×

Welcome to the Slashdot Beta site -- learn more here. Use the link in the footer or click here to return to the Classic version of Slashdot.

Thank you!

Before you choose to head back to the Classic look of the site, we'd appreciate it if you share your thoughts on the Beta; your feedback is what drives our ongoing development.

Beta is different and we value you taking the time to try it out. Please take a look at the changes we've made in Beta and  learn more about it. Thanks for reading, and for making the site better!

California Healthcare Provider Wants Illness-Predicting Algorithm

timothy posted about 3 years ago | from the make-hospitals-smaller dept.

Medicine 341

alphadogg writes "The Heritage Provider Network wants to do for healthcare what technology in the film Minority Report did for police work. In other words, it wants to use technology to pre-emptively predict when illness is likely to strike and take measures to prevent costly hospitalizations. This week Heritage announced that it was offering a prize of $3 million for any developer who successfully created a 'breakthrough algorithm that uses available patient data, including health records and claims data, to predict and prevent unnecessary hospitalizations.'"

cancel ×
This is a preview of your comment

No Comment Title Entered

Anonymous Coward 1 minute ago

No Comment Entered

341 comments

Really? (2, Insightful)

Anonymous Coward | about 3 years ago | (#35665058)

'breakthrough algorithm that uses available patient data, including health records and claims data, to predict and prevent unnecessary hospitalizations.'"

By removing them from the list?
"Sorry, you're statistically not interesting for us anymore..."

Re:Really? (5, Interesting)

click2005 (921437) | about 3 years ago | (#35665132)

or...

"Statistically you're likely to get so your premiums are going up by 588%"

or...

"There is a 22% likelihood that one of your kidneys will fail within 5 years, 44% for leukaemia blah blah.. how about one of our body scans??"

or..

Patient: Can I see a doctor?
HCP: Doctor? We dont need doctors.
Patient: But I'm sick. I think my kidneys are failing.
HCP: I dont think thats very likely. According to our software you only have a 2% chance of kidney failure. Its probably just gas.

Re:Really? (0)

Anonymous Coward | about 3 years ago | (#35665382)

Based on past experience a MSCP could easily handle this task in a manner that would be both computationally efficent and ethically sound. It's well worth looking into proper certification for anyone that's found their career in a hole and wants to find new and exciting opportunities.

Predict and disqualify customers, you mean. (5, Insightful)

mattcsn (1592281) | about 3 years ago | (#35665082)

"New care plans and strategies" sounds like HMO-speak for "cut off people before they cost us more than we soak in from them".

Re:Predict and disqualify customers, you mean. (0)

maxume (22995) | about 3 years ago | (#35665116)

Right, because people aren't at all inclined to be morons or to dislike going to the doctor, so they never-ever put off a visit until things have reached the point of horror.

The only explanation for a care provider doing anything ever is that they are trying to screw their patients (the company running this contest really is one of the ones that sends bills to the insurance industry).

Re:Predict and disqualify customers, you mean. (1)

Kokuyo (549451) | about 3 years ago | (#35665412)

You act as if going to doctors was healthy in the first place. If you have a good one, certainly, but my experience with them makes me wary.

Re:Predict and disqualify customers, you mean. (1)

maxume (22995) | about 3 years ago | (#35665508)

Sure. I only go to the doctor when I think I need it. I can see the value in doing more routine checkups, but I would have to pay for them directly, so I don't.

I'm talking about the people that, for example, wait until their abscess reaches grapefruit size, or ignore their persistent hacking cough, or ignore the loss of feeling in their feet.

Re:Predict and disqualify customers, you mean. (1)

oh_my_080980980 (773867) | about 3 years ago | (#35665924)

Nice one troll. Yeah because it's the sick person's fault. It's not like health insurance companies deny people coverage or treatment because of prior conditions or cost. It's not like Pharmaceutical companies push drugs that patients don't need. It's not like health care bills aren't expensive so people have to chose between food on the table and going to the doctor.

Assuming you're specious argument hols water, you ever think that the person with a large abscess has one because the don't have health insurance or can't afford the treatment.

Ignorant SOB.

Well with the stupid rules in place (1, Troll)

Shivetya (243324) | about 3 years ago | (#35665240)

especially with the nationwide health care bill what did you expect?

The system will allow people to forgo insurance until they are sick. This is akin to crashing your car and then buying insurance and filing a claim. As such, insurance companies need to come up with methods to better manage those they are covering to buffer against those who will abuse the system.

The real problem of course is, people have no incentive to care about their health. Oh, I know, but you say, who doesn't care about their health. Its a matter of degrees. Tell me how many tubbies you know who think they need to change. OK, now tell me how many are doing something about it. When health care's cost do not directly affect the people using it a large number will do nothing to improve their health. They already have been forgiven the responsibility of paying for their health care and quickly associate that with, if anything goes wrong they won't have to pay for that either. Hence the costs are purposely forced upon those who do pay and the companies managing it. Its a perfect system designed to force people out of private insurance for the exact fear you espouse.

So time to break out the predictions. It should not be too hard to get a few things down quickly, there are issues that are attributable based on race, sex, and known health issues (weight, smoking, etc), then compare to results from actual visits if any. It would be a fascinating project. I don't know why anyone is surprised by any of this, my only surprise is that these are not common place now.

Ask yourself this : Why should someone pay for your healthcare if your not an active participant in improving your own health. Then realize that a sizable portion will tell you to go F off all the while not screwing over themselves by not doing anything or worse doing the wrong things.

