Predictive Analytics News

How to Develop a Successful Predictive Analytics Tool

Predictive analytics tools can help simplify workflows for providers, but these tools should avoid certain pitfalls in order to be successful.

predictive analytics, ehr implementation, artificial intelligence

Source: Getty Images

By Kelsey Waddill

- Tom Selva, chief medical information officer of University of Missouri Health Care, knows the power that predictive analytics have to transform the healthcare industry.

Listen to the full podcast to hear more details. And don’t forget to subscribe on iTunes, Spotify, or Google Podcasts

In the busy atmosphere of clinical care, predictive analytics offers the opportunity to catch a patient’s symptoms before they begin to decline or worsen, Selva explained on Healthcare Strategies. Predictive analytics covers a wide range of tools from artificial intelligence and machine learning to effective reporting.

Predictive analytics are often used in operations such as predicting supply chain issues and scheduling applications. But Selva has been focused on using these tools to prevent “never events,” events in which a patient dies from something preventable, specifically sepsis never events.

University of Missouri Health Care needed to improve its mortality rate by reducing the prevalence of sepsis. To address this issue, Selva and his team introduced the national early warning score (NEWS). 

NEWS is a scoring schema that takes into account all of the early signs of sepsis that are easy to miss, such as oxygen saturation, respiratory rate, and blood pressure. The score fits into the EHR platform to alert clinical professionals to a potential sepsis case.

Introducing a tool like NEWS into the clinical setting requires dozens of decisions from how to set the threshold for the alert down to what color should the alert be. To Selva, one of the most critical factors in the tool’s success was having a nurse informaticist who was personally invested in the outcomes and who served as an advocate for the tool.

Having the score is an important step forward, but to Selva the success of a predictive analytics tool largely hinges on what the clinical staff does with the information that the tool provides. 

Selva also emphasized that predictive analytics tools will not be effective if they fail to provide the data to back up their results.

“When you think about machine learning, artificial intelligence, you need to understand that in the medical world, we care more about why you did what you did than what you did,” Selva shared. 

There has been a push to use these tools to simplify healthcare. However, Selva explained that if a predictive analytics tool is so simplified that it provides a result but cannot substantiate its diagnosis, it can break down the user's trust in the tool.

“We have to acknowledge that healthcare is inherently complex and you can only simplify it so far and then it can become dangerous,” Selva said.

Still, Selva predicted a great future for predictive analytics tools, particularly in producing savings.

“Healthcare is a low margin business in America. So predictive analytics will be applied to make it more efficient so that we're spending the dollars where they need to be spent and more effectively, which is good for everyone,” Selva concluded.

________________

About Lark Health

Lark utilizes proprietary A.I technology to simplify the prevention and management of chronic illnesses. Its expanding platform provides users 24/7 access to a virtual care coach through an engaging and personalized text-like interface. Lark currently works with 2 of the 3 largest health plans.

Do Not Sell or Share My Personal Information
©2012-2024 TechTarget, Inc. Xtelligent Healthcare Media is a division of TechTarget. All rights reserved. HealthITAnalytics.com is published by Xtelligent Healthcare Media a division of TechTarget.