Here are seven things to know about how advanced analytics can be used to make predictions in healthcare.
1. This form of advanced analytics has been talked about for nearly a decade — but we are just now seeing large advances in its use in the last 24 months. In 2015, the global healthcare predictive analytics market was valued at $1.48 billion, and is expected to continue to grow at a compound annual growth rate of 29.3 percent through 2025, according to a Grand View Research report.
2. The recent advancements in predictive analytics are powered by a combination of great minds, high-powered data tools and computing, which are being used to attack real-world problems. Two drivers of the recent growth in predictive analytics also include mounting pressure to contain high healthcare costs and increasing consumer demand for personalized medicine, according to Grand View Research.
3. Barriers to deploying predictive analytics have focused on workforce and technical issues. In an interview with Becker's Hospital Review, J. Bryan Bennett, executive director of the consulting organization Healthcare Center of Excellence, noted that before approaching predictive analytics, healthcare organizations needed to collect and integrate relevant data.
4. In the coming year, both the opportunities and challenges for predictive analytics will likely increase, with statistical and predictive data analysts slated to be one of the top 10 most in demand — yet hardest to fill — technology roles in 2017, according to a Forrester report. As the market continues to expand, Grand View Research also notes how key players like IBM, Cerner, Verisk Analytics, McKesson, SAS and Oracle will dominate the space.
5. The applications of predictive analytics range from a variety of business problems, such as staffing issues, to a number of clinical issues around subjects like pneumonia. MedStar Montgomery Medical Center in Olney, Md., recently launched a fall prevention program with analyticsMD's analytics solution, and VigiLanz's has used its analytics methodology, termed Temporalytics, to identify those most vulnerable to developing sepsis.
6. New research suggests predictive analytics can improve healthcare. For example, a study in Radiology concluded that a machine learning model could predict patient survival of heart failure. However, researchers have also faced challenges deploying their findings, as a study in JAMA Cardiology suggests predictive models validated in clinical studies may not prove accurate when used with EMR data.
7. A few recent large-scale deployments of predictive analytics include Rush University Medical Center in Chicago working with GE Healthcare Partners to build a command center, which will use predictive analytics to manage its patient flow, and Penn Medicine in Philadelphia launching a big data initiative to predict which lung cancer patients are at risk for emergency room visits.