Re:Well with the stupid rules in place (1)

WonderingAround (2007742) | about 3 years ago | (#35665334)

There's no arguing it's a twisted concept but these things are the heart of venture capitalism and has the possibility to become a major field in an already struggling healthcare system where the consumer may have an advantage if in good health and suddenly have to deal with intense pain or suffering at some point but get their life back and their money's worth.

Tubbies, huh? (0)

Anonymous Coward | about 3 years ago | (#35665340)

Oh, I know, but you say, who doesn't care about their health. Its a matter of degrees. Tell me how many tubbies you know who think they need to change. OK, now tell me how many are doing something about it.

Hammer, nail, head.

It does go further than just tubbies, though. So, 'healthy' Slashdotters... Who among you works far too many hours per week? Who among you has gone for In-n-Out, or pizza, or Chinese, instead of a dressing-less salad with some sort of rabbit-food smoothie? WHO AMONG YOU DOES NOT KNOW EMPTINESS? ...Uh, sorry, Thusla Doom slipped in there somehow.

Captcha: Chubbier. Hah!

Re:Tubbies, huh? (1)

somersault (912633) | about 3 years ago | (#35665574)

I was starting to get slightly lumpy by the time I hit 25 (I'm tall so it's harder to notice), but I tried various things and when I read about the Atkins diet and ended up eating "low GI" food, I've been in great shape, and more importantly, I feel a lot better.

I'd eat a 12" pizza for one meal if it had a wholemeal base.

Chinese stir-fry is perfectly healthy, if you make it yourself at least and have wholemeal noodles or brown rice instead of normal noodles/white rice. Stuff like steamed dumplings would be fattening (because of the dough, not the meat), but it's okay to have that kind of thing every now and then.

Smoothies with oats in them are amazing btw :) As for dressing-less salad, why? Put as much mayonnaise and dressing on it as you want, you're not going to get fat. However, you only need a bit if you don't want to feel like you're eating a bowl of condiments.

Re:Well with the stupid rules in place (3, Informative)

Isaac-1 (233099) | about 3 years ago | (#35665570)

The real problem is cost of health care, about 6 months ago I fell and broke my back. I have decent if not great insurance, and the treatment for my break (single level split compression fracture if your interested) has been nothing more than a brace and monthly follow up x-rays (and one CT at 4 months)and doctor visits. I was transported to the hospital by ambulance on a back board (cost about $750, $300 out of pocket, kept in the hospital for 3 days base level observation, fall happened on a weekend and I could not be fitted for a $750 custom fitted plastic and foam brace until Monday, hospital bill about $15,000 for 35 hour stay, another $2,000 or so for the 2-3 hours in the ER before being admitted), plus about $515 per month for a couple of x-rays and spinal specialist visits. Total bill upwards of $25,000 so far, out of pocket around $4,000 .

Re:Well with the stupid rules in place (1)

maxume (22995) | about 3 years ago | (#35665676)

For $15,000 we should just tack on another couple thousand and require that they provide you a detailed itemized list of what costs they are using to justify that bill.

Re:Well with the stupid rules in place (1)

wisty (1335733) | about 3 years ago | (#35665736)

>>> hospital bill about $15,000 for 35 hour stay

Holy shit, where did they put you? The Paris Hilton? Did they throw in Paris Hilton? No, your back was bad ...

I hope the food was good ...

Re:Well with the stupid rules in place (1)

cvtan (752695) | about 3 years ago | (#35665860)

My wife has had two knee replacements at $25000 each. Her pacemaker cost about $20000. Your costs seem high.

Re:Well with the stupid rules in place (2, Informative)

Inda (580031) | about 3 years ago | (#35665916)

For comparison in the UK:

CT scan just cost my insurance £450 ($721.22) and the nurse told me that was cheap.

Wife just had gaul bladder removed. In and out of hospital within 24hrs. £4,750 ($7,612.83)

NB. Some of us in the UK get private health care through work.

Re:Well with the stupid rules in place (1)

BVis (267028) | about 3 years ago | (#35665760)

"As such, insurance companies need to come up with methods to better manage those they are covering to buffer against those who will abuse the system."

Or, you know, they could stop screwing over their customers to the point where having insurance is actually affordable, and people would buy it before they got sick.

"They already have been forgiven the responsibility of paying for their health care"

I don't know about you, but I'm sure paying for my health care. To the tune of $1200/month.

"Why should someone pay for your healthcare if your not an active participant in improving your own health"

Maybe because some people actually give a shit about people other than themselves?

Re:Well with the stupid rules in place (3, Insightful)

famebait (450028) | about 3 years ago | (#35665768)

You seem to be arguing from the following premise:
        "costly treatments make people take more care of their health"

Until you bring forth Extraoridnary Evidence (tm) for this Extraordinary Claim (tm), please forgive us for ignoring your random speculations, and for frowning upon your attempt at presenting those ramblings as fact.

You might be surprised to learn that there are many other countries besides the US, employing many different models of health care funding. A first stab at checking your assumtions (don't knock it 'till you've tried it) would be to compare some industrialised countries in terms of public health, healtcare spending, and typical cost to patients.

Seriously - would you or anyone you know actually think "I never really considered getting a serious helath problem, but it the treatment is free, why the hell not?", or is it just "those other people" you collectively accuse of this insanity?

Re:Well with the stupid rules in place (0)

Anonymous Coward | about 3 years ago | (#35665962)

"Ask yourself this : Why should someone pay for your healthcare if your not an active participant in improving your own health." ...and Americans still like to say they are a "Christian Nation." I don't remember Jesus saying "I got mine, why should I help you?"

Your health-care "system" is a world-wide joke. The fact that there isn't a single nation on this planet following the Nixon/Kaiser Permanante model for screwing over patients doesn't bother you...because you dismiss it all as "socialism." Huuah, go America.

