3 actions to get a data analytics project off the ground

Jessica Kim Cohen - Print  | 

Five health IT experts gathered in Chicago to discuss how hospitals can ensure adequate returns on investment when deploying data analytics and health IT projects.

Here are three actions they shared at the Becker's Hospital Review 3rd Annual Health IT + Revenue Cycle Conference on Sept. 21.

1. Be creative. A team embarking on an analytics project doesn't need to start by collecting data. In a hospital setting, a plethora of information already exists in EHRs and feedback surveys. "There's a lot of data out there that we don't act on — BMI, high blood pressure, smoking status, cholesterol — and that's readily available," said Jake Dorst, CIO and chief innovation officer of Truckee, Calif.-based Tahoe Forest Hospital District. "If you think creatively, you can really make an impact."

Brent Stutz, senior vice president of the innovation lab Fuse by Dublin, Ohio-based Cardinal Health, highlighted how taking stock of available datasets can lead to novel ideas during the brainstorming phase of a project. "Make sure you save some budget aside or allow folks to be inquisitive," he said. "I think that's going to be an area that creates a lot of opportunity in an organization. Don't pin everything to a business case and ROI at least out of the gate, but allow people to use the data to be very inquisitive."

2. Identify the problem. The next step to developing an analytics project is identifying a concrete problem and determining whether the data strategy would prove fruitful. "Make sure you have a good business case for it," said Stephen Morgan, senior vice president and chief medical information officer of Roanoke, Va.-based Carilion Clinic. "A lot of times, and I'm assuming this is something everybody struggles with, we'll have solutions brought to us and we don't really have a problem."

Ian Christopher, chief technology officer and co-founder of the tech company Qventus, agreed. He shared how Qventus uses prescriptive analytics and machine learning to forecast how many patients will present at a given emergency room throughout the day. "There has to be that connection between the insight we've found and an action being taken," he said. "There has to be that kind of link between the two."

3. Define the metrics. To ensure a hospital board agrees to develop and deploy a data project, health IT professionals must delineate how the project will improve outcomes or save money — and what metrics the team will measure to ensure it's on track. "That's our problem in the smaller environment," Mr. Dorst said. "It's really proving to the people with the checkbook that we can actually save some money."

Joshua Lee, MD, chief health information officer and vice president of Maywood, Ill.-based Loyola University Health System, also spoke to the role of metrics when gaining administrative buy-in. "You have to be very, very precise in the target that you're going after, that you're pretty sure from the outset is going to produce a measurable remunerative benefit, before you start," he said. "Plan the measurement before you start and build to the measure."

More articles on data analytics & precision medicine:
NIH launches online 'toolkit' to help nurses bring genomics into patient care
Cleveland Clinic, Tempus to apply precision medicine for patients with advanced cancer
Brigham Health deploys Redox systemwide to consolidate patient data

© Copyright ASC COMMUNICATIONS 2019. Interested in LINKING to or REPRINTING this content? View our policies by clicking here.

To receive the latest hospital and health system business and legal news and analysis from Becker's Hospital Review, sign-up for the free Becker's Hospital Review E-weekly by clicking here.