DNA and Insurance

Kevin Drum had an interesting post up yesterday about the problematic interaction between genetic testing of various sorts and the vagaries of the health insurance industry. It's worth emphasizing that this issue is almost certainly going to grow more intense over time as our understanding of genetics increases. Basically, you'll have a situation where unless we can substantially reduce private health insurance's role in the financing and delivery of health care, everyone's going to be going around deliberately ignorant of tons of medically useful information.

Meanwhile, I should add that though I understand the political rationale for trying to basically regulate insurance firms out of the discrimination business rather than simply killing the firms off, I wouldn't be optimistic about success over the long run. Risk assessment and risk screening is the raison d'être of the insurance firm -- it's what the business, fundamentally, is all about. We ought to have faith that market capitalism will quickly enough find ways to de facto achieve whatever risk screening is de jure prohibited. It becomes essentially a question of savvy marketing and product design to make sure that your particular insurance package is mostly bought be a disproportionately young and healthy customers.

Comments

Even in a world where there continue to be private insurance firms (and Tyler Cowen, for one, seems to think that these firms serve the useful function of generating innovative new insurance products), you can still avoid the cream-skimming problem by doing what they do with Medicare HMOs and in several European countries: the government offers insurance companies a fixed per-member-per-month fee, where the fee is adjusted so that it is higher for sicker patients.

Posted by David J. Balan | February 24, 2008 3:24 PM

"raison d'être de facto de jure"

Don't the people deserve a blog that speaks English?

Posted by Dobbs Donuts | February 24, 2008 3:25 PM

Ummm, if its an issue of genetic testing, I fail to see how 'disproportionately young and healthy' is important in risk assessment. The worry is that even heretofore healthy, young folks will be denied insurance, since their risk of a future disease will be clearly delineated, rather than relying on murky, traditional metrics that revolve around lifestyle choices and preexisting conditions.

Posted by EpicureanQuaker | February 24, 2008 3:27 PM

Yep, it is an untenable situation. All these attempts to force insurance companies to treat people as if there wasn't much better information about the relevant risks available than the premiums they are allowed to charge would suggest basically just end up being a very inefficient way to run the rough equivalent of a single payer health care system. The question is really just how much time and money we waste before this become obvious to a critical mass of people.

Posted by DTM | February 24, 2008 3:28 PM

Re: Meanwhile, I should add that though I understand the political rationale for trying to basically regulate insurance firms out of the discrimination business rather than simply killing the firms off, I wouldn't be optimistic about success over the long run.

Genetic testing does not actually increase risk-- the risk exists whether anyone knows about it or not. So it's not as if a world with common and accurate genetic testing will be a world where insurers end up with higher payouts. For that reason I can see the insurance industry supporting this sort of approach: it affects them all equally and does not cost them anything (they are paying for these diseases and treatments already) while lessening public pressure for single payor. Back in 1996 most of the insurance industry was squarely behind HIPAA for the same reason.

Posted by JonF | February 24, 2008 3:36 PM

It becomes essentially a question of savvy marketing and product design to make sure that your particular insurance package is mostly bought be a disproportionately young and healthy customers.

It's an arms race. Companies have famously done just that in the past, and state commissions then regulate those practices out.

Posted by SomeCallMeTim | February 24, 2008 3:46 PM

JonF,

The basic problem is that you can't keep people from knowing their own risks. So if insurance companies have to charge everyone the same premiums (within certain parameters), people with higher risks will buy more insurance. The payouts will increase, and the premiums will also have to increase, and more people with low risks will stop buying the insurance. Which further increases the necessary premiums, and so on.

Now you can turn around and make everyone buy at least some insurance, no matter whether they want to or not and regardless of the cost. But perhaps you then need to subsidize it for the low risk people who can't otherwise afford to pay. And suddenly, all this is starting to look like what I described above: just a really inefficient way to do single-payer.

Posted by DTM | February 24, 2008 3:49 PM

Doesn't community rating solve this problem?

Posted by Korha | February 24, 2008 3:50 PM

My Mom is 77 years old (with excellent, non-deniable Medicare coverage) and had two instances of breast cancer. On the second instance her oncologist told her she probably has a genetic marker for it but that she shouldn't get tested because then any children who get asked for a medical history might be put in jeopardy of getting denied coverage. Her DOCTOR told her not to get more information. Our system is messed up.

Posted by joejoejoe | February 24, 2008 3:57 PM

"Risk assessment and risk screening....it's what the business, fundamentally, is all about."

I'd say it's really "about" risk pooling. That's why it exists and why the consumer cares about having it.

Posted by ed | February 24, 2008 4:17 PM

Good points, and present practices are bad enough to sustain quasi-legally; DNA is just a distraction. The obvious answer is to legislate and regulate as part of the package, but such a loophole is probably the battle that the GOP will latch onto once they know they're losing the war.