Re:Well with the stupid rules in place (1)

oh_my_080980980 (773867) | about 3 years ago | (#35665980)

Death Panels....seriously how stupid do you have to be to not understand why people would not be for this. The biggest issue has to do with misuse. Since an INSURANCE COMPANY is asking for this, it is highly likely, based on past evidence, that insurance companies would deny people coverage. That's a fact.

There other point is THEY'VE TRIED THIS ALREADY.....remember DEATH PANELS. In the healthcare reform debate they talked about creating efficiencies in healthcare by using a similar approach and politicians took that to mean rationing and death panels.

Take your head out of your ass potsy.

Re:Predict and disqualify customers, you mean. (1)

Anonymous Coward | about 3 years ago | (#35665434)

Since the Supreme Court decided that HMOs are in business to make money and not to provide care, I'd say that an announcement of a $3 million prize for an algorithm to "avoid hospitalization" is an end-run around the pre-existing condition exclusion ban.
After all, it's sound business strategy to tailor your business practices to comply with new laws. And statistically proving that insuring a person's health is a sucker bet isn't discrimination.
Until you consider the human suffering.

Re:Predict and disqualify customers, you mean. (2)

PJ6 (1151747) | about 3 years ago | (#35665514)

"New care plans and strategies" sounds like HMO-speak for "cut off people before they cost us more than we soak in from them".

This gets at the heart of why a for-profit model may be inappropriate for some industries.

Re:Predict and disqualify customers, you mean. (0)

Anonymous Coward | about 3 years ago | (#35665700)

You say that like healthcare isn't just a big pyramid scheme...

Re:Insurance against risk is a socialist model (1)

OldHawk777 (19923) | about 3 years ago | (#35665712)

Insurance against risk is a republic (faux-elitist class) business model.

The model says we insure you against reasonable risk? We ensure ourselves from any risk!

Recent example: Derivatives mitigate losses by insuring faux-elitist from their piss-poor decisions.
Recent example: Derivatives ensured earnings for the scam-elitist sellers and assured losses for US, EU... pensions and taxes.

Global corporate socialism/welfare economies work well for others, but never for EU or US folks/citizens.

Socialism/welfare economies are not capitalist economies and are republics, but never democracies.

 

Hmm (2)

Spad (470073) | about 3 years ago | (#35665130)

It's called preventive medicine; the rest of the world has been doing it for some time now...

Re:Hmm (3, Insightful)

goodmanj (234846) | about 3 years ago | (#35665586)

The question is, is this preventive medicine or preventive insurance?

With single-payer health care, this distinction doesn't exist.

Re:Hmm (0)

Anonymous Coward | about 3 years ago | (#35665652)

Insurance companies are supposed to be risk averse. There are ways other than pre-existing conditions to get someone off the health care roles. It's called price them out of the market.

Re:Hmm (0)

Anonymous Coward | about 3 years ago | (#35665812)

The rest of the dumb world that is. I'm not going to take crappy medication for something that I MIGHT get.

Likely to get sick: no healthcare for you! (2)

captainpanic (1173915) | about 3 years ago | (#35665152)

The whole idea of a healthcare insurance is to spread the risk between people... therefore it's pretty much necessary that healthy and unhealthy people pay the same.
If you have a cheap healthcare for all healthy people, and then an unaffordable one for those more likely to get ill, the system crashes, doesn't it?

An insurance is a protection against future problems. Healthy people also must invest in their own unavoidable loss of health.

Re:Likely to get sick: no healthcare for you! (5, Insightful)

Chrisq (894406) | about 3 years ago | (#35665242)

The whole idea of a healthcare insurance is to spread the risk between people... therefore it's pretty much necessary that healthy and unhealthy people pay the same. If you have a cheap healthcare for all healthy people, and then an unaffordable one for those more likely to get ill, the system crashes, doesn't it?

An insurance is a protection against future problems. Healthy people also must invest in their own unavoidable loss of health.

You are assuming that the aim of healthcare insurance is to provide healthcare to people efficiently rather than to maximise profit for the providers.

Re:Likely to get sick: no healthcare for you! (1)

captainpanic (1173915) | about 3 years ago | (#35665358)

You are assuming that the aim of healthcare insurance is to provide healthcare to people efficiently rather than to maximise profit for the providers.

Absolutely!
You may call me a communist now :-)

Re:Likely to get sick: no healthcare for you! (1)

Chrisq (894406) | about 3 years ago | (#35665484)

You are assuming that the aim of healthcare insurance is to provide healthcare to people efficiently rather than to maximise profit for the providers.

Absolutely! You may call me a communist now :-)

You're a communist! (but by that reckoning so am I. I would never swap our NHS for an insurance based system)

Re:Likely to get sick: no healthcare for you! (2)

microbox (704317) | about 3 years ago | (#35665590)

You may call me a communist now :-)

False dichotomies aside, he is pointing out that the healthcare insurance companies are /not/ going to do the job ethically or even properly. Follow the money to see how incentives skew activity. In Canada, billing costs less than 10% of that in the USA. Of course, billing costs go straight to the insurance company. You guys are getting ripped off big-time.

The alternative is not communism. That would be the type of black-and-white thinking one could expect from aspergers.

Re:Likely to get sick: no healthcare for you! (1)

cheekyjohnson (1873388) | about 3 years ago | (#35665930)

That would be the type of black-and-white thinking one could expect from aspergers.

Really? I'd expect it from the average member of society.

