Predictive analytics refers to an advanced branch of analytics that uses data to make predictions about future events. Hence, predictive analytics for healthcare is using advanced branch on analytics to analyze healthcare data and make predictions about patients health and other important healthcare metrics. The process of predictive analysis of data uses techniques such as statistics, data mining modeling, machine learning and even artificial intelligence.

Using these tools, complex decisions and trade-offs can be made based on predictions made from data gotten. Predictive analysis for healthcare makes use of data to determine at-risk patients for chronic diseases such as heart diseases, diabetes, asthma etc. This makes medical decision making more a more informed process devoid of guessing. Now even though predictive analytics for healthcare is of huge advantage to the healthcare system, it won’t necessarily solve the problems in healthcare.

For years, predictive analytics has supported risk stratification in case management and other means. However being able to predict only readmissions (which has been the case) is of little help. That’s like being able to predict that there’s going to be a flood but offering no help as to how to prevent it or protect people from it.  Without protocol and patient-specific outcomes data, predictive analytics is largely vendor smoke and mirrors in all but a very small number of use cases. Therefore, before we start believing that predictive analytics is going to change the healthcare world, we need to understand how it works, technically and programmatically.


When you talk predictive analysis for healthcare you have to back it with the proper infrastructure, resources, and staffing. This is so action can be taken when something is predicted with a high certainty of it happening. If the right resources are not in place to take action then health organizations are unable to harness any historical trends or patterns in patient data.

In order to be successful, healthcare organizations should implement clinical event prediction and subsequent intervention for both content driven and clinician driven actions. These even prescriptions would include evidence, recommendations, and actions for each predicted category or outcome.

Now, this doesn’t mean that health organizations should turn away from comparative data and predictive modeling. These analytical tools are very vital to healthcare. It is instead, a call to be more intentional and researched driven when adopting models. As in fashion, there are also fads in the healthcare and technology industry. Deliberately, but quickly, move your organization up the levels of the Healthcare Analytics Adoption Model. This healthcare adoption model draws upon lessons learned from the HIMSS EHR Adoption Model and describes a similar approach for assessing the adoption of analytics in healthcare.

It might take up to five years, if not more and a new generation of EMRs, patient reported outcomes systems, and activity based cost accounting systems before the healthcare industry can close the gaps in data ecosystem to make predictive models and NLP widely valuable in the industry.  Comparative data will not be as valuable as it should be until its variability is gotten out of common healthcare practices and data definitions of diseases and syndromes are standardized. Variability analysis, not benchmarking, might be the most useful application of comparative data.

“About 85% of my ‘thinking’ time was spent getting into a position to think, to make a decision, to learn something I needed to know. Much more time went into finding or obtaining information than into digesting it … Several hours of calculating were required to get the data into comparable form. When they were in comparable form, it took only a few seconds to determine what I needed to know.”

J.C.R. Licklider

Within Health Catalyst, data modeling and algorithm development is performed using industry leading tools for data mining and supervised machine learning such as Weka, Orange, and R. Ongoing efforts include classification models for a generalized predictor of hospital readmissions, heart failure, length of stay and clustering of patient outcomes to historical cohorts at time of admit.

Most importantly, they have internal access to millions of de-identified hospital records in both the inpatient and outpatient settings and adult and pediatric populations. This training data is crucial to addressing the predictive analytics demands of clients and site customization. So when requests come from healthcare organizations and clinicians – whether it involves classification or clustering or feature selection – Health Catalyst has the tools and the data and the expertise to successfully deliver top performing predictive analytics.