We live in a period of time in healthcare where there is enormous pressure to analyze and compete based on the information buried in IT systems. The volume of healthcare data swells by a staggering 35 per cent per year. Worse than the volume is the fact that 80% of stored health information today consists of unstructured data such as physician notes and discharge summaries, email and satisfaction surveys, reference journals, possibly even text messages or social media. This unstructured data is stored inconsistently and is generally incomplete and overly difficult to access. The solution to wrestling the unstructured data beast and domesticating it is available today and comes from a “Watson ready” offering from IBM. This powerful IBM Content and Predictive Analytics (ICPA) for healthcare solutions enhances clinical and operational decision making by analyzing both unstructured and structured data in context against past, present and future scenarios.
Using ICPA, healthcare organizations can delve into their own patient diagnostic and treatment records, clinical policies, medical literature and other sources of qualified information in ways they never thought possible. One key example is the work currently underway at Texas-based Seton Healthcare to help increase the survival rate of congestive heart failure (CHF) patients. Studies have shown that proactive treatment of CHF patients can help prevent the readmission rates on those types of patients. Seton Healthcare is using the Content and Predictive Analytics solution to analyze patient data to determine the high-risk patients who would benefit from interventions. The project mines the unstructured data of patient records, developing predictive models to determine the potential for readmissions, applying that knowledge then to improve outcomes and reduce costs. With the increased pressure to reduce readmissions and costs, this predictive analytics approach is very timely for the proactive healthcare organization.
The impressive idea behind this solution is combining key technologies and pre-integrating them to reduce the time-to-value for deploying the specific application. One of the key components of the solution besides IBM Content Analytics and SPSS Modeler is a healthcare solution accelerator (HSA – as if we need more acronyms). HSA harnesses the natural-language processing powers of Watson to extract medical facts and build structured relationships through pre-built annotators for advanced text analytics. The exciting aspect of this accelerator is its ability to identify trends, patterns, deviations, and anomalies in data. HSA can expand predictive analytic capabilities beyond traditional modeling and scoring to predict outcomes! A physician could analyze a large group of patient treatment and outcomes records, along with recent medical literature or research guidance, then apply it to his or her own patient records to develop a highly personalized treatment plan. This solution has the power to be a game changer for chronic conditions, the identification and treatment of cancer or managing at-risk populations in general. ICPA has the potential to allow large, sophisticated healthcare organizations to extend world-class treatment options to small towns and remote locations – a big win for telehealth.
The applications for this combination of content analytics and predictive modeling are extensive including a favorite headache for most healthcare teams called manual chart abstraction. The average hospital can spend hundreds of thousands of dollars a year and tie up valuable nursing resources reading charts to complete medical records, prepare them for billing and summarize statistical quality data. This solution, coupled with modern paper scanning optical character recognition software, could also radically improve the quality and volume of valuable information hidden in paper records for analysis and, most importantly, predicting outcomes. Many physicians will tell you that the most valuable part of a medical record is the notes and IBM is providing the capability to prove it.
The best secret is last: IBM has a Jumpstart program to help a healthcare organization develop a strategy for implementing this solution and selecting a best path-to-value for a rapid return on investment. It doesn’t take a hardware system the size of the supercomputer that won on TV either, reasonable Power7 systems that are tailored for complex analytics of unstructured data and the compute intensive needs for predictive analytics. To summarize, your healthcare organization does not have to wait for Watson technology to come to you, the future is here now and ready to address your demand for predictive analytics and outcome management.
Would you like to learn more about Predictive Analytics and its capabilities? Come discuss it with us in Booth 1274 at the HIMSS show!
