidelines). We leverage AWS and our on-premises infrastructure to enable our researchers and clinicians to utilize data and supporting IT systems more efficiently and effectively. As we enable our teams to get quality data faster, we enable them to discover, write papers and collaborate with other scientists. faster. We believe this will translate into better treatment for our patients.

We also believe that IT and informatics are key enablers to enhance the speed and quality of research. We can also help drive business costs down, which therefore leaves more dollars for additional research and technology.

For many of us in health systems who are struggling to manage massive and highly variable datasets, the cloud offers options to help IT departments provide the services that need to be provided. However, rapidly implementing newer IT technologies has never been a strength for health care systems. Patient care is a top priority and for good reason. The perceived risk of changing to new technologies does not justify implementations that might jeopardize patients' lives and hospitals' reputations. There are exceptions, but there is still hesitation from traditional health care organizations toward using newer technologies. Fortunately, we are one of those exceptions.

From our experience, we find there are many advantages for health care systems to adopt cloud-based technologies. However, particularly in the health care space, it is not a global solution to all that ails health care IT departments. We think of it as just another tool in our arsenal. We are challenged with extreme data sizes, variable data shapes, semantics and an extremely educated and demanding user base. We need everything we can get.

There is also the core business case for cloud, which is difficult to prove. It's hard to determine ROI when it comes to patient care, particularly in the translational research space, where much the cost benefit of the data-driven research will not be "realized" in a particular fiscal quarter or on cash flow. The best way we have done it is by comparing the alternatives, which is the traditional datacenter. There will be more detail on that in a subsequent article.

The overall goal for health care is to create better outcomes for patients and a sustainable cost. Cloud and new IT technologies can be a key enabler. Those who are implementing these new technologies correctly have begun to create a competitive advantage over the slow adopters.

Our team, anchored by cloud-based technologies including AWS, are well on our way to creating a secure, efficient and scalable IT and data infrastructure that will help our institute realize its mission of using genomic and clinical data to improve patient care. Will cloud services become part of mainstream services for health care systems? Are you using one of them?

About the Author

Aaron Black is the director of informatics for the Inova Translational Medicine Institute (ITMI), where he and his team are creating a hybrid IT architecture of cloud and on-premises technologies to support the ever-changing data types being collected in ITMI studies. Aaron is a certified Project Management Professional (PMP) and a Certified Scrum Master (CSM), and has dozens of technical certifications from Microsoft and accounting software vendors. He can be reached at @TheDataGuru or via LinkedIn.