Jonathan Halvorson

Health care policy and scientific method. Interests of late: causal inference, AI, markets vs spending caps for cost control.

[Jonathan Halvorson @son_of_halvor]

@mcuban The carriers responded to intense public pressure. They were demonized, and have remained so ever since.
Those forces will demonize you too, if you ever get close to implementing a plan that strangles the golden goose.
The barrier is not the carriers, it is public sentiment.

[Jonathan Halvorson @son_of_halvor]

https://t.co/eiALfGMfwW

[Jonathan Halvorson @son_of_halvor]

I've written a short essay on how our talk about cause and effect is often a disguised way of assigning moral responsibility. Not so much to trick our opponents, but because we have no other choice. 1/2

[Jonathan Halvorson @son_of_halvor]

We spend a lot of time trying to understand how disagreements about what is good arise and persist.

I find a related point more fascinating: differences in beliefs about what is good also drive differences in beliefs about facts. Morality doesn’t just get tied up with an outlook on what is good, but an outlook on what is.

[Real Doc Speaks @realdocspeaks]

Here is the interesting part of plain films. When I started out, they were developed on film, and they had to be filed.

The hospital had to pay for the film, the developing, file room clerks, file folders, and surgeries were canceled because no one could find the films.

Now it is all digital, and the image appears immediately. There are no costs for:
• Developing the film
• The physical film
• Storage of the film
• Salaries for file room clerks

Where are the savings? If this were any other industry, the price of a plain film would have followed that of computers.

We need to develop a free market to obtain reasonable prices.

[Jonathan Halvorson @son_of_halvor]

The average physician's income from Medicare has gone up since the 90s, not down. And commercial income has gone way up. Overall physician income has increased faster than the general rate of inflation.

The problem is not that physicians aren't paid enough. Physicians are quitting because FFS volume-driven billing always (in every nation) requires controls to avoid fraud, waste and abuse. Entrepreneurial physician billing practices have resulted in many complex, tedious, and sometimes abusive insurance controls that doctors hate, and the more costs go up the more insurers try to clamp down.

And as insiders know, many specialties pay far more than primary care. That's where the really ambitious doctors go, reducing primary care supply. Primary care docs make invidious comparisons and feel left out, cheated compared to cardiologists and dermatologists. Fix that imbalance, and move away from FFS.

[Ed Livingston @ehlJAMA]

It is absolutely wrong to refer to the AMA as America’s physicians. There are hardly any active practicing physicians who belong to the AMA. They say they have 350,00 members because they consider all 300,000+ students and residents as members of the AMA.

I can tell you from having a worked there that they don’t even tell their own employees how many actual members they have it’s embarrassingly small. I also know because I used to manage the CME portfolio that they provide very few CME credits for their members – another indicator of how small their membership is. When I was there they would issue about 20 to 30,000 CME credits per year which almost all came from JAMA. Less than 10% of their CME came from AMA activities. This is in contrast hundreds of thousands or millions of credits given by other by other organizations that actually do have very large memberships

When I first went to work there I was shown a poll that showed 70% of physicians have a unfavorable view of the AMA.

2/3 of their income comes from their ownership of CPT and HHS mandates that CPT be used for billing. Without CPT there would be no AMA.

I had an awful experience working there as some of you might know.

The physician community should absolutely not let the AMA present themselves as representing American medicine

[Jonathan Halvorson @son_of_halvor]

The claim "HIPAA was a mistake" is making the rounds. What people forget is that it was created in another world where guaranteed issue and community rating in the individual market didn't exist (except for New York, which was in a death spiral). A major goal of the privacy rule was to make sure people weren't denied insurance or charged far more due to knowledge of pre-existing conditions.

Those days are gone, and critics are right that HIPAA gets in the way of scientific progress.

Time to seriously update and re-weigh the value of scientific progress against the value of privacy in the current environment.

[Chris Anderson (in London) @chr1sa]

"HIPAA was a mistake". Yes, and this is true for almost all of the "precautionary principle" regulations, from GMO restrictions in Europe to AI regulations across US states.

Rather than imagining all the things that might go wrong and seeking to prevent them, let the marketplace work and if and when things actually do go wrong, regulate *then* based on real-world harms not fantasies.

[Jonathan Halvorson @son_of_halvor]

@TSRooseveltRE @DRsLoungePod @anish_koka Isn't that your problem, not mine? Your view that the probability of every event is 0 or 1 is determinism. To make sense of action, you need a perspective from which aspects of the future are undetermined until we act.

[Jonathan Halvorson @son_of_halvor]

@TSRooseveltRE @DRsLoungePod @anish_koka That's just metaphysical determinism. We still have to make predictions about individuals and estimate likelihoods. The criticism in the case above is that not enough of the right variables were used to make the prediction.

[Jonathan Halvorson @son_of_halvor]

@BadEconomicTake @ScottHickle Neither of us is retarded or insane. Try being more useful in your comments.

[Jonathan Halvorson @son_of_halvor]

@AeonCoin @ScottHickle Seems like it in the end.

[Jonathan Halvorson @son_of_halvor]

@AeonCoin @ScottHickle With a 90% medical loss ratio, no insurer will have a profit. All fully insured business lines have at least 10% admin cost. There are some TPAs and ASO business lines that have an admin cost ratio slightly below 10% but the MLR rules don't apply to those.

[Jonathan Halvorson @son_of_halvor]

No, it is 3-5% for health insurance. It has been for decades. This is easy to verify.

You're completely misunderstanding this "20% limit." It applies to administrative expense and profit, and it is created by the 80% floor for medical expenses (actually 85% in a lot of cases). There is no insurer on Earth that has 0% administrative expenses, so 20% is the profit limit only in theory, but in reality the limit is about 10%, and the average as I've said is 3-5% because admin costs are often close to 15% and insurers often don't hit the lower bound on medical expenses but spend a little more than that.

[Jonathan Halvorson @son_of_halvor]

Insurance profit of 50% just means you're paying nearly twice as much premium as you would in an efficient market for insurance. Yes, if I'm overpaying for premium and on the hook for almost all my medical costs, I'll use fewer medical services.

Why would an insurer with a super high deductible get a discount from the provider on price?

Your example doesn't make sense to me. If the doctor gets the cash price of $128, then the insurer hasn't paid anything. Its profit has nothing to do with the cash price it didn't pay.

Also, health insurer profit margins are typically 3-5%, nowhere near 20%. An MLR of 80% gets added to a 15% AER (administrative expense ratio) to yield a 5% profit.

[Jonathan Halvorson @son_of_halvor]

Exactly, I can list out a dozen reasons the US has the highest prices in the world. The ACA's MLR rule would not be in the top 10.

People forget: the MLR was set at a level that 90% of plans already met. Pretty much all employer sponsored insurance met it. There were some individual market plans (mini-meds and very high deductible plans) that did not. People don't understand insurance, and so didn't know what they were buying, and thousands went bankrupt with these plans that were often unscrupulously marketed.

[Jonathan Halvorson @son_of_halvor]

A counterpoint to data center hysteria:

[Jonathan Halvorson @son_of_halvor]

You missed the point. I said that enrollment fraud that adds people who don't use services drives up the cost to government. What it doesn't do is drive up the premium rate per enrollee.

For example, take a scenario where 50% of enrollees don't know they're enrolled, pay zero premium themselves, and aren't using any services. The premium per enrollee will go down because the average enrollee is using less services. Let's say premium goes down by 20%. The total premium spending by the government in that case goes up by 20%, since 0.8 x 1.5 = 1.2.