The Challenges with Prescriptive Analytics in Healthcare

Prescriptive Analytics is not a new concept in healthcare, however, with the current landscape that focuses a lot on data, every healthcare organization has been forced to pay more attention to it. The landscape of predictive analysis can be quite broad so healthcare organizations have to take on outside partners that are more knowledgeable so they can understand both the research and commercial side of predictive analysis.

Prescriptive analysis for healthcare organizations should be both content-driven and clinical-driven. This means predictive analysis tools should only be used when needed and not for everything all the time. Predictive analytics can also serve as a decision-making tool that brings the management of the organization together. Decision makers cannot be isolated or far removed from the actual point of decision.

So, considering these facts about predictive analytics, how does one go about integrating it into a healthcare delivery system? How does a healthcare organization convert the information gotten from predictive analytics data to action? Prediction is only useful when it is transformed into action. The process of converting predictive analysis into action is called prescriptive analysis or integrated prediction. These concepts involve the interpretation of data into action and recommendations in order to achieve the desired outcome. Predictive analysis should carefully link to clinical priorities and measurable events, such as cost-effectiveness, clinical protocols, or patient outcomes.

For example, a healthcare organization can tackle their readmissions using predictive analysis. With predictive analysis, they can forecast the readmission rate for the next 30 days. They can also forecast associated cost, real-time hospital census bed counts, pending medication reconciliation, or adjusting order sets for education material and in-home follow-up and caregiving.  After putting all this data together in clear chats, a hospital can determine which patients are at a higher risk of o being readmitted. The for these patients, they can pay more emphasis to them during discharge, by educating them properly, making sure they understand their medication and even organizing a home care team for them if needed. After discharge, they can also ensure a timely and frequent communication between the patients and their primary care physicians and also between them and the healthcare organization.

The Usefulness of Prescriptive Analytics in Healthcare Technology

Technology and healthcare have continued to go well together in recent decades. This new marriage is a good thing and has served so many advantages. As with any boom period, there continues to be new technology flooding the healthcare industry that sometimes it can be hard to keep up. It makes one wonder how useful some of this technology is. Because they are still relatively new, there is no strict regulation or established vetting process for new technology in healthcare. Hence, there are new innovations that are gimmicky and not particularly useful. If a healthcare organization does not have a proper technology framework, they end up wasting time and money.  

For example, the technology that can sequence the entire DNA of an individual (known as genome sequencing) has been applauded as a great advancement in healthcare. However, the usefulness of this is not that interpretable or applicable. Beyond that, even if the full information gets passed along efficiently, but lacks the context of metadata or interpretation and classification of what it means and what to do about it, again the remarkable technology has fallen short of its full potential. And even if it is indeed applicable, it would require an entire framework that requires infrastructure that is not cost effective for healthcare organizations. In other words, the technology is there, but the means to deliver and interpret actionable data has yet to be fully developed.

Prescriptive Analytics and Population Management

  • Predictive analysis can help with risk stratification and population management. Predictive analytics was listed as level 7 out of the 8 possible levels on the Healthcare Analytics Adoption Model, there are many keys and pitfalls that can occur at such a level if not properly prepared.  However when it comes to big data, population health and predictive analysis, the following should be noted.
  • More data does not equate to more insight: It can be difficult to extract robust and clinically relevant conclusions, even from reams of data or big data. So just gathering big data is not the answer, being able to analyze it and properly draw useful conclusion and predictions from it is what’s key.

  • Insight and value are not the same: While many solid scientific findings may be interesting, they do little to significantly improve current clinical outcomes. Something might be insightful and very interesting but that does not mean that it is practical.
  • Ability to interpret data varies based on the data itself: Sometimes even the best data may afford only limited insight into clinical health outcomes. Good data does not mean that it can be useful for all purposes. Sometimes it just provides a tiny little information but information or insight plays a bigger role in the overall outcomes and in the big picture.
  • The implementation itself may prove a challenge: Leveraging large data sets successfully requires a hospital system to be prepared to embrace new methodologies; this, however, may require a significant investment of time and capital and alignment of economic interests. Sometimes implementing the findings of predictive analysis means executing a huge strategic plan that the organization may not be fully ready for. Or it might not be financially viable at that time. That does not mean that it should be abandoned, instead, a proper plan should be made to maybe break it down to smaller projects or cut down on other less viable aspects of the healthcare organization’s running strategies.