Were you ever told to go with your gut?
While this has been widely held belief, increasingly the smartest industries are running as far away as possible from instinct-driven decisions. And this includes health care. Instead, clinicians have mountains of data that can give them clues to not only what their patients want, but how to treat them and when.
Clinicians have stepped up to this big data challenge with the deployment of real-time analytics platforms. Previously, a care facility might have tracked its patients by pulling batch data. Think of batch data like this: A pharmacy wants to keep limit the number of behind-the-counter purchases of pseudoephedrine per customer. So at the end of each day, its data management system pulls records for each order filled that day and adds it to the total data accumulated over the past year.
However, that pace isn’t wholistic or realistic. A patient could go to multiple pharmacies in a row. Therefore, pharmacies must use a real-time data capture at the state level to prevent this behavior. To do this, a facility can incorporate contextual analytics to this picture — a practice where an enterprise tracks behavioral data to be able to determine if something is amiss — it becomes that much easier to flag customers who are breaking the rules.
Contextual analytics could also let a health care facility know if its employees are potentially breaching the internal security measures put in place on its data. For instance, an after-hours log in to review patient information from an employee that normally doesn’t perform that behavior could let a company know about nefarious behavior before it becomes a bigger issue.
The need for this kind of real-time data relay is hitting many industries. And the Internet of Things (IoT) has created countless inbound data points for clinicians to gather and track more information. The trick is getting an analytics program to turn that around data in real time to maximize revenue and patient convenience.
A large-scale example of contextual real-time analytics benefitting a big business when processed in real time was Disney’s deployment of MagicBands. With amusement park income slumping, Disney made a $1 billion bet that their problem was a data problem. These wristbands act as credit cards, hotel room keys and park passes for patrons. But they are so much more to Disney.
The park analyzes in real time what its visitors are doing. If it notices foot traffic is above an enjoyable threshold in one area, employees are instantly alerted to create a spontaneous diversion to steer the crowd away, giving everyone a more pleasant park experience.
Patient care can be coordinated in the same data-driven way. And the desire for this type of information is coming from both patients and providers. According to a PwC Health Research Institute report, 88 percent of patients are willing to share their personal data for treatments and 60 percent are willing to have a virtual doctor visit. And 81 percent of clinicians say mobile access to medical information helps them coordinate patient care and more than half would rather provide a portion of their care virtually.
These health care providers know there is power in real-time data analysis. Telehealth is projected to be at $34 billion industry by 2020, and high-end machine learning platforms can process 700 hospital security data points per second. What good is all of this innovation if finding important health care trends is still regulated to batch data analysis?
Synchronoss’ numbers suggest companies with traditional analytics spend 80 to 95 percent of the time collecting data, with only the leftover sliver devoted to creating actionable items based on the data.
Health care providers need to invest in a software-as-a-solution model to autonomously marry the data they collect with what decisions they need to make. By connecting gross data collected through an existing M2M system with a powerful and easy-to-use software system, a hospital could relay a patient’s changed status to a select host of doctors and nurses, ultimately making patient success rates higher and hospital staff smarter.
The typical time to develop a real-time data analytics solution in house is 24 months. However, there are solutions like Synchronoss Analytics that can be implemented within 90 days and integrated into your back-end systems through web APIs and batch processing systems.
Big data doesn’t have to be a big headache for health care. Harnessing the power of the emerging fields of real-time and contextual analytics empowers providers to operate smarter, not harder, and gain better results in the process.