Follow this blog:
RSS

Predict who will end up in the hospital: There’s a $3 million algorithm for that

By | April 4, 2011, 8:41 AM PDT

Imagine there’s an algorithm that can predict if you’ll get sick 15 years down the road.

Soon there might be a math equation that can predict just this. The algorithm just has to ask the right questions.

Did the patient have recent lab tests? Does the patient also have diabetes? Or high cholesterol? Is the person not taking his medications? All these factors (and more) would theoretically be plugged into an equation. The goal is to predict if the person will likely end up in the hospital in the near future.

The Heritage Provider Network wants to bring this kind of predictive health to the table. It announced the $3 million Heritage Health Prize. The competition officially launched today, April 4.

Using health records and claims data, HPN believes it’s possible to predict the chance of hospitalization. The winning algorithm should be able to identify which patients will end up in the hospital.

The HPN hopes this algorithm will help lower the cost of hospitalization…especially the unnecessary cases. According to the American Hospital Association, 71 million people in the United States were hospitalized last year. But in 2006, more than $30 billion was spent on unnecessary cases. It’s a resource and cost issue. Both can be better managed.

Recently, Jonathan Gluck from the Heritage Provider Network, called the current system a sick care system during a presentation. Watch Gluck talk about how the algorithm might help us move towards a predictive health care system.

Now that’s one way to bring innovation into the health care system…and there might very well be an algorithm for that. Spending less on hospital costs, means more money for research and the development of treatments.

More than 100,000 data scientists will likely try for the prize, according to Computer World. The developers have two years to program the perfect algorithm and will have access to 700,000 (unidentified) patients’ electronic health records.

Creativity can go a long way to bring innovation into the health care system. And in this case, there’s a huge incentive to do so.

The startup company, called Kaggle, is organizing the competition. The company has already launched 16 data prediction competitions, Computer World reports.

Hat tip via Popular Science. Source: Heritage Health Prize

Start your week smarter with our weekly e-mail newsletter. It's your cheat sheet for good ideas. Get it.

Boonsri Dickinson

About Boonsri Dickinson

Boonsri Dickinson was a contributing editor for SmartPlanet from 2010 to 2012.

Boonsri Dickinson

Boonsri Dickinson

Contributing Editor

Boonsri Dickinson is a freelance journalist based in San Francisco. She has written for Discover, The Huffington Post, Forbes, Nature Biotech, Technewsdaily.com, Techstartups.com and AOL. She's currently a reporter for Business Insider. She holds degrees from the University of Florida and the University of Colorado at Boulder.

Follow her on Twitter.

Boonsri Dickinson

Boonsri Dickinson

In the unlikely event that Boonsri has a professional or financial relationship with a company she writes about, it will be prominently disclosed.

She writes for SmartPlanet and is not an employee of CBS.

If you liked this, don't miss...
8
Comments

Join the conversation!

Follow via:
RSS
0 Votes
+ -
RE: Predict who will end up in the hospital: There's a $3 million algorithm for that
Great! If this works then Health Insurance companies will know who not to insure.
Posted by kynth@...
4th Apr 2011
0 Votes
+ -
RE: Predict who will end up in the hospital: There's a $3 million algorithm for that
Unfortunately, as good an idea as this may be, it does not matter how good your data and systems are. Unless the healthcare system is going to be directed to turn people away, the medical dependants will still come and the wasted billions will always remain.
Posted by 12312332123
5th Apr 2011
0 Votes
+ -
RE: Predict who will end up in the hospital: There's a $3 million algorithm for that
Or just raise their rates so high that they have to drop the insurance. Or threaten the company who hired them with higher rates so they get fired on some pretext.

I can't imagine who on earth would think that this is to help patients!
Posted by slccom
5th Apr 2011
0 Votes
+ -
RE: Predict who will end up in the hospital: There's a $3 million algorithm for that
That's why you need to go to a doctor who is off the grid...country doctors or the like...that way big insurance won't know anything about you when you need hosp[italization
Posted by tech_ed@...
5th Apr 2011
0 Votes
+ -
RE: Predict who will end up in the hospital: There's a $3 million algorithm for that
Folks you are so right. Docs and nurses and probably the housekeepers can tell by the way they walk up to the desk whether someone needs to be admitted. That is not what they are looking for. They are looking for how to do the least and make the most money. The people who are running hospitals and hmo's and other insurance companies are just plain miserable cruel people. They have many excuses for their behavior; we have to pay the electric bill don't we? And they all get new cars every year for a bonus plus cash and such. After all, a non profit hospital has to do something to bleed off all that money and keep them in the non profit range of 2-3%. It is sure not the business I started in back in the early 60's. Trust me.
Posted by IMWeira
5th Apr 2011
0 Votes
+ -
"I'm sorry sir, your HPN Index is too high for us to even talk to you...."
"...I don't care how bad you feel, your numbers are too good for you to qualify for any services we provide--have a good day!"
Posted by klassman6
7th Apr 2011
0 Votes
+ -
RE: Predict who will end up in the hospital: There's a $3 million algorithm for that
What struck me on reading the rules and description was how
long the prediction time
period was. I?m an old time doctor and this is what I?ve seen
doctors, hospitals and
patients do:
?A lot of patients don?t follow through with treatments because
they simply can?t
afford it.
?Physicians don?t always accurately communicate or record
critical factors in their
admission decision. In other words, the reason why a patient
was admitted isn?t
always recorded.
?People get lazy and don?t follow instructions.
?Many doctors and nurses fail to communicate to the patients
what they need to
do to take care of themselves.
This is how I currently make my living, by making these
predictions. The only predictive
model that I?ve found that works has a prediction time line of zero
days. In other
words, simply by predicting whether a patient needs to still be
hospitalized tomorrow saves a
tremendous amount of lives and money.
If anyone is interested in my methodology, more details
atwww.medicalscientificresources.com.
Posted by DocOsgood
10th Apr 2011
0 Votes
+ -
Benefits would flow to insurers
I can certainly see the benefits of this to the health insurers.

Finding benefits to patients is less clear. If you wanted to maximize benefits to patients you would design the competition quite differently - like figuring out who might benefit from additional tests or follow-up vs. who wouldn't. Dr. Osgood's "predicting whether a patient needs to still be hospitalized tomorrow" would be a sub-set of that question.

But it is hard to believe anyone but the insurance companies would find value in the question as posed.
Posted by jimgggg
16th Apr 2012
Join the conversation
Formatting +
BB Codes - Note: HTML is not supported in forums
  • [b] Bold [/b]
  • [i] Italic [/i]
  • [u] Underline [/u]
  • [s] Strikethrough [/s]
  • [q] "Quote" [/q]
  • [ol][*] 1. Ordered List [/ol]
  • [ul][*] · Unordered List [/ul]
  • [pre] Preformat [/pre]
  • [quote] "Blockquote" [/quote]

Join the SmartPlanet community and join the conversation! Signing up is fast and free. Don't wait -- we want to hear your opinion!