Re:Likely to get sick: no healthcare for you! (1)

geminidomino (614729) | about 3 years ago | (#35665498)

You are assuming that the aim of healthcare insurance is to provide healthcare to people efficiently rather than to maximise profit for the providers.

Personally, I consider the fact that this is no longer the case to be rather compelling evidence that humanity is beyond redemption at this point.

Push the button and reboot. Maybe the cockroaches will fuck up less.

Re:Likely to get sick: no healthcare for you! (1)

Motard (1553251) | about 3 years ago | (#35665610)

Health insurers are definitely not aiming to maximize the profits of providers. Quite the opposite in fact.

Re:Likely to get sick: no healthcare for you! (1)

geminidomino (614729) | about 3 years ago | (#35665788)

Minor Nit: The term "Provider" within the medical billing system is horribly overloaded. Both physicians and insurance companies can and often are referred to as "providers" (with different modifiers, hopefully).

Re:Likely to get sick: no healthcare for you! (1)

Motard (1553251) | about 3 years ago | (#35666004)

No, in this context, insurers are not providers. That term is used because providers can include midwives, ambulance companies, durable medical equipment sellers, pharmacies, etc.

Re:Likely to get sick: no healthcare for you! (3, Insightful)

mcelrath (8027) | about 3 years ago | (#35665662)

And that, in a nutshell is what's wrong with for-profit insurance providers: the profit motive of the company is directly opposed to the health motive of the customer.

Because of that very fundamental fact, the only medical insurance scheme that makes any sense is a socialized one.

Re:Likely to get sick: no healthcare for you! (1)

Motard (1553251) | about 3 years ago | (#35665864)

Incorrect. Live, healthy premium payers are good for health insurers. Live, unhealthy premium payers are bad for health insurers. And dead people don't pay premiums, so they're no use to health insurers.

The profit motive ensures that the plans remain viable and competition ensures that the insurers will keep the premiums as low or lower than their competitors. And if an employer suspects that the insurers are raping them, then they can self insure and just pay an administrative fee to a Third Party Administration company and assume the risk themselves or with them help of a reinsurance company.

Profitable healths plans are good as long as there is competition. It's sustainable. If you look at the U.S. national debt, you'll immediately see that our government is not so good about managing these things, so we are left worrying about the sustainability of Medicare and Social Security.

Re:Likely to get sick: no healthcare for you! (5, Insightful)

mcelrath (8027) | about 3 years ago | (#35666012)

Incorrect. If an insurance company has the opportunity to remove unprofitable members from the rolls, they will take it. If they have the opportunity to refuse treatment, they will take it. If they can select which new customers they will take and which ones they won't, they will use that. If they can write long obtuse contracts outlining things they won't pay for, and have their army of lawyers enforce it, they will do it.

It is a general fact about any kind of insurance that the interests of the insurer are misaligned with the interests of the insuree. They're predatory industries who rely upon promising more than they will deliver and tricking their customers wherever possible.

Only in the circumstance that the insurer is required to insure everyone does the profit motive go in the direction of the patient's interests (in the form of preventative care). Preventative care is a long term investment that wall street doesn't see.

Re:Likely to get sick: no healthcare for you! (1)

Kludge (13653) | about 3 years ago | (#35665366)

... therefore it's pretty much necessary that healthy and unhealthy people pay the same.

Yeah, the fat guy who smokes, drinks, and eats cheeseburgers everyday should pay the same as me, the skinny guy who eats salads everyday. I have really low life insurance rates compared to most other people. Why should health insurance be different?

If you have a cheap healthcare for all healthy people, and then an unaffordable one for those more likely to get ill, the system crashes, doesn't it?

The system won't crash entirely as long as there are some people who are willing to pay for health care. However, that is the problem with US health care, fewer and fewer people can afford it because doctors and hospitals limit the amount of care available, driving up costs.

Re:Likely to get sick: no healthcare for you! (2)

timeOday (582209) | about 3 years ago | (#35665640)

Yeah, the fat guy who smokes, drinks, and eats cheeseburgers everyday should pay the same as me, the skinny guy who eats salads everyday.

Based on nothing but the above, your risk is probably lower than his, but it might or might not be. You might simply have bad genes. Without accurate predictive models, what you call "risk factors" don't have much predictive value - that's what risk fators are - factors in some sort of model. So I support better modeling.

However, we must apply them carefully. I don't think insurance rates should reflect anticipated health costs, but rather the individual's choices related to anticipated health care costs. There's no point punishing or incentivizing things that aren't choices, such as race. Without good predictive models, it's all too easy to manipulate people by stirring up vindictive emotions on the false premise that only irresponsible people would ever need treatment.

Even so, we still necessary to quantify risk factors that aren't personal choices to effectively practice (preventative) medicine.

Re:Likely to get sick: no healthcare for you! (1)

kingramon0 (411815) | about 3 years ago | (#35665488)

The whole idea of a healthcare insurance is to spread the risk between people... therefore it's pretty much necessary that healthy and unhealthy people pay the same.

That's not how car insurance works. People with a higher risk of accidents pay more. That's why your premiums go up after a speeding ticket, and go down when you get married / have kids. Your premiums change with your statistical level of risk.

I don't know how other forms of insurance work, so maybe someone can enlighten me, but I assume they also charge based on risk.

So, why would health insurance be different?

Re:Likely to get sick: no healthcare for you! (3, Insightful)

Sponge Bath (413667) | about 3 years ago | (#35665612)

"So, why would health insurance be different?"

Ultimately it is different because without car insurance you walk, without health insurance you die [1]. Maybe you are fine with the concept of the poor and people who don't live the way you feel they should just dieing of treatable illness, but that fundamental difference between car insurance and health insurance remains.

