Let’s navigate through what it takes to predict health outcomes and cost. What if we could use machine learning in your organization to reduce the cost of care for both the organization and the members receiving that care? Have you thought about what data you need to collect? How you might need to enrich that data to gain more insight in to what is driving those outcomes and cost? Or what types of machine learning algorithms you might utilize in order to most effectively target patients who are likely to be high cost? We are going to look at not only the tech behind the predictions, but also examine the business and data relationships within the healthcare industry that ultimately impact your ability to deliver an effective solution.

Now that we have discussed various types of predictive models, let’s take a look at which models are appropriate for the business case we are trying to address and how we can evaluate their performance. For example, is using the same performance metric appropriate to use when making predictions about individual vs. population health? In this module we'll discuss how layering appropriate decision support methods on top of predictive analytics and machine learning can lay the groundwork for significant improvements in overall outreach and productivity, as well as decrease costs. Finally, we will discuss the key to blending decision support into the existing ecosystem of your business workflow and technology infrastructure.

Now that we know the importance of decision support and predictive modeling, we are going to take that one step further. Not only do we need to predict, but more importantly, we need to prescribe. It is not enough to just implement alerts and reminders - we need to offer guidance and recommendations for healthcare professionals. Let’s take a look at how analytics can improve the patient experience and their overall health status.