Wharton’s Jean Lemaire and Arupa Ganguly of the Hospital of the University of Pennsylvania discuss the implications of genetic tests on insurers.

An inexpensive gene test for Alzheimer’s disease by 23andMe has been used for years in countries such as Sweden, and it hasn’t made a ripple because of universal health care. But in the U.S., genetic testing could lead to a perfect storm for the insurance industry. Wharton statistics professor Jean Lemaire and Arupa Ganguly, a genetics professor who runs the Genetic Diagnostic Laboratory at the Hospital of the University of Pennsylvania, recently analyzed the implications of genetic tests on insurers on the Knowledge at Wharton show, which airs on SiriusXM channel 111.

An edited transcript of the conversation follows.

Knowledge at Wharton: Let’s start with the APOE 4 gene and its role in Alzheimer’s disease.

Arupa Ganguly: The APOE 4 is one of the four forms (or alleles) in which this gene comes. There are different versions of it — e2, e3, e4. People who carry this APOE 4 allele have a higher risk of getting Alzheimer’s disease. The e2 allele is a risk-reducing allele, whereas e4 is a risk-enhancing allele. Someone who is born with two copies of the e4 will have a 15-fold increased risk. Someone who has only one copy of it has a four-fold increased risk of having Alzheimer’s.

But this is not really the familial form of Alzheimer’s disease for which there is a whole different set of genes that are tested. The e4 allele is what we talk about when we are talking about people without any symptoms or family history, but who are concerned about their future risk.

Knowledge at Wharton: Obviously, it’s a big impact on a person’s life when they find out they have this gene and the potential for the disease.

Ganguly: True. This allele testing usually is done in the context of someone who is already affected, is going into a clinical trial or just before they are symptomatic. … But this test will allow an individual like me to just spit into a tube and get the results and find out the status of the gene in my system. That can tell me what is my risk from this particular gene of getting Alzheimer’s. It impacts because it’s not an easy piece of information to deal with, mainly because you cannot prevent it. Genetic information is something that one has to live with; they cannot change it.

“For insurance companies selling long-term care insurance, this is actually the perfect storm.” –Jean Lemaire

There aren’t any guidelines right now on how to prevent the onset of Alzheimer’s disease. Maybe it can be delayed somewhat, but the final onset will happen. The allele increases the risk, but it is not the only factor that is going to determine whether the person will get Alzheimer’s or when.

Knowledge at Wharton: Is there hope that through greater understanding we will be able to find a cure for Alzheimer’s?

Ganguly: Yes. The company that does the testing, 23andMe, will allow you to share this information with your physician and others so that you can be part of future clinical trials, future preventive protocols and things like that. But I think personally, the information has so much emotional impact, so much economic impact — meaning what’s going to happen, who’s going to take care of me in the future — that it leads us all to think about long-term care insurance.

I’m concerned that if someone wants to be proactive, doesn’t want to burden his or her family … financially, what are the options they will have? If you can afford it, try to plan to have long-term care insurance. And if [those plans don’t work out], that’s going to be a very devastating consequence of having this information and not being able to do anything.

Knowledge at Wharton: Jean, when you look at the insurance industry as a whole, long-term care options have been dwindling over the last decade or so. What is the impact on the industry of having this type of diagnostic ability with Alzheimer’s?

Jean Lemaire: The long-term insurance is in trouble already. A few years ago, about 100 companies were offering it. Now it’s down to 12. Essentially, people signing up for that policy are much sicker than expected, and this is the problem for insurance companies. Policyholders do self-select, people buying annuities do live longer, people buying term insurance live a shorter life, people buying long-term insurance have problems. The problem that we know is adverse selection, people trying to take advantage of some information that they have and that they don’t have to reveal to their insurance company.

For insurance companies selling long-term care insurance, this is actually the perfect storm. There’s an inexpensive test, less than $200, which is highly predictive of a very expensive disease that consumers may not have to reveal to their [insurer]. It seems to be a real worst-case scenario for insurance companies.