I'll just repeat that you really ought to link to Drum on the Blogroll, for my convenience. The number of posts about him surely means you rank him up there.

Posted by John Haber | February 24, 2008 4:18 PM

How exactly you can continue to leave healthcare in the hands of those that will refuse to cover the treatment of those who are most in need is entirely beyond me.

Here in Britain the NHS assists all who require it, thus ensuring that those with a family history of cancer stand as fine a chance of getting the provisions they need to defeat it as someone who is unlikely to contract it.

This does not suit the rich, of course, but they can still pay for healthcare. It is hardly as if we have outlawed such organisations.

Posted by Revamp | February 24, 2008 4:39 PM

When I was at risk for a genetic disease, my desperation to be tested beat out the knowledge I might be denied insurance one day. I really cannot describe how much I had to know, in order to do whatever I could to protect myself, not financially, but physically and mentally.

Fortunately--or not--the disease I was at risk for cost very little to treat, since there really is no treatment.

Posted by KathyF | February 24, 2008 5:01 PM

This is almost right. The fundamental issue is that:

1) Getting insurance means I reduce my expected future income in order to reduce the variance of my future income.
2) Genetic testing reduces variance. Never to zero, because there are catastrophes, which are still pretty random, but it reduces variance.
3) Thus, as time goes on there is less and less reason to trade off expected value for variance reduction.
4) A regulated "insurance" market screws everything up because it places a cap on the cost of insurance. Thus, some people are guaranteed money losers. An unregulated market would charge these people some price, but it would be very high.
5) Genetic testing, in a regulated market, increases the number of guaranteed money losers.

In my view, the solution is top-to-bottom deregulation plus massive subsidy for anyone who can't afford the high prices.

It's a very thorny "solution" because when you subsidize something, the price can just go up forever. So, it's tricky. But I would like to see more proposals in this direction.

Posted by mk | February 24, 2008 5:28 PM

Re: So if insurance companies have to charge everyone the same premiums (within certain parameters), people with higher risks will buy more insurance.

A) Most people obtain insurance through their employer, so they simply have to take what they can get. Buying more insurance is only an option for a small fraction of the population.

B) The type of risk we are talking about is very long term. This is not the same situation where a person is diagnosed with a disease needing treatment in the near or immediate future. To the extent that people do buy more insurance to handle long term risks this actually benefits the insurers since these people will be paying premiums for years before they have any claims (and most oifus run pretty high health care bills sooner or later in our lives anyways). Too, this type of testing remains very inexact; it certainly cannot oredict the date one will fall ill, or even, on most cases, the certainty that one will).

Re: On the second instance her oncologist told her she probably has a genetic marker for it but that she shouldn't get tested because then any children who get asked for a medical history might be put in jeopardy of getting denied coverage.

I have never yet heard of an insurer asking for family medical history. And there was also an article in our newspaper about genetic testing and health insurance. Turns out there's not a single insurance company requiring or even utilizing this information. Moreover I believe the 1996 HIPAA law still applies: as long as one's coverage has not lapsed over two months you can't be turned down no matter what your health history (It's surprisng the number of people who are ignorant of this law). And this is not even history. In fact, since insurance comes with a statute of limitations on pre-existing conditions (if you are free of an illness for, I think, a year, you cannot be denied coverage for it) I wonder how insurers even could deal with these tests. Maybe increase premiums (but only on individual policies, which, again, are a small minority of all policies), but there would be no mechanism for denying coverage outright since there is no pre-existing condition to key off of.

I do agree strongly with enacting laws to prevent genetic data from being used in any way by insurers, but it appears that right now this is much ado about nothing.

Posted by JonF | February 24, 2008 5:38 PM

In 2005, I was a diagnosed with hereditary hemachromatosis. This means my body can't remove iron, so it builds up inside me and ultimately destroys my organs resulting in a painful death.

I found out because I was in a minor car accident and my liver enzymes were elevated and stayed that way. My doc was suspicious and I took a blood test for this illness and lo and behold I am homozygous recessive for it. I am 50% Swedish, and HH is fairly common for northern Europeans. Many of you out there are are carriers (heterozygous), likely.

Anyway, bad news, right? Not really. I found out young, so the treatment is: I give blood every 3 months. That's it. No painful death or other side effects.

The real bad news is not medical, it's on the insurance end. I can't get life insurance. I have 3 little kids. I was denied a policy at my work, through my wife's work, and through a private carrier. So if I die, they have no safety net. This is very worrying.

How can this be if it is so treatable?

