Saving lives – planes & patients
Big data has been one of the top tech concepts lately. As its name suggests, Big Data is compelling in part just due to its size – and the fact that it is tied at the hip to another major technology: it typically necessitates use of the cloud’s processing power to effectively run analytics on the massive datasets.
One of the biggest names in the news related to these prominent technological concepts is General Electric. GE is moving aggressively toward the public cloud, with more than nine of every ten new apps deployed in a cloud setting last year. One of its engines produces approximately 500 GB of information each flight. That means GE has to handle 2 TB of additional information every time an aircraft touches down. That occurs every six seconds.
What does it really mean to be in the information age? It means that we are cranking out data, much of it incredibly useful but untapped. GE is just the tip of the iceberg: 2.5 quintillion terabytes of data were produced daily in 2012. The same amount of information is now created in 48 hours as was produced between the origin of humankind and 2003.
Obviously healthcare organizations aren’t operating planes, but big data holds promise just the same – in both cases, it could save lives. For GE, it could prevent crashes. For healthcare, it can help researchers pinpoint new treatment approaches. However, issues with compliance and interoperability in healthcare make it difficult for healthcare technology professionals to make the most of it.
“While other industries have been far more successful at harnessing the value from large-scale integration and analysis of big data,” explained Mayo Clinic associate professors Nilay D. Shah and Jyotishman Pathak, “health care is just getting its feet wet.”
HIPAA compliance and the general issue of patient privacy make big data sound daunting – but its value simply can’t be ignored. Shah and Pathak constructed a basic roadmap of what the healthcare industry needs to do as it warms up to big data.
Healthcare big data roadmap
Data standardization
Many organizations have combined various streams of data so that they can better analyze their users and reshape their approaches. Heterogeneity of data is a healthcare obstacle that must be overcome.
Interoperability is often framed in terms of devices working properly together on a moment-by-moment basis; patient care suffers when data can’t transfer between medical devices in real time. However, the issue of data integration stretches far beyond the interoperability of clinical medical machines.
“The vast amount of data generated and collected by a multitude of agents in health care today comes in so many different forms,” said Shah and Pathak, “from insurance claims to physician notes …, images from patient scans, conversations about health in social media, and information from wearables and other monitoring devices.”
Combining data so that it is useful needs support from both business and government. The National Institutes of Health recently created a program entitled Big Data to Knowledge Initiative to foster big data projects within biomedical research. Other collaborative efforts include the National Patient-Centered Research Network and Optum Labs.
Insights
Integrating data is of course just one part of the process. The real question with all this information is how to make sense of it. That’s the realm of predictive analytics. By pulling in data from various sources, providers could make better educated guesses about what type of treatment would work best for a particular patient.
“These predictions may help identify areas to improve both quality and efficiency in health care in areas such as readmissions, adverse events, treatment optimization, and early identification of worsening health states or highest-need populations,” commented Shah and Pathak.
Revolutionary analytics methods have been developed, but they aren’t typically utilized in the healthcare field. Examples include the graph analytics and machine learning that have been used to advance the objectives of retail companies. The former method is starting to gain traction in healthcare. Artificial intelligence projects are underway as well, such as the Mayo Clinic’s use of IBM’s Watson supercomputer to find patients that would be the best fits for clinical trials.
Action
Once data has been pooled together and insights have been garnered, the next step is practical application. That requires a cultural shift.
The knowledge obtained from analytic projects can proliferate across a wide spectrum, related to
- How safe and effective certain treatments are
- Outcomes of one care framework versus another
- Predictive strategies to further refine methods, especially with new approaches.
The cultural shift that’s required is simply to understand that what the data suggests may mean that policies and practices have to adjust. The knowledge coming from computers is becoming a significant corollary to the knowledge provided by doctors, administrators, and other stakeholders.
Meeting the big data challenge head-on
“Ultimately, what should drive this initiative and others is to addresses the complexities, unmet needs, and challenges facing patients,” said Shah and Pathak.
The first place to start when you consider serving your patients is with a provider that offers a HIPAA-compliant cloud, properly protecting their health information. For your big data initiative, use the ClearDATA HealthDATA™ Cloud Platform, run through the only healthcare-exclusive cloud in the world.