Ganguly: There is a dichotomy there. We want insurance, but we cannot get insurance if we reveal why we want insurance. That’s why it is a perfect storm. The people who find out they don’t have the bad allele will not go for this long-term care insurance, which is expensive. I have long-term care insurance through the university, but it’s still not inexpensive. And the sooner you start, the less expensive it is.

But as time goes on and people become more and more aware of it, I don’t know how people are going to play it out in terms of disclosing whether they have had gene tests and what are the results of the gene tests. As much as I can understand from the insurers’ point of view that this is like a selection for the sick people who want to have the insurance, I don’t know the solution without a universal health care system or something like that.

Knowledge at Wharton: There is the potential for states to opt out of certain elements of health care, and long-term care could be one of them. That could have quite an adverse effect on consumers and the industry.

Lemaire: Yes. It is a tremendously expensive disease. If you are to stay in a good nursing home in the United States, you are talking about $100,000 a year. With Alzheimer’s disease, you are talking about possibly five, six, seven years. It is a very, very expensive claim. We do face this possibility that only high-risk people will purchase insurance. Insurance companies will need to increase their prices because people who test negative will drop out of the market. We call that the death spiral, meaning policies becoming more and more expensive, insuring a smaller and smaller proportion of people. Insurance may not be the solution or the only solution there.

Knowledge at Wharton: It seems like we could be heading to a point where the companies would logically assume that if somebody is buying long-term care insurance, they are sick.

Lemaire: Yes. Right now, they are not allowed to ask any questions. Society has to decide which questions can be asked and which questions cannot be asked. Questions about race, religion, national origin, sexual orientation cannot be asked even though there may be significant statistical differences for claims. Society has to decide whether questions about genetic testing can be allowed or not. It’s a question about who in society is subsidizing whom. If insurance companies are not allowed to ask any genetic test question, they will continue to drop out of the market, and the taxpayer will pay for that in the end.

“Society has to decide whether questions about genetic testing can be allowed or not.”  –Jean Lemaire

The main question during the underwriting process is whether insurance companies are allowed to ask whether people took the test. Obviously, it’s very sensitive information for consumers. They look at it as very private information, their genetic makeup. They don’t want to reveal it. Companies want to know about it. Companies want a level playing field. Insurance companies will never request anybody to take a test, but if they have taken it, they want the prospective policyholders to reveal that they have taken the test.

Ganguly: But I think that is problematic. Why would an individual who knows they are at higher risk of getting Alzheimer’s or any other disease that’s going to cost a lot in terms of care, and therefore cost a lot in terms of the insurance policy, reveal this? People need long-term care for many other diseases. Just because someone is carrying an APOE 4 allele, that person could be deprived of having an opportunity to buy this long-term care insurance, as opposed to many others who will have other conditions for which there is no predictive testing at the moment but who will need very long-term care. Even if you want everyone to disclose test results, individuals may be carrying the APOE 2 allele, but that doesn’t mean they won’t need long-term care insurance for some other cause. So, I see that this information can be misleading.

Knowledge at Wharton: Could this open the door to these types of questions being asked in a job interview, which you’re not allowed to do now because it’s discriminatory?

Lemaire: Absolutely. Down the road, insurance companies and even employers are afraid of what could happen. The U.S. is the one country in the world where employers pay for health care insurance. Consumers may fear employee discrimination if their employers find out that they have potentially a very expensive disease. So, it’s not only the insurance market that’s going to be affected but possibly the entire employment market. The fear is that real, reliable information would be available at a fairly low price. Insurance companies are using a ton of information. Your cholesterol level, your blood pressure — this is cheap to obtain and it’s predictive, but it’s not that highly predictive. We are not talking about a 15-fold increase of the probability they’ll develop the disease here, so most insurance companies do use geneticists to try to figure out what could potentially happen.