Because it is grouped by the insurance companies with another far more serious illness called bronze diabetes, which is just what it sounds like: you are so toxic with iron that your skin is orange! And you have diabetes because your pancrease has been pummeled by the effects of toxic levels of iron in multiple organ sites.

My doc actually got the big insurance book down and showed me, there was nothing he could do - the two were assigned to the same category. If I were untreated, if I were to not give blood and let my iron levels accumulate for decades, then I would indeed die a painful death somewhere in my 60s from illness, potentially bronze diabetes.

So there you have it. I am a member of a permanent genetic underclass already - a 'victim' of genetic screening. I have been denied independently 3 times and it wouldn't bother me if I were single. But sometimes it stresses me out a bit when I'm just idling. And I don't smoke, do drugs, or drink. I religiously go for my treatments and keep my ferritin levels within very low limits. I am only 37.

Just FYI.

Posted by whippet | February 24, 2008 5:40 PM

JonF,

Last I saw, circa 2005, the number of people getting health insurance through an employer (either as group member or dependent) had slipped below 60%. But that is actually beside the point: often people can choose between multiple levels of insurance through their employer, and you can always add supplemental insurance outside your group plan. So, nothing is stopping most people from buying more insurance.

As for the term of the risks, I'm actually confused about your point. As soon as genetic testing indicates you are entering a relatively high risk period, it may make sense for you to buy more insurance at artificially low rates. That is true even for relatively long high-risk periods because the whole idea of insurance is to have the total premiums slightly outweigh the total average payout over the whole term of the policy. So if you have good reason to believe your total payout will be considerably higher than average over the course of the whole term, you should buy more insurance. And if you happen to know the risks won't be likely to materialize until some future period, then until that time you are in the low-risk, opt-for-less category, and then you can switch over into the high-risk, opt-for-more category, when the relevant period starts.

Now it is true genetic testing is far from perfect at predicting your likely medical expenses. However, it is getting better all the time. And the bottomline is that if the participants know a lot more about themselves individually than the insurance companies are allowed to find out and act upon, then this becomes a losing game for the insurance companies. And that will be true unless the behavior of the participants is regulated (meaning people are forced to buy insurance).

Posted by DTM | February 24, 2008 6:21 PM

Excellent analysis of the insurance business.

One caveat is that genetic testing is overblown as a method of predicting an individual's health future. Genes don't exist to kill you, they exist to help you survive long enough to reproduce and help your descendants reproduce.

Really, really bad gene variants, such as the one that causes Huntington's disease, tend to die out of the gene pool within a limited number of generations of first cropping up as a negative mutation - basic natural selection.

Most other medically troublesome genes have some positive effects that balance their negative effects -- for example, the gene for the sickle cell is very bad if you have two copies but it can help you survive falciparum malaria if you have one. So, in the malaria regions, it reaches a stable optimum (but here in America it was dying out).

This logic suggests that many illnesses are caused more by infections than by genetic flaws (germs are evolving new offenses as fast as we can evolve new defenses), so investment in unglamorous antibiotic and antiviral research might pay off more than investment in fashionable genetic research.

Posted by Steve Sailer | February 24, 2008 6:27 PM

For more on this, everyone should watch Gattaca. A great film which for some reason never got the plaudits it deserved.

Firstly its a great piece of film-making. However, it also portrays a convincing dystopic future,in which the genetic underclass is created through a combination of carefully selected embryo's and genetic testing at birth.

For anyone that needs any convincing, Gore Vidal was in it. Also, don't be put off by Jude Law. At that time he was an actor rather than a star and put in a great performance.

Posted by shariq | February 24, 2008 6:39 PM

Jesus, Gattaca was the boringest movie ever. It felt like (and I would not be surprised if it was) a ten-page story from Asimov's padded out to two hours with modernist architecture.

Posted by Senescent | February 24, 2008 6:53 PM

(shrug) It's what happens when you go from a probabilistic model to a deterministic. It takes all of "the wonder" out of it.

And Gattaca was NOT the boringest movie ever. The Unbearable Length Of A Movie (aka unbearable lightness of being) was.

Posted by sherifffruitfly | February 24, 2008 7:40 PM

do with Medicare HMOs and in several European countries: the government offers insurance companies a fixed per-member-per-month fee, where the fee is adjusted so that it is higher for sicker patients.

The Medicare HMOs (Medicare Advantage) are a giant rip off for the government because of precisely the issues Matt points out -- the risk adjustment is not nearly sufficient to counteract how well the companies select good risks for themselves. And European countries that do this (most notably Holland) A) have significant risk selection problems too, and B) have *far* tighter regulation of insurance companies than we do.

Matt has his finger on the essence of the problem. THe "bad" face of insurance (risk selection) is always going to be an easier way to make a profit than the "good" face of insurance (organizing care in some nifty and more efficient new way).

