When a 'Wildly Irrational' Algorithm Makes Crucial Healthcare Decisions (theguardian.com) 38
"Thousands of disabled and elderly people in more than a dozen states have had to fight against decisions made by an algorithm to get the support services they need to remain in their homes instead of being institutionalized," reports the U.S. edition of the Guardian:
The cuts have hit low-income seniors and people with disabilities in Pennsylvania, Iowa, New York, Maryland, New Jersey, Arkansas and other states, after algorithms became the arbiters of how their home health care was allocated — replacing judgments that used to be primarily made by nurses and social workers.
In Washington D.C., "on the worst end, we've had clients who actually died, because their services were cut and they were not receiving the care that they needed" Tina Smith Nelson, supervising attorney with AARP Legal Counsel for the Elderly, said about the effects of a new algorithmic system introduced in 2018. Over 300 seniors have had to file administrative appeals after their home care was cut by a new algorithmic system. "I think as a society we move into unsettling territory when we rely solely upon algorithms and data to make determinations about health care needs," Nelson said. "We reduce a person's humanity to a number...."
The situation is reflective of a reality increasingly affecting all users of American healthcare: algorithms — ranging from crude if-then charts to sophisticated artificial intelligence systems — are being deployed to make all sorts of decisions about who gets care. Government officials have touted algorithmic decision-making systems as a way to make sure that benefits are allocated even-handedly, eliminate human bias and root out fraud. But advocates say having computer programs decide how much help vulnerable people can get is often arbitrary — and in some cases downright cruel. The underlying problem, experts say, is that neither states nor the federal government provide enough funding to allow people needing health assistance to remain safely in their homes — even though these programs usually end up being much less costly than putting people in institutions. The algorithms resort to divvying up what crumbs are available...
Kevin De Liban, an attorney with Legal Aid of Arkansas, began fighting the cuts after severely disabled patients started calling "en masse" in 2016.... De Liban's legal team revealed flaws with the algorithm in court. It turned out, De Liban said, that the calculations had failed to factor in things like whether a patient had cerebral palsy or diabetes. A single point in the scoring system — for instance a point added because the patient had had a fever in the last three days or had open pressure sores — could make a huge difference in how many hours they received for the entire year... "As the algorithm worked, it was, to our eyes, pretty wildly irrational," said De Liban...
After years of court battles, Arkansas' use of the algorithmic system was finally thrown out in 2018... But across the nation, the battle continues. In Washington D.C., Pennsylvania and Iowa, legal services attorneys are plagued with calls from seniors complaining they have lost their care because of the algorithms recently adopted in those states.
The Guardian ultimately tracked down the designer of the algorithm, University of Michigan Professor Emeritus Brant Fries, who acknowledged that the system isn't even designed to calculate how many hours of care people actually need, but to try to allocate whatever scarce resources are available in the most equitable way.
"We're not saying that the size of the pie is correct... But whatever the money is there, I'm dividing it more equally!"
In Washington D.C., "on the worst end, we've had clients who actually died, because their services were cut and they were not receiving the care that they needed" Tina Smith Nelson, supervising attorney with AARP Legal Counsel for the Elderly, said about the effects of a new algorithmic system introduced in 2018. Over 300 seniors have had to file administrative appeals after their home care was cut by a new algorithmic system. "I think as a society we move into unsettling territory when we rely solely upon algorithms and data to make determinations about health care needs," Nelson said. "We reduce a person's humanity to a number...."
The situation is reflective of a reality increasingly affecting all users of American healthcare: algorithms — ranging from crude if-then charts to sophisticated artificial intelligence systems — are being deployed to make all sorts of decisions about who gets care. Government officials have touted algorithmic decision-making systems as a way to make sure that benefits are allocated even-handedly, eliminate human bias and root out fraud. But advocates say having computer programs decide how much help vulnerable people can get is often arbitrary — and in some cases downright cruel. The underlying problem, experts say, is that neither states nor the federal government provide enough funding to allow people needing health assistance to remain safely in their homes — even though these programs usually end up being much less costly than putting people in institutions. The algorithms resort to divvying up what crumbs are available...
