Better designed business intelligence can accelerate AI initiatives

There is a business intelligence suite in every health system. Its purpose is often broad and ambiguous, such as “to increase revenue and patient safety while complying with all regulations and standards.”

Business intelligence groups are called upon when the health system faces a particularly thorny challenge such as a new shared risk driver contract that requires careful risk mitigation, among other potential examples.

In the business world, business intelligence is usually conducted around a product: how many downloads, how many times the user clicks on the red button, etc. However, in healthcare, business intelligence is usually implemented only at the highest level: the number of lives saved, the dollars spent.

Healthcare BI teams spend time segmenting the number of patients, then tracking the results for these segments of patients. The problem, however, is that this top-down approach often overlooks the complexities of interventions in the middle. There are often a lot of confounding factors: What intervention actually worked?

Connecting patients to interventions and ultimately to outcomes requires that business intelligence teams spend more time on the details. This related understanding requires the same rigor that AI initiatives need to be successful, secure, and effective. And this understanding only happens through precise engineering.

At Penn Medicine, we’ve built integrated product teams around our AI applications. These product teams consist of data scientists, clinicians, and software engineers, to name a few.

This year we started including a Business Intelligence Analyst as well, to help us make better design decisions so that connecting patients to interventions and outcomes is feasible and easily reported via dashboards and reports.

In our first joint program, we deployed a machine learning application to better find incomplete patient records. Our pilot showed an immediate improvement, but those gains began to wane after deployment.

Because of the level of engineering that went into our BI dashboard, we got an insight into what was going on in those middle steps, and were able to quickly focus on the problem and make corrections.

Business intelligence in healthcare is about making the right decisions. Data science in healthcare is all about providing insights that allow for better decisions. Health systems that take advantage of the natural compatibility of these two disciplines are more likely to see better results faster.

Mike Droglis is the chief data scientist at Penn Medicine.

What do you think?

Written by Joseph

Leave a Reply

Your email address will not be published. Required fields are marked *

widespread fraud in reporting vaccine infections; Withheld data, pretending to "Safe and effective"

Tips for packing the perfect picnic basket