Posted by mq | February 24, 2008 7:48 PM

Gattaca might have been interesting if the main character had actually been limited by his genetics.

Posted by Barbar | February 24, 2008 7:49 PM

Gattaca might have been interesting if the main character had actually been limited by his genetics.

I always thought that made it a little more interesting. After all, genetics is NOT destiny, so this aspect is more realistic than what is superficially the premise of the society in the movie. At the same time, the main character is training to be some kind of astronaut. Leaving aside the weird police-state aspects of the constant testing, this is exactly the sort of super-elite profession where extensive genetic profiling might well be justified. (You really want to try to avoid one of your 8 person crew developing a crippling heart condition on a 3-year round trip to Mars or whatever.) So is the hero bravely breaking free from the box wrongly assigned to him by a society over-reliant on genetic testing? Or is he just ripping off his employer and putting his crewmates at risk?

Posted by jack lecou | February 24, 2008 8:55 PM

mq,
I'm not sure you're right about the risk-adjustment being screwed up in the Medicare HMO's; that had not been my understanding, but I might have been wrong about that. But the basic point is the same: the continued existence of private health insurance companies (whether because they serve a useful function a la Tyler Cowen or simply because we will prove unable to get rid of them) does not necessarily mean that we are forever doomed to have cream-skimming problems; beating that problem is no more or less difficult than getting the risk-adjustment right.

Posted by David J. Balan | February 24, 2008 9:00 PM

JonF is right. This won't increase risk, it will just distribute the costs of the risk in a different way. We'll be back to square one - we'll still have to figure out how to insure everyone without raising healthcare costs. This isn't an argument for or against single payer, it's an argument for insurance reform. Nothing new to see here. But, if we want genetic testing to some day evolve into designer drugs and more cures, we probably shouldn't emulate the NHS as some have suggested. We need to encourage innovation.

Posted by John | February 24, 2008 10:11 PM

d-squared had an interesting piece a while back on how genetic testing basically kills private health insurance dead. There's a fair amount of technical stuff in the post on insurance markets:

If you're correlating your risks and reducing the size of your pools, then you need to hold more capital against possible tail outcomes, and this can jack up your costs to the point at which it's uneconomic to provide insurance in the first place.

And the conclusion is this:

Solutions? Sorry, don't really have one, unless one seriously thinks that the genie of genetic screening can be pushed back in the bottle. I'd note, however, that the engine of most of these "problems of asymmetric information" (in this case, the adverse selection problem which makes the pooled equilibrium solution with private information untenable) is usually an embedded option. In this case, it's the option of the insured party to choose whether or not to buy insurance. Since you can't force them to buy the product, they will only do so when it's to their advantage, and this turns out to be enough to knock down the existence of the market. I speak as a member of a health insurance scheme (the National Health Service) which doesn't have the property that you can refuse to buy it if you don't want it, and would humbly suggest that something along these lines might help the insurance industry out of what might turn out to be a nasty hole.

Posted by pseudonymous in nc | February 25, 2008 12:39 AM

Whippet,

I found your story interesting and started Googling to see what I could learn about your condition and insurance. From what I read, it seems like it is possible to get life insurance, esp. if your doctor will attest to your iron levels. I assume this requires finding the right insurance company.

Of course, you shouldn't have to work harder than me to find someone to insure you, but that's out of our control at present... except to the extent that public shaming of those practicing this genetic discrimination might help, and for that I intend to do my tiny bit by blogging about it.

I hope you keep looking and find something that will cover you.

Posted by Jim Hu | February 25, 2008 2:20 AM

You say:

I understand the political rationale for trying to basically regulate insurance firms out of the discrimination business rather than simply killing the firms off.

How about explaining that some time? The only rationales I know of are either pure cant (single payer is socialist!!) or else pure power-politics. What is this alleged "policy rationale"??

Posted by Egypt Steve | February 25, 2008 8:23 AM

@ Barbar:

Gattaca might have been interesting if the main character had actually been limited by his genetics.

Actually, he was.

He was able to 'pass', and eventually go on to Mars, solely due to access to someone else's genetic material -- through financial arrangement with a member of the privileged class (That, and a sympathetic physician at the last moment).

If he hadn't had that, then he would have stayed a janitor. His life choices, his access to goods, services, and technology, would have been limited. Because of his genes.

Posted by Jemand von Niemand | February 25, 2008 10:04 AM

Dobbs Donuts: "People" and "deserve" are both originally French words too. You should have written: Aren't the folk worthy of a blog that speaks English?

Posted by Egypt Steve | February 25, 2008 10:10 AM

MESSAGE

Posted by ISHMAel back | March 5, 2008 4:01 AM