The use of big data and analytics technology can help increase revenue, enhance customer service and improve decision making in a variety of industries. In healthcare, it can help save lives.
The Evangelical Lutheran Good Samaritan Society, the largest not-for-profit provider of senior care and services in the United States, has launched a big data analytics project that the organization says will enable it to predict and, in many cases, prevent heart attacks and others catastrophic health events.
Based in Sioux Falls, S.D., Good Samaritan has more than 240 locations across the country and 27,000 individual clients in 24 states. The society’s more than 20,000 employees provide a range of services including rehabilitation/skilled care, assisted living, senior living and home care. Its primary mission is to enable seniors to live healthy, independent lives at home for as long as possible.
The organization’s IT infrastructure includes database technology from IBM, Microsoft, Oracle and other vendors, according to Rustan Williams, the society’s vice president and CIO.
Good Samaritan is in the midst of building a foundational data warehouse that will house petabytes of data gathered from a number of diverse sources, including sensors, to promote wellness and safety for clients in their homes.
By gathering a huge volume of data in a large data warehouse and then applying advanced analytics, Good Samaritan is hoping to create a more accurate picture of its clients’ health, which will enable better early detection and intervention, says Williams.
PUTTING THE PIECES TOGETHER
A centerpiece of the Good Samaritan’s strategy is IBM’s Big Data and Analytics technology, which provides the society a suite of healthcare and wellness software and services, including predictive analytics. Good Samaritan began deploying the IBM technology in January 2014.
Among the key components of the IBM offering are IBM Enterprise Content Management, which allows companies to capture, share, and govern data; and IBM Case Manager, which supports ECM and provides healthcare workers with access to analytics and dashboards.
Also part of the IBM solution is Patient Care and Insights, a configurable suite of products that combines advanced analytics and care management capabilities to help health organizations plan care.
With the IBM systems, organizations can mine structured and unstructured data such as vital metrics and test results, doctors’ notes and diagnostic reports. Providers can identify and visualize trends, patterns and anomalies, and determine their relationship to medical outcomes.
IBM’s Big Data and Analytics technology is at the heart of Good Samaritan’s LivingWell@Home program, which uses passive sensor devices, including motion and temperature detectors, to track how active clients are, observe their sleep patterns and monitor their falls. The sensors provide two other functions -- monitoring vital signs and serving as personal emergency response devices. Data is sent via WiFi to the organization’s servers.
“There are three distinct data sources from those services and they need to work in an integrated fashion, because what happens is when you put all of that data together it paints a [more complete] picture of the health status of the person,” Williams says. “If there are any deviations from an activity baseline, a lot of times that indicates a change or pending change in health status, so with these early indicators or alerts you have an opportunity to intervene.”
By using IBM’s Big Data and Analytics software, Good Samaritan can consolidate sensor information along with clinical and operational data on clients, Williams says.
Data is gathered continuously and delivered securely through an integrated dashboard to licensed nurses who review the information daily.
Staff at Good Samaritan checks the data for changes in a client’s normal routine that might indicate a developing medical condition or a safety concern, so they can address issues in real time before they become problems.
The technology allows Good Samaritan caregivers to proactively monitor, manage and collaborate to ensure successful home-based eldercare for clients. The Internet-based services are not restricted by location. The services can be used regardless of care setting by clients, doctors and family members.
To date, Good Samaritan has monitored more than 5,000 people in the program. It is currently tracking about 1,500 and anticipates that by the end of next year the number of actively monitored clients will approach 3,000.
Before the IBM systems were deployed, data coming in from various sources wasn’t integrated, and users had no easy way to correlate information to assess clients’ health.
“The data was siloed in its own disparate applications that were providing specific functionality within the given application,” Williams says.
Good Samaritan has begun developing metrics to track the benefits of the big data and analytics tools. Yet the society says it can already quantify some gains.
For example, the ratio of monitoring nurses to clients for the LivingWell@Home program has improved from one-to-100 to one-to-350. In other words, one registered nurse is now able to monitor the healthcare status of 350 clients, up from 100 in the past. And the IBM technology is one of the reasons for the improvement, which has led to cost savings by allowing the Society to increase monitoring volume without increasing monitoring staff, Williams says.
“In the past, nurses had to go to three different data sources and do an integration on the fly, which takes time,” Williams says. “Now we can put all the data together in one user interface.”
With all of the disparate data sources within the organization delivering information into a single pool, Good Samaritan is now positioned to perform deeper analytics to predict possible events that could jeopardize a client’s well-being.
“This sets the stage for us to use analytics so that we can be more proactive in the health care of our clients,” Williams says. “We’re just now starting to develop some predictive models as they relate to healthcare events and what the information is telling us in terms of emerging [client] healthcare changes.”
The implementation of the big data and analytics tools has come with several challenges, Williams says. One is keeping costs under control to stay within budget. Another is determining the optimum timing for consolidating data for assessments due to each client having his or her own unique schedule (for example, when they eat, go to sleep and wake up).
Good Samaritan has also had to work out data integration issues between the IBM systems and those of other vendors, Williams says. “We have completed significant work with the vendors to address this,” he says.
Finally, there are issues related to security and privacy. As a provider of healthcare services, Good Samaritan must comply with Health Insurance Portability and Accountability (HIPAA) regulations, including the privacy and security of patient information.
“We have had to put processes in place where we can authorize who has access to information,” Williams says. “If a family member wants to access information they have to first be authorized.”
In the future, Good Samaritan will use data analytics for various diagnosis-related groups, modes of care and other services to help it define additional cost-effective care solutions to prevent readmissions and other potentially negative outcomes for clients.
For now, based on the response of Good Samaritan employees who work with client data, the big data and analytics project has been a huge success and staffers see many opportunities for using the IBM technology.
“We’ve whetted peoples’ appetite with these tools; they see the promise and the value and we’re getting overwhelmed with requests” for data consolidation and analytics, Williams says. “We’ve heard some of our [LivingWell@Home staff] say they can almost predict from passive sensor monitoring and actual vitals that a healthcare event may be coming. If we can [better] understand those patterns, we can be predictive on a grand scale, and provide information to physicians, home healthcare providers and even to family members to prevent these events from happening.”