Defining Predictive Analytics in Healthcare

Predictive analytics is rapidly becoming one of the most-discussed, perhaps most-hyped topic in healthcare analytics.  Machine learning is a well-studied discipline with a long history of success in many industries.  Healthcare can learn valuable lessons from this previous success to jumpstart the utility of predictive analytics for improving patient care, chronic disease management, hospital administration and supply chain efficiencies. The opportunity that currently exists for health care systems is to define what “predictive analytics” means to them and how can it be used most effectively to make improvements within their system.

However, predictions made solely for the sake of making a prediction are a waste of time and money.  In healthcare and other industries, prediction is most useful when that knowledge can be transferred into action. The willingness to intervene is the key to harnessing the power of historical and real-time data. Importantly, to best gauge efficacy and value, both the predictor and the intervention must be integrated within the same system and workflow where the trend occurs.

The following Health Catalyst paper, “Using Predictive Analytics in Healthcare:  Technology Hype vs Reality” is a good summary of both the hype and hope of predictive analytics in healthcare

HC-banner-ER-using-predictive-analytics

How To Get Started With Predictive Analytics

If there are many pitfalls to avoid in healthcare predictive analytics, then where do you get started?  The most important starting point is to establish a fundamental data and analytic infrastructure upon which to build.  Deliberately but quickly move your organization up the levels of the Healthcare Analytics Adoption Model.  This model draws upon lessons learned from the HIMSS EHR Adoption Model and describes a similar approach for assessing the adoption of analytics in healthcare.  This model starts a level 1 foundation of an integrated, enterprise data warehouse combined with a basic set of foundational and discovery analytic applications.

1.  Start With an Integrated Data Warehouse and Analytics Measurement System

gearsEnterprise Data Warehouse

You need data across the entire continuum of care in order to manage patient populations.  This requires an enterprise data warehouse (EDW) platform.  An EDW is the central platform upon which you can build a scalable analytics approach to systematically integrate and make sense of the data.

Health Catalyst deploys a unique Late-Binding™ Data Warehouse that enables healthcare organizations to automate extraction, aggregation and integration of clinical, financial, administrative, patient experience and other relevant data and apply advanced analytics to organize and measure clinical, patient safety, cost and patient satisfaction processes and outcomes.

Foundational Dashboard and Reporting Applications

Today, most of your analytic resources are tied up doing manual reporting.  New requests often take weeks or even months to clear the queue.  You need to free up these analytic resources to focus on identifying and solving real problems instead of producing reports.

Health Catalyst provides a series of Foundational Applications which significantly accelerate the manual reporting work of many analysts into near real-time executive and department reports and dashboards.  For example,  Population Explorer, one of our most popular apps can produce the equivalent of 1000 dashboards with over 100 metrics.

Discovery Applications

Making sense of your volumes of population health data can be overwhelming.  You need a systematic way to understand and evaluate the important opportunities and allocate resources to work on those key processes.

Health Catalyst provides a series of Discovery Applications that will help you prioritize the opportunities within your system and help you focus on the areas of greatest opportunities in terms of quality of care improvement, cost reduction, revenue enhancement or patient injury prevention.

Once you have your foundational enterprise data warehouse and analytics applications in place, you are better prepared to begin more advanced analytic tasks such as predictive analytics in the higher level of analytics adoption.

2.  Use the Three Basic Steps of Predictive Modeling

predictive-modeling

The following is a simple schematic of the predictive modeling process.  For predictive analytics to be effective, Lean practitioners must truly “live the process” to best understand the type of data, the actual workflow, the target audience and what action will be prompted by knowing the prediction.

  1. The first step is to carefully define the problem you want to address, then gather the initial data necessary and evaluate several different algorithm approaches.
  2. Step two refines this process by selecting one of the best performing models and testing with a separate data set to validate the approach.
  3. The final step is to run the model in a real world setting.

The more specific term is prescriptive analytics, which includes evidence, recommendations and actions for each predicted category or outcome.  Specifically, prediction should link carefully to clinical priorities and measurable events such as cost effectiveness, clinical protocols or patient outcomes. Finally, these predictor-intervention sets are best evaluated within that same data warehouse environment.

So many options exist when it comes to developing predictive algorithms or stratifying patient risk. This presents a daunting challenge to health care personnel tasked with sorting through all the buzzword and marketing noise.  Healthcare providers need to partner with groups that have a keen understanding of the leading academic and commercial tools, and the expertise to develop appropriate prediction models.

Follow 4 Key Lessons Learned for Adopting Predictive Analytics in Healthcare

Given that predictive analytics are 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.  Fortunately for healthcare, there are numerous existing models from other industries that can be combined with past healthcare examples to ease some of the potential pains and pitfalls.  Highlights of some those key lessons include:

  1. Don’t confuse more data with more insight:  While many solid scientific findings may be interesting, they do little to significantly improve current clinical outcomes.
  2. Don’t confuse insight with value:  While many solid scientific findings may be interesting, they do little to significantly improve current clinical outcomes.
  3. Don’t overestimate the ability to interpret the data:  Sometimes even the best data may afford only limited insight into clinical health outcomes.
  4. Don’t underestimate the challenge of implementation:  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.

The following Health Catalyst Executive Report, “4 Essential Lessons for Adopting Predictive Analytics in Healthcare”  expounds more in detail around each of these 4 lessons:

HC-banner-4-essential

In order to be successful, we feel that clinical event prediction and subsequent intervention should be both content driven and clinician driven.  Importantly, the underlying data warehouse platform is key to gathering rich data sets necessary for training and implementing predictors.  Notably, prediction should be used in the context of when and where needed – with clinical leaders that have the willingness to act on appropriate intervention measures.

In the end, the overall goal is to leverage historical patient data to improve current patient outcomes. Predictive analytics is a powerful tool in this regard.

Health Catalyst Predictive Analytics Solutions

Health Catalyst not only has the expertise to develop prediction models, but our underlying warehouse platform is key to gathering the rich data sets necessary for training and implementing predictors.Notably, our prediction is only used “in context” – meaning when and where needed, with clinical leaders that have the willingness to act on appropriate intervention measures. Most important, however, these predictor-intervention sets can best be monitored and measured within that same data warehouse environment where otherwise not possible.

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, we 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.

We have a number of analytic applications that can be used in predictive analytics initiatives, including  Population Explorer, Readmission Explorer, Predictive Readmission Analyzer, and Patient Flight Plan Predictor.  So when your request comes – 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.  If you have interest or questions on any of these applications, feel free to contact us or schedule a demo by filling out our online form.

Read More About This Topic

4 Essential Lessons for Adopting Predictive Analytics in Healthcare 
David K. Crockett, Ph.D., Senior Director of Research and Predictive Analytics

What is Data Mining and its Use for Predictive Analytics in Healthcare?
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
Dale Sanders, Vice President, Strategy

Why Predictive Modeling in Healthcare Requires a Data Warehouse (white paper)
David K. Crockett, Ph.D., Senior Director of Research and Predictive Analytics

Prescriptive Analytics Beats Simple Prediction for Improving Healthcare 
David K. Crockett, Ph.D., Senior Director of Research and Predictive Analytics

The Power of Geo-Analytics (and Maps) to Improve Predictive Analytics in Healthcare
David K. Crockett, Ph.D., Senior Director of Research and Predictive Analytics

Products and Services

    Collect key Knowledge Center content to share with colleagues or yourself by dragging it here.

    compose message

    Here are the articles you added to "My Folder" in the Health Catalyst Knowledge Center.