Kevin De Liban, an attorney with Legal Aid of Arkansas, began fighting the cuts after severely disabled patients started calling "en masse" in 2016.... De Liban's legal team revealed flaws with the algorithm in court. It turned out, De Liban said, that the calculations had failed to factor in things like whether a patient had cerebral palsy or diabetes. A single point in the scoring system — for instance a point added because the patient had had a fever in the last three days or had open pressure sores — could make a huge difference in how many hours they received for the entire year... "As the algorithm worked, it was, to our eyes, pretty wildly irrational," said De Liban...
After years of court battles, Arkansas' use of the algorithmic system was finally thrown out in 2018... But across the nation, the battle continues. In Washington D.C., Pennsylvania and Iowa, legal services attorneys are plagued with calls from seniors complaining they have lost their care because of the algorithms recently adopted in those states.
The Guardian ultimately tracked down the designer of the algorithm, University of Michigan Professor Emeritus Brant Fries, who acknowledged that the system isn't even designed to calculate how many hours of care people actually need, but to try to allocate whatever scarce resources are available in the most equitable way.
"We're not saying that the size of the pie is correct... But whatever the money is there, I'm dividing it more equally!"
Re: (Score:2)
The machine may be rational, but the algorithm is not. Literally, it is not "based on or in accordance with reason or logic".
Re: (Score:2)
tensorflow.
this likes a solution that be applied here.
for use by corperations.
but for use by people
How do you own common sense...? (Score:1)
Re: (Score:2)
In the USA it would be...
The Governments!
The Insurance Companies!
NOT for the patients or the medical personnel.
The Governments and the Insurance Companies have been making important decisions for patients and overriding the patients' wishes and needs as well as the Medical Doctors and medical staff. IMO, it is horrendous when Governments and Insurance want to voilate the relationship between patient-doctor and let non-medical personnel decides the medi
Re: How do you own common sense...? (Score:1)
size of the pie? What "pie"? (Score:1)
Damn people aren't even leaving the tin
The pie is a deception
Correct Pie size (Score:3)
Pi is supposed to be 3.14159265358..
Instead they set the pi to 0.03
Re:Maybe we should get government out of healthcar (Score:5, Interesting)
These are presumably Medicare patients; if they didn't have Medicare they likely wouldn't have health insurance at all because at their age, they're uninsurable.
Except for Medicare and Medicaid the government stayed out of health care for many decades, and through those decades US spending per capita has outstripped comparable advanced countries while outcomes fell behind. We are ranked #37 in the world by health care quality, but are #1 by per capita spending. The #1 country in the world by outcomes is France, which has universal government health insurance and spend half what we do per capita. Even *Italy* is kicking our ass. They spend almost exactly 1/3 what we do per capita and are #2 in the world.
Re: (Score:3)
And, of course, the VA, to care for those of us who served and didn't come back whole.
Re:Maybe we should get government out of healthcar (Score:5, Interesting)
Much of that is due to monkey-wrenching by people who are determined to prove that government is an inferior solution even if they have to make it true themselves.
Re: (Score:2)
This isn't an accident. Nobody in red states is willing to pay to keep someone else's grandparents alive and free, because nobody in red states is willing to pay to keep -anyone- alive and free.
Re: (Score:2)
(Also, anyone else find it hilarious that the idea of using data to make decisions is being painted as a bad thing? How else do you want to make decisions? Random guessing?)
Not all uses of data are equal. For example, would you applaud a new data driven tax law if the computation was left shoe size in mm * the square root of the number of letters in your middle name + the current lunar day / the number of times "The Trouble with Tribbles" has aired that month +2? It is data driven, after all.