[1] Earlier than necessary due to treatable illness you can't afford.

Re:Likely to get sick: no healthcare for you! (1)

maxume (22995) | about 3 years ago | (#35665734)

Insurance still seems like a terrible way to socialize medicine.

(Mandatory insurance does apparently work a great deal better than mandatory treatment without mandatory insurance)

Re:Likely to get sick: no healthcare for you! (1)

Qzukk (229616) | about 3 years ago | (#35665740)

It's not how any insurance works.

You get your car wrecked and your insurance company pays to fix it, you go on.
You get your house burned down and your insurance company pays to fix it, you go on.
Your employee falls down on the job and your insurance company pays to fix it, you go on.
A customer falls down and your insurance company pays to fix it, you go on.
You die and your insurance company pays to bury you, your kids go on.

You get cancer and your insurance company pays and pays and pays and pays and...

Sure, the model might work for something simple like a broken leg, but we have too many incurable chronic diseases that are expensive to treat, and to the chagrin of the moral police, not all of them are the fault of the sufferer.

Re:Likely to get sick: no healthcare for you! (0)

Anonymous Coward | about 3 years ago | (#35665974)

Of the people close to me that have gotten cancer (ages 50-60 or so), 2 have died within a year or 2 and 1 received about 12-15 months of treatment in two separate bouts, and has not required aggressive treatment for about 10 years (I don't know the details but would assume there are some tests and whatnot that are ongoing).

Cant do it. Not allowed. (1)

140Mandak262Jamuna (970587) | about 3 years ago | (#35665164)

Kodak, Microsoft, IBM, Motorola and about 25 more companies claim they have already patented it. When pressed they admitted they have pretty much patented everything that could ever be done on a computer.

Re:Cant do it. Not allowed. (1)

Sponge Bath (413667) | about 3 years ago | (#35665654)

No doubt. People have masturbated to erotic images for all recorded history, but one or more of these companies likely have patents covering that activity in combination with the internet.

The Actual Problem in Pursuing this Prize (5, Informative)

sarbonn (1796548) | about 3 years ago | (#35665174)

As a healthcare professional who does data analysis for a number of hospitals, this sounds like a great idea, but at the same time I also realize the limitations of conducting this algorithm process. To begin with, HIPAA compliance laws make it very difficult to share specific data about patients, which means someone trying to put together this type of information, or statistical based program process, is going to have to do it sans data, creating false data that isn't actually real case information. Which then means that even if you are capable of providing an algorithm that fulfills the functionality, the designers of the prize program are most likely going to stand up and say that it's not transferrable to real cases because you didn't account for the specific variables that are present in real world data (meaning you can't predict data that is actually already there due to the amount of errors in guesswork involved). If they made available the actual data they want extracted, this might be a possible process. But until they do, it is like guessing statistical outcomes of a presidential race without knowing anything about the people who might be actually running.

Re:The Actual Problem in Pursuing this Prize (0)

Anonymous Coward | about 3 years ago | (#35665392)

It's it possible to scrub the data of all identifiable info? Give the developer "real data" with no means to trace it back to actual persons? Or even use the recently deceased as possible data mine?

I would think there are ways to do it and keep privacy intact,

Sean D.

Re:The Actual Problem in Pursuing this Prize (1)

ColdWetDog (752185) | about 3 years ago | (#35665466)

Oh stop. This is very easy. No coding required:

Your gonna die.

Could do this in Visual Basic!

Re:The Actual Problem in Pursuing this Prize (2)

Motard (1553251) | about 3 years ago | (#35665696)

They will be providing access to a de-identified database. So you should have all the data that would be available in a real world application.

Re:The Actual Problem in Pursuing this Prize (1)

FridayBob (619244) | about 3 years ago | (#35665982)

As a healthcare professional who does data analysis for a number of hospitals, this sounds like a great idea, ...

How can you say that when the US health insurance industry has such an awful track record of discriminating against applicants who admit that they have a medical history with any of a very long list of possible ailments? AFAIK, this is not something that people are subjected to anywhere else in the world. This algorithm represents a double-edged sword that, once they have it (if they ever get it), they will use the wrong way: "What's that, no history of any medical problems, you say? Well, our algorithm predicts that later on you will likely develop costly problems anyway, so your application is rejected. Get your health insurance elsewhere!"

I'm confused (1)

OzPeter (195038) | about 3 years ago | (#35665178)

The Heritage Provider Network wants to do for healthcare what technology in the film Minority Report did for police work. In other words,

They're going to have doctors using jetpacks to rush to medical crisis? They're going to have have huge data entry systems where you need massive upper body strength to work all day? They're going to have iris scanning all over the place so you're viagra ads are targetted directly to you?

Oh .. you mean the use of *human* pyschics to predict the future! That sure is a weird definition of technology.

Count carbs (1)

hsthompson69 (1674722) | about 3 years ago | (#35665180)

Super duper easy. The chronic diseases of civilization that cost us the most money (obesity, diabetes, cancer, heart disease, etc), all have their source in the effect of the hormone insulin. Insulin levels are raised by high blood sugar levels, and blood sugar levels are raised by carbohydrate intake.

Now, you probably won't find that data in people's medical records, but if they started tracking that, I think they'd have an excellent predictor of future problems.

Re:Count carbs (1)

Kr1ll1n (579971) | about 3 years ago | (#35665280)

As a type-1 diabetic diagnosed late in life, I can tell you that this is not an end all, and your understanding of this is horribly flawed. The effects of carbohydrates can easily be countered with enough exercise, but even then, that is not a perfect solution. Look at the primary source of carbohydrates in the United States. If you think it is sugar, you would be wrong. It is HFCS, or high fructose corn syrup. Sugar hase a simple chemical structure, and is easy for the body's insulin to breakdown. HFCS on the other hand, is a complex chemical structure, and is more difficult to break down.

