Machine Learning: The Life-Saving Technology That Will Transform Healthcare

Health Catalyst believes machine learning is the life-saving technology that will transform healthcare.

This technology challenges the traditional, reactive approach to healthcare. In fact, it’s the exact opposite: predictive, proactive, and preventative—life-saving qualities that make it a critically essential capability in every health system.

Health Catalyst is on a mission to help health systems save lives by making machine learning routine, actionable, and pervasive through™ (models built into every Health Catalyst application),™ (free, open source software), and its healthcare analytics platform (the foundation).

There are limitless opportunities for machine learning in healthcare

Some may ask whether this is just a technology fad or whether it will provide true value in healthcare.  Health Catalyst believes the introduction and widespread use of machine learning in healthcare will be one of the most important, life-saving technologies ever introduced.  We believe the opportunities are virtually limitless for the technology to improve and accelerate clinical, workflow, and financial outcomes.  The following are just a few examples:

  • Reduce readmissions. Machine learning can reduce readmissions in a targeted, efficient, and patient-centered manner. Clinicians can receive daily guidance as to which patients are most likely to be readmitted and how they might be able to reduce that risk.
  • Prevent hospital acquired infections (HAIs). Health systems can reduce HAIs, such as central-line associated bloodstream infections (CLABSIs)—40 percent of CLABSI patients die—by predicting which patients with a central line will develop a CLABSI. Clinicians can monitor high-risk patients and intervene to reduce that risk by focusing on patient-specific risk factors.
  • Reduce hospital Length-of-Stay (LOS). Health systems can reduce LOS and improve other outcomes like patient satisfaction by identifying patients that are likely to have an increased LOS and then ensure that best practices are followed.
  • Predict chronic disease. Machine learning can help hospital systems identify patients with undiagnosed or misdiagnosed chronic disease, predict the likelihood that patients will develop chronic disease, and present patient-specific prevention interventions.
  • Reduce 1-year mortality. Health systems can reduce 1-year mortality rates by predicting the likelihood of death within one year of discharge and then match patients with appropriate interventions, care providers, and support.
  • Predict propensity-to-pay. Health systems can determine who needs reminders, who needs financial assistance, and how the likelihood of payment changes over time and after particular events.
  • Predict no-shows. Health systems can create accurate predictive models to assess, with each scheduled appointment, the risk of a no-show, ultimately improving patient care and the efficient use of resources.

From Reactive to Predictive: How Machine Learning Saves Lives

Many healthcare organizations around the country have already started improving outcomes and saving lives by partnering with Health Catalyst and using its analytics.

Example:  Using machine learning to reduce hospital-acquired infections

Health Catalyst machine learning has already been deployed at multiple client sites with promising outcomes. A recent article published by Hospitals and Health Networks shares the approach and results of one, integrated health system who is using predictive analytics in its CLABSI prevention initiative. They want to be able to predict which patients are most likely to develop a CLABSI so clinicians can proactively intervene more quickly and deliver optimal care for her or his patient.

They use a Health Catalyst enterprise data warehouse (EDW) and analytics to bridge information gaps in its EMR data and to develop a complete picture of patients’ CLABSI risk. Models on the data were developed and tested by using algorithms such as logistic regression and random forest—the workhorses of the machine learning world. This led to the development of a CLABSI risk prediction model that is built into a unit-level dashboard used by nursing staff to identify patient-level care gaps. In addition to the risk score, the top three risk factors for each high-risk patient are displayed, providing immediate insight into specific actions that can reduce the CLABSI risk for these patients.

According to a recent Scottsdale Institute report on the project, the risk-prediction tool—which auto-mines the EHR daily—has successfully predicted 85 percent to 90 percent of CLABSIs.

The results in the first six months were impressive, including:

  • An 87% predictive accuracy with a false positive rate of only .16.
  • 20% decrease in CLABSIs.
  • 30% drop in harm events overall.

The Problem: Machine Learning Isn’t Routine in Healthcare

Machine learning is part of everyday life for most Americans, from navigation apps to internet shopping, and widely used in other industries, such as retail and banking. But it isn’t routine in healthcare because of the complexity and limited availability of data—and the lack of access to highly skilled data scientists and teams required to turn that data into meaningful improvements.

The Solution: Making Machine Learning Routine, Actionable, and Pervasive in Healthcare

Health Catalyst has a three-pronged solution to making the technology routine, actionable, and pervasive in healthcare:

1.— our models and strategy for building machine learning into every Health Catalyst product. machine learning models are appearing in applications across all Health Catalyst product lines and our strategy is to build models into every single Health Catalyst product.

