Sign up for email updates!

Don't miss out on what matters. Sign up for email updates!

Stay informed! Sign up for e-mail updates:

Monday, October 28, 2019

Racist Algorithm: Your Medicare-for-All Future

A computer algorithm prioritizes hospital patients based on their economic value. Some people call it racist; in reality, it is a socialist invention.


Remember Microsoft's foray into artificial-intelligence tweeting? It ended in a picture-perfect disaster after the AI tweet machine had endorsed Adolf Hitler and made other comments that would get a human tweeter banished to the darkest corners of the virtual universe.

Using artificial intelligence and algorithms to make ethical decisions is not even remotely as simple as the 25-year-old programmers at Microsoft thought it would be. After having wasted a whole lot of company money, and made a fool out of artificial intelligence, you would think that the people behind the "Tanya Tay" tweetbot had set a bad enough example for the rest of this century.

Not so. Algorithm lovers have struck again. Behold the Wall Street Journal, Friday October 25, p. A3, print edition:
Black patients were less likely than white patients to get extra medical help, despite being sicker, when an algorithm used by a large hospital chose who go the additional attention, according to a new study underscoring the risks as technology gains a foothold in medicine. ... the algorithm gave healthier white patients the same ranking as black patients who had one more chronic illness as well as poorer laboratory results and vital signs.
In plain English, this means that the algorithm allocated as much health care resources to the healthier patients as it did to those who had more serious medical conditions. This, of course, is stupid from a common-sense viewpoint, but it is perfectly rational if you are an advocate of single-payer health care. We learn why as the WSJ article continues:
The algorithm used cost to rank patients, and researchers found health-care spending for black patients was less than for white patients with similar medical conditions. 
There are two parts to this. First, the algorithm elevated the healthier patients because it equated cost with value. When you don't use medical professionals to determine where medical resources should be spent, you inevitably end up with an algorithm that places a monetary value on a patient's treatment. 

Secondly, even though the article does not explain this point, the difference in cost has to come from somewhere else than a blatantly racist preference. It is likely that the white patients came with better insurance than the black patients; if black household income is lower than white household income, it stands to reason that a larger share of black patients than white patients would come to our hospitals with Medicaid rather than private insurance. 

If this is the case, then given the lower reimbursement rates in Medicaid compared to private insurance the black patients would be ranked lower by this algorithm. 

This little story should be a big wake-up call for all those who contemplate a single-payer health care system, i.e., "Medicaid for All". In such a system, the finite budget is set not by what patients bring to the table in the form of private insurance, but by the bottom line in a health-care budget passed by Congress. This bottom line is then paired with guidelines for how the finite resources be distributed among patients. As I explained in my book The Rise of Big Government, these guidelines - known in Europe under the acronym QALY - put a distinct economic value on the treatment of the patient. Given the patient's expected recovery, the allocation of treatment resources depends on the economic value of that recovery.

The economic value, in turn, is often tied to a patient's ability to pay taxes in the future. If the patient can be expected to deliver at least the amount in taxes that his or her treatment cost, then the patient is definitely worth treating; if not, the loss to government has to be weighed against the alternative use of the treatment resources. If there is another patient with what government considers to be a better cost-benefit balance, then it will steer the resources in that patient's direction.

We do not have a single-payer system in America (thank God) but the algorithm reported on in the Wall Street Journal article is a microcosmic example of what a QALY-style algorithm would do to the entire health care system in America - and all its patients. The first order of business would be to place an economic value on each patient. The second order of business would be to assess the cost of the treatment the patient needs. Third, the algorithm would balance the economic value against the cost of treatment and then rank patients based on where the least amount of resources can be expected to deliver the most amount of benefits - to government. 

The true lesson, of course, from this is that we need to reinforce the private sector in our health care system. As I recently reported, the Republican party's Study Committee has presented a good, workable reform plan to roll back the most excessive stupidities of Obamacare. Check out Part 1 and Part 2 of my analysis of the reform plan. 

No comments:

Post a Comment