Natural Sugar=riding a bike around the block
HFCS=riding a bike around the block, while towing an anchor.

Re:Count carbs (0)

Anonymous Coward | about 3 years ago | (#35665332)

As you well know, the parent is clearly pointing the finger at fat people. So stop being such a pedantic prick.

Re:Count carbs (1)

Kr1ll1n (579971) | about 3 years ago | (#35665374)

And where his finger is pointing matters how? The message is still the same. Oh, and obesity would also decrease if we could get away from HFCS. Stop being an AC and learn to contribute instead of blab out insults. It's a discussion forum, not a divorce case....

Re:Count carbs (0)

Anonymous Coward | about 3 years ago | (#35665338)

Natural Sugar=riding a bike around the block
HFCS=riding a bike around the block, while towing an anchor.

Hmmmm so eating sugar is like light exercise and HFCS is like strenuous exercice... but easier than actually exercising. Okay, I'm sold. How much for your HFCS health supplements?

Re:Count carbs (1)

maxume (22995) | about 3 years ago | (#35665476)

HFCS, is, literally, the components of sucrose. The ratio of glucose to fructose in HFCS isn't 1:1 like it is in sucrose, but it is something like 55:45. Chemically, it is maybe a little simpler, but probably better described as the same.

When humans eat sucrose, it is rapidly converted to free glucose and free fructose, which are individually absorbed by the small intestine. HFCS is already composed of free glucose and free fructose and is absorbed by the small intestine.

So any theories about the evils of high fructose corn syrup have to rely on some sort of signal being disrupted by not requiring the first step of digestion, or perhaps by the slightly higher amount of fructose. There is also the remote possibility that some impurity does something medical.

Re:Count carbs (1)

hsthompson69 (1674722) | about 3 years ago | (#35665710)

Type 1 diabetes is a completely different beastie, but you've made a couple of important mistakes:

1) insulin doesn't "breakdown" sugar. It tells fat cells to hold onto fat, and forces the body to burn the sugar in the blood for energy. It's not some sort of solvent.

2) HFCS is simply a different ratio of fructose and glucose (55/45, instead of 50/50 in common table sugar). The fructose and glucose in HFCS are metabolized exactly the same way the fructose and glucose in common table sugar is.

Re:Count carbs (1)

Dcnjoe60 (682885) | about 3 years ago | (#35665832)

Super duper easy. The chronic diseases of civilization that cost us the most money (obesity, diabetes, cancer, heart disease, etc), all have their source in the effect of the hormone insulin. Insulin levels are raised by high blood sugar levels, and blood sugar levels are raised by carbohydrate intake.

Now, you probably won't find that data in people's medical records, but if they started tracking that, I think they'd have an excellent predictor of future problems.

And your medical degree was from where? While it is true that once one is diabetic, carbs are an issue. Obesity is not so much carb related but the input of calories exceed the expenditure of calories, that could be carbs, more likely fats, and definitely a sedentary life style., Most heart disease has nothing to do with carb intake and actually after a heart attack, most survivors are placed on a high carb low fat diet.

Carbs do raise your blood sugar levels, of course so do protein and fats, unless you are on a low carb diet. But for normal diets, just about everything with calories raises your blood sugar levels as that is because the cells in the body, particularly the brain need glucose to function. Carbs are not bad, you couldn't live without carbs. People in Southeast Asia eat relatively high carb diets, mainly rice, and have low incidences of of any of the diseases you mentioned. There are also a good number of vegetarians that have very high carb diets that are not obese or with any of the ailments you list.

Reduced carbs are good for losing weight, and dealing with glucose levels for those that glucose is a problem, such as diabetics, but it is not the cure all and be all for the general public.

Anyone capable of doing that should know better (0)

Anonymous Coward | about 3 years ago | (#35665184)

...than to deliver it for just 3 million (especially when they are meant to be paid at the discretion of an insurance company).

It's a good idea, but... (1)

Overzeetop (214511) | about 3 years ago | (#35665194)

It's a good idea, but as soon as you include "claims data" in your modeling, it becomes an insurance/actuarial process. Why would claims data matter? Besides, isn't this what primary care doctors are currently tasked with? Maybe they'd prefer to work both sized of that insurance equation - raise premiums and reduce personnel costs?

Re:It's a good idea, but... (1)

Motard (1553251) | about 3 years ago | (#35665562)

They're using claims data because that's the data they have. They don't have the full medical records. They're a provider network. They don't set premiums.

Spending is negative, preventing is invisible (1)

germ!nation (764234) | about 3 years ago | (#35665202)

Isn't the wider problem that no one has "Money that didn't have to be spent" on their balance sheets? If people regularly claim on their health insurance (I assume that's how it works? UK resident here) won't their cover suffer in some fashion down the line even if the times they picked to claim where 100% right decisions that removed the need for much more expensive future claims?

Re:Spending is negative, preventing is invisible (1)

maxume (22995) | about 3 years ago | (#35665252)

Despite all the yelling, "employer" is a much better predictor of the level of insurance a person in the U.S. has than "medical history".

That medical history has an impact can be much more emotionally offensive so it isn't that shocking.

Sounds good to me, in my dreams (3, Funny)

symes (835608) | about 3 years ago | (#35665212)

So there I was, walking down the street minding my own business... when a van screeches to halt in front of me. Five (5!) scantily clad nurses throw me to the ground and give me the kiss of life. Who knew I was about to be run over?