Ganguly: I think what you’re referring to is like the genetic information derived from family history, which is easier to ask and is allowed at the moment. But what I think we should think about is how this information is available in other countries like Europe, Sweden, and other places. 23andMe has been offering this test in those countries for almost 10 years now. But because they have universal health care in many places, these things have not really had any consequence. It is going to have a big consequence in our country.

You brought up the employability aspect, which is not allowed right now because of GINA, the Genetic Information Nondiscrimination Act, but things can change. We are in a changing environment now. These things can really sway the wave of information gathered and the consequences on that in terms of health insurance, life insurance and employability. It’s going to be very concerning in the coming days and months.

Knowledge at Wharton: Did I read it correctly that the original intent of these tests by 23andMe was ancestry?

Ganguly: It was. A few years ago, it was actually stopped by the Food and Drug Administration because it was giving out information that the FDA thought people are not ready for. It has now gone back and allows dissemination of this kind of test results. People will have to opt in to get results for Parkinson’s and Alzheimer’s, and they have the option of going to a genetic counselor because many individuals may not understand before they get the test result what it means. One of the problems that we have known for a long time is that we don’t know how that individual is going to react to that information. They can have all kinds of psychological impacts, so it’s very important that the test results are communicated in the right context and the right way.

Knowledge at Wharton: How separate is the long-term care industry from the rest of the health care industry?

Lemaire: They are somewhat separate. Very few insurance companies now provide long-term care insurance. Thousands of companies in the U.S. provide life insurance, health insurance, but long-term care is a very specialized part of the insurance field. And it is, unfortunately, decreasing. Last year, only 100,000 people bought it out of over 300 million Americans. That’s not good for insurance. Insurance works on the idea that many people buy the policies, so the few who get the disease are being subsidized by many others who don’t get it. If the market continues to dwindle, the policy may not be offered anymore.

“People don’t think of buying long-term care insurance when they’re 20.”  –Arupa Ganguly

Knowledge at Wharton: We’re looking at a potential crisis if we have a dwindling marketplace for long-term insurance, yet we still have millions of people who will be dealing with diseases like Alzheimer’s. We’re going to get to a point where these people are not going to have long-term care as an option.

Ganguly: They will not. As I understand it, people will not be able to buy the long-term care insurance because it will be too expensive. People don’t think of buying long-term care insurance when they’re 20. They start to think about it as they get older. By that time, you have already reduced the pool of people who are going to contribute. As Jean mentioned, insurance works if many people buy the policy and the sick people are taken care of by the not-so-sick people, in that sense. But it will not happen if the insurance companies also ask, “What is your genetic test result?” People will shy away from getting long-term care insurance at that point.

Knowledge at Wharton: The other part is what the future looks like for the companies that provide this insurance. It’s gone from 100 down to 12. Who’s to say it won’t go from 12 to zero in the next decade?

Lemaire: It could very well be. It’s also little bit of a U.S. problem. Two-thirds of the people in nursing homes are on Medicaid. It is a strategy if you have Alzheimer’s to go through your assets in order to be eligible for Medicaid, then the government pays for it. Is that an optimal solution? Probably not. But that’s the only solution that many people have for the moment. The situation in the U.S. is totally different than the situation in other countries where they have single-payer universal health care. In some countries, companies are allowed to ask questions. In others, they’re not. It’s not obvious whether you want to take the tests or not because you’re not testing yourself, you’re testing your whole family. As was mentioned, a lot of counseling is needed for people to really understand the tests. It has the potential of being a complicated situation here.

Ganguly: Any time a genetic test is done, which is followed by casket testing, meaning that other family members are tested, this changes the family dynamics in many situations. You bring into your living room a storm that was completely not asked for before. But it also has economic consequences for the people who are actually going to take care of this sick person in the future. In a way, I feel like not knowing this would have been better for many people. We’ll have to see how people are going to go forward with it.