Re: (Score:2)
Re:Maybe we should get government out of healthcar (Score:5, Informative)
Re: (Score:3)
Determining healthcare choices doesn't just require authority to do so, it requires resources. 'People' and charities can stand by and complain when everybody doesn't get every treatment. They can even intervene and assist in a handful of rather arbitrarily-selected incidents. But what they cannot do is step in and give everybody every treatment. And if they tried, they would immediately run into the same problems faced by
Re: (Score:3)
Re: (Score:2)
Re: Maybe we should get government out of healthca (Score:5, Insightful)
Re: (Score:2)
The idea of data being used isn't being painted as a bad thing. The thoughtless use of too little data (eg not including data on diabetes) and overweighting of certain data types (eg data on hopefully transient phenomena such as pressure sores) is what's being critiqued
Re: (Score:2)
Hear, hear!! You have room for Grandma and Grandpa at your place, right? Since there won't be any Medicare in your world, people like you should be stockpiling their prescriptions now. Put your money where your mouth has been.
The crucial part (Score:5, Insightful)
The crucial part is right there in the text:
So, it seems the problem is we shrank the government a bit too much. And no, letting families and communities figure it out on their own is not a solution. Unless you're an an-cap who wants society to return to a neo-feudal Charles Dickens times where everyone was flat broke except for a select few. That experiment in laissez faire capitalism didn't end too well either.
You tell me how other countries much smaller than the US, can get better results for half the cost per capita? It's because the evicted the profiteering gluttons from their systems.
Re: (Score:2)
You tell me how other countries much smaller than the US, can get better results for half the cost per capita?
Well, assuming your assumption is correct, that they CAN get better results (which is open to debate), the answer is BECAUSE they are MUCH smaller and more homogeneous.
Re: (Score:1)
Well, assuming your assumption is correct, that they CAN get better results (which is open to debate), the answer is BECAUSE they are MUCH smaller
How is size relevant? Medicare and Medicaid are administered by States, so it scales by size.
and more homogeneous.
And how is this relevant either? How is the colour of skin a big factor in organising healthcare? If you looked at, for example, percentage foreign born, then the USA is actually slightly less diverse than countries such as the UK.
They are just excuses for inaction.
Re: (Score:2)
HOLD UP! (Score:4, Insightful)
What about when we have "wildly irrational" people making laws? ;)
Re:HOLD UP! (Score:5, Interesting)
Re: (Score:3)
Laws are a much LOOSER algorithm than computer programs. They tend to depend on the 'legal' definition of generic words that get specified by judges in response to actual real life situations. More importantly, they have a systematic correction method from appealing to a higher court.
Computer algorithms tend to start out as specific as possible, either defined by bureaucrats based on money rather than situations, or worse, by Bayesian based learning programs fed a small, prejudiced sample.
Stop blaming algorithms (Score:5, Insightful)
Testing (Score:3)
Computer says you die.. (Score:1)
Re:Computer says you die.. (Score:4, Insightful)
And yet it's supposed to be in socialism that there are 'death panels' deciding whether you live or die.
Hey America, isn't it time to embrace western democracy instead?
Equally vs equitably (Score:2)
"But whatever the money is there, I'm dividing it more equally!"
It doesn't take a computer to divide it equally. You just divide by n to get each share.
He means "equitably." Unfortunately, equity is in the eye of the beholder. Since he doesn't seem to understand that, he is an idiot.
The pie isn't big enough (Score:2)
Do not blame the algorithm (Score:2)
So, the critical sentence is at the end of the summary:
"But whatever the money is there, I'm dividing it more equally!"
Government can do rationing two ways: the usual, and the equal. One has hidden, and the other has outrageous harm.
Usually, government puts people on never ending waiting lists. You want free housing voucher? Wait 28 months. Want a green card for your mom? Wait 10 years. Want institutional care? Again wait many years. But once you get it, it is essentially yours forever.
The other case: Every