Re:Sounds good to me, in my dreams (1)

symes (835608) | about 3 years ago | (#35665288)

Actually, I'm confused... The algorithm is to predict costly hospitalisations? Surely there are some diseases where hospitalisation is the best way of reducing premature death (heart bypass?) In this case, would the algorithm be expected to leave these guys alone as they are more likely to suffer death than hospitalisation? And... surely the BEST way of improving health (if this is what they are really interested in) is to get people when they are young and steering them away from unhealthy lifestyles (alcohol, tobacco, drugs, etc.)? Furthermore, surely age is the biggest predictor of hospitalisation, not neccessarily because they are more likely to contract disease but because they are more vulnerable. Would the model be expected to discount age? There sure seems to be a striking lack of clarity here, that or I should read TFA...

Re:Sounds good to me, in my dreams (1)

vlm (69642) | about 3 years ago | (#35665390)

get people when they are young and steering them away from unhealthy lifestyles (alcohol, tobacco, drugs, etc.)?

Can't fix stupid. However, if they've already pushed the self destruct button, its possible that they could be steered toward a cheaper, faster demise rather than a slow expensive trainwreck. I can see the ads now "For patients whom drink more than 12 beers every day, and only patient whom drink more than 12 beers every day, the Surgeon General recommends that methanol and OJ is a better drink than ethanol and OJ; bottoms up!"

Remember they're not in the business of improving health, they're in the business of improving profit.

Re:Sounds good to me, in my dreams (1)

xtracto (837672) | about 3 years ago | (#35665684)

5 != 5!

Why don't they submit their prediction competition to Kaggle [kaggle.com]. There are quite a lot of "impossible" prediction competitions over there already.

Computerized discrimination? Donot want! (0)

Anonymous Coward | about 3 years ago | (#35665290)

Donot want... mmm.... donut want. Time to head down to Krispy Kreme's.

I'd rather write a heuristic!! (1)

Quato (132194) | about 3 years ago | (#35665346)

Aren't algorithms always supossed to reach a termination result? I don't want ot be terminated!

As a doctor (1)

Dunbal (464142) | about 3 years ago | (#35665378)

I find this disturbing. So the "Heritage Provider Network" is willing to submit healthy people to weekly or even daily blood tests in order to feed these algorithms? When you add up the total cost of the "prediction", it ends up being greater than the total cost of treating the condition. Or perhaps they think that they can use magic to predict who is going to get sick when. I mean, statistics can give you a general idea, but they absolutely cannot say what will happen in your particular case. The secret to healthy living is the one people have been hearing all along. Don't smoke. Don't drink to excess. Look both ways before crossing the street. Eat more vegetables and fruit than carbs and meat, eat small portions 5 times a day, do at least an hour's worth of exercise per day, keep your weight down, and see your family physician regularly for primary (preventative) care. Unless you have some genetic predisposition to disease that you can't do anything about, the above sentence will carry you well into your 9th decade.

Re:As a doctor (1)

bws111 (1216812) | about 3 years ago | (#35665822)

Where does it say anything about daily (or otherwise) blood tests? It specifically says they want to analyze available information, like claims data. So if they get a claim for diagnostic code x for someone, then months later get a claim for code y, then code z, they want to be able to analyze that and recommend some preventative measures so the person does not wind up hospitalized. Yes, this is the sort of thing a doctor would normally do, but many people go to different doctors for different conditions, and don't always keep their primary care physician up-to-date.

Heathcare in US = failure of capitalism (0)

Anonymous Coward | about 3 years ago | (#35665380)

This is just an insurance company doing what it is chartered to do as a corporation: maximize shareholder value. No matter that, in practice, maximizing shareholder value means that they are no longer providing healthcare insurance at all. Providing real insurance is fraught with risk and downside - better to provide prepaid healthcare services to people who won't need all the money they pay in = profits maximized.

The benefit of capitalism is better products and services through competitive pressure. Where's the competition for insurance companies that drives them to serve the customer better?

Maybe there are just certain industries and services where capitalism is a net negative?

predicting those at risk doesnt help much (2)

meander (178059) | about 3 years ago | (#35665492)

The commonest disabilities in the western world are heart disease, stroke, diabetes, and multiple arthritis from being too heavy for your poor bones to handle.

I'm a GP doctor; as folk walk in the door, it is usually obvious who is going to be at risk for future problems. They are fat, overflowing my poor abused seats, they groan as they stand up, they are obviously unfit. As a added bonus, I can often smell the cigarettes on their breath.

Does knowing who is at risk help? Sadly, all too often it doesnt.

Lazy fat slobs will on average die considerably younger of way too many diseases, and I have not even mentioned chronic crappy disabilities like back pain, hip pain, knee pain. I pride myself that I turn a few of these folks to the bright side of eating a bit better, exercising a tad more, and thus living longer and actually enjoying those healthy extra years.

You don't need an algorithm to work out who is at risk of future disease, it is bloody obvious (can I have my $3M now?). The problem is getting these fat, unfit folk to realise there is more to exercise than driving to buy their next greasy pizza.

Oh! they deliver now as well...

Re:predicting those at risk doesnt help much (1)

Dcnjoe60 (682885) | about 3 years ago | (#35665664)

The commonest disabilities in the western world are heart disease, stroke, diabetes, and multiple arthritis from being too heavy for your poor bones to handle.

I'm a GP doctor; as folk walk in the door, it is usually obvious who is going to be at risk for future problems. They are fat, overflowing my poor abused seats, they groan as they stand up, they are obviously unfit. As a added bonus, I can often smell the cigarettes on their breath.

