Machine learning is a term that crops up often in healthcare lately, but it’s important to understand what really constitutes learning in this context. What some call machine learning is actually unintentional programming, but true learning is derived during the process of building a predictive model. This article delves into the nuts and bolts of a healthcare machine learning model and describes the training process a model undergoes to impact outcomes for patients. The key ingredient is data and the key deliverable is to complete the feedback loop so those responsible for managing care have actionable information at their disposal. Machine learning is beyond conceptual; it’s incorporated into a growing list of predictive models for various disease classifications.
Choose by Topic
Tom Davenport, one of the top three business/technology analysts in IT and one of Fortune’s top 50 business school professors, pioneered the concept of “competing on analytics” and is the opening keynote speaker for the 2017 Healthcare Analytics Summit. Having written/edited 18 books and more than 100 articles, Tom is cofounder of the International Institute for Analytics and senior advisor to Deloitte Analytics. Tom is known for making the most complex concepts accessible—a critical skill an industry as complex as healthcare. And when it comes to big data, Tom is dedicated to equipping healthcare leaders with a clear, useful understanding of what it means from a technical, consumer, and management perspective so health systems can make fact-based, data-driven decisions that lead to outcomes improvement.
What do health systems risk when they hold onto an older enterprise data warehouse (EDW) perspective? By thinking about the EDW as a tool for only historic data that’s not highly reliable and can’t support important decisions, organizations miss out on near real-time (NRT) reporting and valuable decision-making resources. Far from an outdated tool, today’s EDW is capable of meeting rising demands for timely, quality data. Health systems can ensure their EDW reaches its full potential by prioritizing it among their technology and properly supporting it—with the best equipment and human resources. The well maintained EDW is not stuck in the past, but rather, an invaluable tool to move healthcare analytics forward.
The Medicare Access and CHIP Reauthorization Act (MACRA) appears to be a reporting challenge for many healthcare provider systems with few resources for managing the menagerie of measures. Indeed, with more than 270 measures in play, many systems have yet to jump in, but the deadline is inevitable. A plan of action is possible by recognizing and acting on these eight challenge areas:
Challenge #1: High-level performance insight
Challenge #2: Defining measure specifications
Challenge #3: Data quality reporting requirements
Challenge #4: Benchmarking data
Challenge #5: Proactively increasing measures surveillance to enhance outcomes
Challenge #6: Strategically aligning measures on which to base risk
Challenge #7: Identifying measures with the largest financial impact
Challenge #8: Taking risk in multi-year, value-based contractsMid-to-large size provider groups need a strategy around MACRA quality measures and a tool to help them make sense of all the reporting requirements.
Workers in today’s healthcare systems need dashboards with more power, interactivity, and visual feedback than traditional static reports are able to provide. Users also need to understand how and where to make improvements based on the dashboard’s information. To provide such deep insight to the data, a healthcare dashboard should have the following characteristics: be easily accessible, display reliable data, contain relevant data, be up-to-date for the task at hand, and include trends and/or benchmarks. When the right type of dashboard is combined with a late-binding data warehouse, users will gain access to the knowledge their data holds to drive lasting and effective improvement initiatives.
There are limitless opportunities for machine learning in healthcare. Defined as, “a field of computer science that uses pattern recognition to identify historical relationships in large data sets using an algorithm to create a generalized model of behavior,” machine learning is one of the most important life-saving technologies ever introduced to healthcare for several reasons:
- Reduces readmissions.
- Prevents hospital acquired infections (HAIs).
- Reduces hospital length of stay (LOS).
- Predicts chronic disease.
- Reduces one-year mortality.
- Predicts propensity-to-pay.
- Predicts no-shows.
- Improves sepsis outcomes.
Digital magazine and website Computerworld has named Health Catalyst to its 2017 Best Places to Work in IT list. Health Catalyst joins 100 IT companies that are leading the way in employee satisfaction and engagement with generous salaries, exceptional benefits, ongoing learning, and more. According to Ken Mingis, executive editor of Computerworld, IT employee satisfaction is increasingly vital: “As technology moves to the strategic center of every business, the ability of the enterprise to attract and retain skilled IT talent has become critically important.” Some of the team member-reported attributes that make Health Catalyst a best place to work include:
- Above-market compensation.
- Great work-life balance, thanks to unlimited PTO, company holidays, a work-from-home policy, and maternity and paternity leave.
- Companywide bonus structure.
- Fitness benefits, including onsite gym with fitness classes.
- Education/training reimbursement.
Outdated technology and antiquated costing methodologies have left health system CFOs unable to see the true cost of the services they provide and impacts on patient outcomes. The move from fee-for-service to value-based contracts, however, means that CFOs need this information more than ever. Health Catalyst® has partnered with industry-leading health systems to develop a next-generation costing system: the CORUS™ Suite. Two integrated products comprise the suite:
- Activity-Based Costing delivers accurate and actionable data from across the continuum of care in a scalable and maintainable tool.
- Cost Insights analyzes and delivers early insights through a customizable dashboard powered by embedded logic and access to the most granular level of activity and costing data.
Learn from the Best in Healthcare Data Visualization at Health Catalyst University™ During HAS™ 2017
Too often, the hard work of collecting and transforming data into meaningful insights is betrayed by a critical step in the journey: the visualization. Data visualizations should always make data easily consumable and digestible and accelerate outcomes improvement. This is where the Health Catalyst University Visualization Track comes into play. It’s one of four tracks available leading up to the 2017 Healthcare Analytics Summit. Class attendees will learn how to:
- Describe why visualization is important
- Recognize commonly accepted presentation rules
- Identify weakness in existing visualizations
- Execute the critical steps for effective chart creation
Care management programs play a large part in many health systems’ population health strategies. However, these programs can consume a lot of resources. It is important to know if a care program is effective, and eventually, to show a positive ROI. Many roadblocks stand in the way:
- Complexity of Environment
- Prolonged Time to ROI
- Lack of Access to Disparate Data
- Difficulty Engaging the Patient