Prescriptive Analytics Beats Simple Prediction for Improving Healthcare
Those who work in and around Big Data quickly come to understand that oodles of information alone does not directly equate to knowledge and insight. It is often said that data plus context equals knowledge. In other words, for meaningful use to be derived from large amounts of information—the associated context of that data must also be readily available and simultaneously considered. An additional caution is not to confuse simple insights with real dollar value. Nowhere is this more true than in health care.
The Challenge with Predictive Analytics in Healthcare
With all the current hype surrounding big data and predictive analytics, it’s challenging for organizations to sift through all the buzzword and marketing noise. Healthcare providers need to partner with groups that have a clear understanding of the leading academic and commercial tools available, along with the expertise to develop appropriate prediction models.
In order to be successful, clinical event prediction and the resulting intervention should be both content-driven and clinician-driven. Prediction tools should only be used in the context of when and where needed – with clinical leaders that have the willingness to act on appropriate intervention measures. Decision makers cannot be isolated or far removed from the actual point of decision.
Prescriptive Analytics in Healthcare and Clinical Action
So how can we successfully integrate predictive analytics into a healthcare delivery system? In simple terms, prediction is most useful when that knowledge is conveyed into clinical action. This is what is meant by “integrated prediction” or prescriptive analytics. This “in context” prediction should include not only the evidence, but also the interpretation and recommended actions for each predicted category or outcome. Most importantly, prediction should carefully link to clinical priorities and measurable events, such as cost effectiveness, clinical protocols, or patient outcomes.
A great example of this in context prediction is seen when dealing with hospital readmissions. Many stand-alone applications simply display the forecast of patients likely to return to the hospital in the next 30 days, along with a few other fancy looking charts. A better use of the predictor would be to drive decisions from the same dashboard and see associated cost simulation, real-time hospital census bed counts, pending medication reconciliation, or adjusting order sets for education material and in-home follow-up. Armed with this additional information, hospital staff can determine those patients at highest risk of readmission and take action to mitigate this risk, such as emphasizing patient education at discharge or ensuring timely communication with primary care physicians and acute care facilities.
Healthcare Enterprise Data Warehouse
Unfortunately, much of the related information discussed above does not reside in any one system or data set. An underlying data warehouse platform is essential to gathering complete data sets necessary for training and implementing clinical predictors. Furthermore, to best judge the efficacy and value of forecasting a trend and changing behavior, that predictor-intervention set must be integrated into the same system and workflow where it originally occurs. For this reason, prediction within a robust data warehouse environment will always be superior to standalone or siloed applications.
Not surprisingly, the cutting edge enterprise data warehouse platform and Late-binding™ approach of Health Catalyst excels in this space. Return on investment does not reside in data itself, but in timely interpretation and appropriate intervention. Remember, “Information plus context equals knowledge. But predictions made solely for the sake of making a prediction are a waste of time and money” (quoting from my whitepaper: 4 Essential Lessons for Adoption Predictive Analytics in Healthcare).
Have you seen predictive analytics used in a stand-alone fashion or out of context in your organization? What opportunities might present after you’ve seen the information in context?
Read More About This Topic
Using Predictive Analytics in Healthcare: Technology Hype vs. Reality by David K. Crockett, Ph.D., Senior Director of Research and Predictive Analytics
3 Reasons Why Comparative Analytics, Predictive Analytics, and NLP Won’t Solve Healthcare’s Problems by Dale Sanders, Senior Vice President, Strategy
4 Essential Lessons for Adopting Predictive Analytics in Healthcare by David K. Crockett, Ph.D., Senior Director of Research and Predictive Analytics
In Healthcare Predictive Analytics, Big Data is Sometimes a Big Mess by David K. Crockett, Ph.D., Senior Director of Research and Predictive Analytics
Predictive Analytics: It’s the Intervention That Matters (Webinar, Slides or Transcript) by Dale Sanders, Senior Vice President and David K. Crockett, Ph.D., Senior Director of Research and Predictive Analytics