Does knowing who is at risk help? Sadly, all too often it doesnt.

Lazy fat slobs will on average die considerably younger of way too many diseases, and I have not even mentioned chronic crappy disabilities like back pain, hip pain, knee pain. I pride myself that I turn a few of these folks to the bright side of eating a bit better, exercising a tad more, and thus living longer and actually enjoying those healthy extra years.

You don't need an algorithm to work out who is at risk of future disease, it is bloody obvious (can I have my $3M now?). The problem is getting these fat, unfit folk to realise there is more to exercise than driving to buy their next greasy pizza.

Oh! they deliver now as well...

If you truly are a physician, It's hard to understand why you would choose to stay in practice with the arrogant attitude you have towards your patients. I am so glad that in your humanitarian care, you have chosen to turn a few of these burdensome patients of yours to the bright side. If what you posted it really how you feel, then it is truly sad for the patients you treat. They deserve better than that, whether sick or not, whether overweight or not, whether you like their choices or not.

Counter-Countering (0)

Anonymous Coward | about 3 years ago | (#35665528)

OTOH - Hackers and Cracker networks will probably find funding for attempts at code that foresees the foreseeing programs. Effectively for-foreseeing them (44seeing, or 8seeing, or 16seeing, foursquare-seeing) algorithms. Combined with the usual phishing and worming around, probably. A smarter hacker could put it all in some VIPs name, in multiple contracts, paid by one of those Iraq accounts that make 10s of bill...ions "dissapear", unaccountably, unredeemably, unprosecutedly. MAkes sense, in a rocambolesque, art-nouveau, steampunkish, sort of way.

The "healthcare providers" are our true heroes! (1)

Evi1M4chine (2029370) | about 3 years ago | (#35665550)

Because opposing to what all you nanny-state friends out there claim, they care very much for your health.
In that they make extremely sure you are never ever sick... in any way that they have pay you a single cent.

</Colbert>

Noble efforts or Evil plan? (1)

Dcnjoe60 (682885) | about 3 years ago | (#35665588)

While it may be a noble effort to try and predict when people will be sick and need major medical care or even use the information to provide preventative measures before the events occur, there is also much potential for the information to be misused. The same information to predict serious illness can also be used to deny coverage to those individuals that have negative predictions

70 years ago, splitting the atom led to the expectation of cheap, safe, unlimited energy. However, as the world found out, that same effort also lead to bombs of unbelievable destructive capability. Splitting the atom was a neutral thing. How it is used became a moral thing. Likewise, predicting illness is a neutral thing, but how it could be used will very much turn into a moral thing.

LOL (1)

Charliemopps (1157495) | about 3 years ago | (#35665726)

as a person that has had multiple undiagnosed illnesses over the years that I had to finally diagnose myself and then demand tests that my healthcare provider didn't want to administer, only to find out I indeed had the diseases I was concerned about (graves disease and pernicious anemia) I can state for a fact that there is a very simple way to preemptively diagnose disease.

In both cases I had symptoms that CLEARLY indicated the diseases I had. In both cases, standard rudimentary tests that cost just a few dollars would have been enough to diagnose the disease. My doctor had a screen in which he could order blood tests. It was a single screen with check boxes. All of the tests required to diagnose me were on those screen... but he'd only check 1 at a time. Then require me to come back and he'd run another. After months even years of getting nowhere I found out that I could request my test results. I found out that the doctors I had seen were running the same, useless tests over and over. Usually for iron deficiency or blood sugar level. They never checked my TSH level, they never checked my basic blood count. In the end I looked up the tests I needed and demanded them much to my Doctors dismay.

Want to diagnose people early? Run ALL the tests. TSH level should be a standard test run once a year. Thyroid problems affect 30% of the population! Most go undiagnosed. Pernicious anemia is also common, usually appears at age 30 and usually goes undiagnosed until it becomes so severe that it induces dementia at age 60!!! Both of the tests for these disease also cover dozens if not hundreds of other common ailments. How much money was my health care provider saving by skipping these tests? How much would it eventually have cost them for not treating it and then having to put me in a nursing home at age 60 because of the brain damaged suffered because of the anemia?

Tests are cheap, and the more you run then, the cheaper they will get. There's absolutely no reason not to do a full blood panel and test for a wide range of disease once a year at their physical.

A good thing (0)

Anonymous Coward | about 3 years ago | (#35665902)

If this works, it will be a very good thing. Getting it to work is the tricky bit.

I have often advocated (in my country, where we have a sensible, national, healthcare system) the government making available an (optional) annual check up to all citizens. Full CAT scan, full blood tests, every non-invasive test under the sun, once per year, to anyone who wants it. The costs would be massive, but so would the benefits. Think of all the illnesses (particularly tumours) that could be caught early, before they become (a) serious and (b) expensive to treat. Such a plan might even pay for itself in the long run.

How about .... (0)

Anonymous Coward | about 3 years ago | (#35666008)

Probability of illness = ( doughnut intake * daily cigarette count ) / Insurance premium
Load More Comments
Slashdot Account

Need an Account?

Forgot your password?

Don't worry, we never post anything without your permission.

Submission Text Formatting Tips

We support a small subset of HTML, namely these tags:

  • b
  • i
  • p
  • br
  • a
  • ol
  • ul
  • li
  • dl
  • dt
  • dd
  • em
  • strong
  • tt
  • blockquote
  • div
  • quote
  • ecode

"ecode" can be used for code snippets, for example:

<ecode>    while(1) { do_something(); } </ecode>
Sign up for Slashdot Newsletters
Create a Slashdot Account

Loading...