Steward Health Care case study – DataRobot


[MUSIC PLAYING] If we do 1% reduction
in staffing, meaning that we staff
to volume and not staff to average census, we
actually are saving $2 million a year for eight hospitals,
and now we’re 38. If we actually do 0.1
reduction in the patient’s length of stay, $10
million in savings. So the ROI for us is already,
we’re already seeing that, and we’re only getting
better from it. It’s uncharted territory
for health care. A lot of people
use the buzz word of predictive analytics
or machine learning, but we’re actually doing it. We have a product that’s out
on the hospitals’ floors now and we see it work. We took our
historical data and we trended all of our
data to find out when our peaks and valleys were. So we were over-staffing
to our volume, and we needed to have
a better way to staff to the peaks and
valleys of our volume. DataRobot has the tools
to allow us to manipulate the data to learn from it. Health care is failing if
they are being reactive. We want to see how we’re
going to do next month and take proactive
measures to plan for those bumps
along the road, and I think machine
learning is the way that we’re going to do that. Not only can we tell
you that there’s going to be 20 patients arriving
through the ED tomorrow, we can tell you 20 of
what type of patients. That’s what we’re shooting for. The smarter you get,
the more you want. I never thought I’d be
doing this two years ago, but now that I’m
doing it, I can’t believe that we weren’t
doing it sooner. [MUSIC PLAYING]

Tags:, ,

Add a Comment

Your email address will not be published.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