My Iowa post last week got a whole lot more attention than I thought it would have and provoked several great conversations. One of the questions was about how to deal with this extreme corner case and if a high cost risk pool made sense for this use case. This is the perfect example for a high cost pool.
And then there was a great discussion on how to design such a program. And this leads to a discussion about clustering. Bigger risk pools are more efficient and effective at spreading risks, so we’re going to talk about clustering for a bit.
@onceuponA @matthewherper @afrakt @aaronecarroll @larry_levitt @ASlavitt @LorenAdler Ideally national pooling not 51 state pools as random clustering would be ugly for small state much less non-random cluster of doom
— David Anderson (@bjdickmayhew) April 21, 2017
Let’s imagine a hypothetical high cost risk pool of the top 1,000 individual claim years in the country. The average claim will be $5,000,000 for the year. Let’s simplify things and say 990 are randomly distributed by population and 10 are a non-random cluster that we can insert into any state at any time. The first run through is with fifty one state (and DC) based high cost risk pools. We’ll look at two states, California and Wyoming, for this run.
California has about 10% of the population. California should expect to see 99 people in this hypothetical pool plus an expectation that one of the ten non-random people would be expected to be in California. Their expected high cost risk pool budget is $500 million. Now if all ten of the non-randomly clustered people are in California, they increase the expected pool costs by 9%. California is big enough and rich enough that a surprise $45 million dollar medical expense does not destroy their budget.
Now Wyoming should expect to see between 1 and 2 people qualify for the high cost risk pool. Let’s assume the Wyoming state government is very cautious and they allocate $10 million for the high cost risk pool. That works great in a normal year. But if the travelling roadshow of catastrophic medical expenses arrive in Cheyenne, the state is now on the hook for twelve qualified individuals. They are 500% over budget now and the state budget is underwater.
This thought experiment is amazingly unrealistic.
Even if we are to assume that extreme medical cost cases are randomly distributed, we should expect several states to be surprised at the number and expense that they face as Pennylsvania could reasonably expect to see anywhere from 45 to 51 qualifying individuals from the scenario above in any given year just do to random chance.
More importantly, we know that diseases are not randomly distributed. My ongoing freak-out about Zika is based on the fact that this is a concentration of very high need and high cost individuals on states with low Medicaid funding. Genetic disorders are tightly clustered due to both the combination of most people live near their families rather than being randomly distributed and localized clusters of diseases have led to local medical-industrial clusters of medical knowledge and treatment. For instance, maple syrup urine disease is a common genetic disorder among Amish families, so there is a good deal of knowledge on treating that disease clustered in Lancaster County, Pennyslvania and Holmes County, Ohio. Sickle cell disorders are overwhelmingly a disease of African Americans, so it is more common in Mississippi than Montana.
From a financial perspective, there is a chance that there is enough sample size that although one state will have more of one genetic disorder it washes out as another disorder it is light in is dis-proportionally prevalent in another state so the cash flows balance out. That is an empirical question that I don’t know enough to answer. But even if genetic disorders balance out, localized outbreaks like Zika won’t balance out.
State based high cost risk pools would remove some of the falling knife incentives that I described in Iowa but they will be underfunded and overwhelmed at times of high need. National level pooling is far more efficient and effective.
Wag
Makes sense to anyone who isn’t a republican.
ArchTeryx
A national high risk pool doesn’t change the fundamental nature of them, though: The ghettoization of the sick specifically to encourage them to die quickly by grossly under-funding the pool. They have never worked at any scale.
Someone in the Iowa thread commented that the cynical solution would be to bribe a doctor to make a “mistake” and slip enough sedative into the patient’s IV to cause them to die.
What he didn’t say is that high risk pools have exactly the same goal – just making sure it is done out of sight, and done by withholding care or making it unaffordable. In other words, the status quo ante prior to the ACA. AKA, murder by spreadsheet.
It’s what the Republicans want to take us back to, too, and I still feel I dodged a bullet aimed straight at my head when ACA/Medicaid repeal fell short in Congress.
[EtA: Reinsurance is actually a form of high-risk pooling done right: The patients get the same access to benefits at the same cost as healthy people, but a larger entity picks up paying for the edge cases ‘behind the scenes’, to assure they keep getting the care they need. Reinsurance has the OPPOSITE goal of high risk pools, though: To keep the truly sick in the main pool without bankrupting their insurers.]
JPL
Once again a very informative post, although I had to google maple syrup urine disease.
dr. bloor
@ArchTeryx: David pointed out the problems of reinsurance with chronically expensive people in that very post.
A national pool doesn’t have to be a death ghetto at all if properly structured.
It will be highly entertaining watching those rugged individualists, states-right fetishists in Congress argue against having cripplingly expensive cases taken off their hands.
David Anderson
@dr. bloor: correct, reinsurance is not necessarily a ghetto. It is a risk spreader. The key question is funding of course
ArchTeryx
@JPL: Yeah. It’s basically a disease which disables an enzyme complex that processes three amino acids out of the 20 present in proteins. Those amino acids build up endlessly in the blood and eventually become lethal toxins, while the body’s cells are starved for them.
These sorts of frank genetic defects are rare specifically because they tend to be lethal at a very early age, way before the unfortunates that have them can have kids of their own. Unfortunately, this one can hang around in populations invisibly: Certain Mennonite populations have a huge incidence of it.