Health Catalyst is an outcomes company, so we focus on how models will be used, what workflows they affect, and what decisions they drive. When we developed a predictive model for Central Line Associated Bloodstream Infections (CLABSIs), it wasn’t enough to just show which patients were at high risk for developing a CLABSI. The prediction was displayed along with the modifiable factors that were driving the risk, giving nurses actionable information about how to reduce the risk.

In addition to our CLABSI risk prediction model, Health Catalyst has leveraged to deploy numerous predictive models for clinical, financial, and operational decision support:

Health Catalyst has already built over a dozen predictive models for clinical, financial, and operational decision support:

  • CLABSI risk
  • CHF readmission risk
  • Diabetes future risk
  • Diabetes net likely complication
  • Forecast incurred but not reported claims/year-end expenditures
  • Propensity to pay risk
  • Appointments no show risk

Health Catalyst has even more predictive models planned for the first quarter of 2017:

  • Clinical and operational decision support for bowel surgery
  • Hip and knee surgery
  • Coronary artery disease
  • Geo-spatial network referral and leakage prevention
  • Early detection of CAUTI and sepsis
  • Expected mortality and length of stay
  • And more.

2.—our way of stimulating the adoption of machine learning in healthcare. is a community resource for learning and collaborating on machine learning in healthcare, with weekly blogs and broadcasts where users can learn and interact with the Health Catalyst data science team. It also includes a software package that automates tasks and lowers the barrier of entry to adopting this technology. is free and open source because we believe the faster we can get systems to adopt machine learning, the sooner they can start using analytics to improve care and save lives. With, health systems have the people to do data science right now. has three key benefits:

  1. Free access to machine learning educational and collaboration resources.
  2. Free, open-source machine learning software that democratizes machine learning by lowering barriers to entry.
  3. Makes deployment easy and ensures production quality.

3. Healthcare Analytics Platform—our platform is second to none as a backbone to the most complex machine learning problems.’s effectiveness is closely tied to Health Catalyst’s proven ability to integrate high-volume data from virtually every internal and external source available. Because multiple sources of data are required for machine learning to develop the models that drive predictive analytics, the technology is more effective the more data is present.

Most companies providing machine learning solutions require customers to figure out how to connect as many as 100 different data sources to make the technology work. By contrast, Health Catalyst’s Healthcare Analytics Platform integrates 120 different data sources, including the electronic health record (EHR), claims, key financial, operational and patient satisfaction systems.

Health Catalyst Knows How to Implement Technology in Ways that Deliver Outcomes

Most importantly, Health Catalyst combines these three technologies together with our proven improvement methodologies to produce meaningful outcomes improvement results.

Most existing machine learning solutions vendors provide academically-appealing, standalone models without an understanding of how to translate them into meaningful, scalable outcomes.  As a result, there are few substantive examples of widespread machine-learning aided outcomes improvement in healthcare.

On the other hand, Health Catalyst knows how to make outcomes improvement work.  We know how clinicians use data to make decisions. We understand the context in which machine learning insight needs to be delivered and the right time and modality to deliver that insight.  That knowledge is built into, scaling machine learning and outcomes improvement for use by virtually any organization. The bottom line is that health systems can save more lives and improve more care, while saving money at the same time.

Learn More About Machine Learning

Join Health Catalyst’s Machine Learning Community

Learn from your fellow citizen data scientists about how to use to start using machine learning within your health system.

Receive weekly notifications, learning tips and live broadcasts. Blog
Our blog focuses on healthcare data science, including machine learning, visualization, R, Python, the predictive packages, as well as using these tools to understand and improve population health outcomes.

How Makes Machine Learning Accessible to Everyone in Healthcare
Levi Thatcher, Director of Data Science

The Benefits of Machine Learning in Healthcare
Levi Thatcher, Director of Data Science

What Models has Health Catalyst Created with
Levi Thatcher, Director of Data Science

Applications of Healthcare Machine Learning
Levi Thatcher, Director of Data Science

View Webinar on How to Use

Levi Thatcher and his data science team hosted a webinar titled “Machine Learning Using A hands-on Learning Session” with several learning objectives:

  • Describe and install
  • Build and evaluate a machine learning model
  • Deploy interpretable predictions to SQL Server
  • Discuss the process of deploying into a live analytics environment

View Webinar On-Demand

Speak with Someone at Health Catalyst About Using Machine Learning

If you’re interested in learning more about using machine learning to improve outcomes, contact the Health Catalyst Data Science team or request a demo.

Join our growing community of healthcare leaders

Stay informed by getting the latest news and exclusive updates from Health Catalyst.

This site uses cookies

We take pride in providing you with relevant, useful content. May we use cookies to track what you read? We take your privacy very seriously. Please see our privacy policy for details and any questions.