Hands-On Healthcare Data

Chapter 7. Trends in Healthcare Analytics

While data engineering is a critical component in how we leverage real-world data, it is not the most glamorous of topics. Of course, all of this cleaning and processing of the data is a means to an end. Whether we are on the IT side, the data side, or even the business/clinical side, the end goal is to derive insights of the data that will improve decision making.

For business use cases, this may be to support more efficient care, improve cost savings, or even increase revenue (which is still important, even for nonprofit institutions). For clinical use cases, this may be to enable earlier and more accurate diagnoses of diseases, connecting care teams to the right information about the patient at the right time or determining the appropriate staffing levels based on patient acuity. This last point is particularly interesting since it represents a subset of use cases where there is both a clinical and business benefit.

On one end of the spectrum, data analysts rely on business intelligence tools such as Tableau or Cognos. On the other hand, data scientists rely on NLP and machine learning (including deep learning) to build predictive models. While we often break “data science” or “analytics” tasks into simple buckets like “business intelligence” and “machine learning,” the reality is that these lines are becoming increasingly blurred as we integrate data deeper into both our clinical and business workflows.

In this chapter, we will discuss ...