[EtA: One of the things that the medical industry has yet to really grapple with are the social implications of allowing people with previously-lethal genetic defects to survive to adulthood and beyond. In closed populations, this could lead to a huge spike in cases as the recessive gene loci gets spread far and wide. I know I’ve grappled with this with my Crohn’s: The likely outcome is that my fiancee and I are going to be childless. Neither of us want to risk our kids getting what we grapple with every day.]
Lee
If I’m understanding this post right: Here is how Alaska solved the problem but then again Alaska has a long history of supporting socialism ;)
JPL
@Lee: Thanks for the link. Texas is seeing pockets of the Zika virus, and they would be prudent to follow Alaska’s lead.
ArchTeryx
@David Anderson: Reinsurance isnt a ghetto, high risk pools are…they don’t HAVE to be but political forces assure they always become so. Water flows downhill and so does sewage.
dr. bloor
@ArchTeryx: Remind us again how you’re going to convince Joe’s Reinsurance, Inc. to reinsure an Anderson Health patient that everyone knows is going to cost $12 million the next calendar year.
Emily68
There are a few really bad (and expensive) diseases, such as ALS, where those diagnosed with it are eligible for Medicare no matter what their age is. Would it be possible for people with extremely high medical expenses to be Medicare-eligible, too? Congress would’t have to pass a law listing the conditions, just the minimum amount of expenses.
Frank Wilhoit
@ArchTeryx: This is why NO form of risk segmentation can work. Risk is either pooled or else it is segmented: insofar as it is segmented, it is not pooled; you can’t straddle that. Any argument against risk segmentation based on geography works equally well against any other segmentation criterion.
To your point at #9, charity will always be stigmatized and delivered in such a way as to humiliate its recipients. This is the essential sadism of humanity and runs far deeper than ANY societal action will ever be able to touch.
ArchTeryx
@dr. bloor: Contract. Contracts, in our messed up society, hold more force then the laws our politicians write – assuming they write them at all. It’s why reinsurance is still a thing, but high risk pools inevitably became ghettos and death panels, even in states as progressive as CA, and why Obama specifically did away with them.
If a particular re insurer can’t take a $12 million hit they aren’t a very good or very large one.
EtA: Insurers want perfect predictability, and while statistical modeling provides a good measure of it, the way human bodies fail is dictated as much by Ifni (the Random Number Goddess of David Brin’s books) then by anything they can predict. Thus, reinsurance.
ThresherK
I also never heard of maple syrup urine disease, and as a Yankee I’m relieved it’s only genetically-oriented.
ArchTeryx
@Frank Wilhoit: High-risk pools aren’t charity, though…not really. You’re thinking more of Medicaid, and at least NYS isn’t constantly stigmatizing me for having it, again, thanks to the ACA. (Individual private-practice doctors are a different story, but that’s thankfully been very rare. SNAP, on the other hand, is county-administered and they go out of their way to infantilize you).
High risk pool participants weren’t stigmatized or humiliated, AFAIK. They were simply frozen out of care until they died, or caught a lucky break and got out of the ghetto. Look up some of the stories of people dying on CA’s high risk pool waiting list – and those that died AFTER getting off the list, for much the same reason.
David Anderson
@ArchTeryx: They can take a $12 million dollar hit if it is only a sometimes hit.
When a reinsurer writes a contract, they go through the claims experience and the relevant local market knowledge to determine what the probability of actually having to pay out will be. If during that claims analysis that see that Mr. Doe is a $12 million dollar claim with a diagnosis that strongly implies that he is a $12 million dollar hit next year AND they see that Mr. Doe is on the reinsurance year membership file, they are going to charge the full cost of Mr. Doe’s treatment.
Reinsurance is not magic. It works against possibilities not certainty. If there is a certainty, the reinsurer is charging a premium equal to the full cost of the certainty plus admin, overhead and profit. It is cheaper for the carrier to keep the certainty in-house.
Reinsurance works for you. The carrier will pay for your maintenance meds and routine care. They want protection against your guts going into full fledged rebellion necessitating either a month in the ICU or a transplant. That is a probabilistic judgement that this very high cost scenario will apply to a very small percentage of Crohn’s sufferers and they don’t know who that unlucky person is. Again, they’ll self-insure against the expected number of crisis cases but they want coverage where if they expected four people like this but they actually get nine in a given year. This is where reinsurance does wonderful as it is effectively a risk transfer from healthy pools to sick pools.
Victor Matheson
I am proud to say that I didn’t have to look up maple syrup urine disorder. Brains and good looks all in one package.
But seriously, in response to ArchTeryx, there should be a ton of places where an insurer can’t take a $12 million hit. The exchanges insure about 12 million people. If these are equally distributed across the country, a state like Idaho (the 40th largest state by population) will have about 60,000 on the exchanges. If you want a healthy marketplace with 3 or more insurers, that is only 20,000 members per insurer even if each insurer gets exactly 1/3 of the market. That’s a $600 hit per insured person. Given an average plan cost of about $4,000 for an individual, that works out to a 15% increase in the price of insurance plans for that company. And the death spiral begins.
And this hypothetical example is pretty close to what happened in Iowa. For any plan with less than 30,000 members in a particular state, we would expect a $12 million patient to increase costs by 10%. And we might expect quite a few insurers with less than 30K members in a state.
Victor Matheson
And thanks to Mayhew for bringing up the Iowa case. Fascinating. It has generated a bunch of discussion here at my work.
David Anderson
@Victor Matheson: Thanks, the Iowa case has generated way more conversation than I expected as it is an awesome corner case/teaching example/policy quandary.
TenguPhule
You can be assured to learn something new every day here.