Is 23andMe selling Health/Genetic info to Insurance companies?

Question

I just finished the 654 "Health" questionnaire  at 23%me.

At least 10% of the questions were not related to or had anything to do with genetics, or genetic inheritance.

Questions such as where were you born, your mother, your father. Where did you grow up, near a farm, in the city, where do you live in a farm or the city.

Do you smoke and drink, are you around people who smoke.

(The last two are always found on Health and Life insurance applications, and are used to underwrite -price-the policy)

There were more, and none of them had anything to do with genetic and genetic inheritance.

Here is a link to the NCBI which addresses the use of genetic discrimination in insurance. Of course all of these laws are subject to being overturned.

https://www.ncbi.nlm.nih.gov/pmc/articles/PMC1288333/

Comments

Hi William. You might want to remove the last couple statements. I hate to see our forum inviting political arguments. There are enough arguments about genealogy and wiki editing. :-)

---

Those questions are actually related to the research on genetic inheritance. For example, if you reported having been diagnosed with COPD (I don't know if this is a question they've asked), but said you do not smoke and you are not around people who smoke, that info might be helpful in identifying genetic connections to non-smoking related COPD. Similarly, where you were born and where you grew up may indicate the presence or absence of nongenetic factors affecting your health.

---

I disagree, for one thing those weren't the only such questions that had nothing to do with genetics. Where my parents, or myself were born, has nothing to do with genetics. Whether or not my mother or father lived within a city or 1/4 mile from a farm has nothing to do with genetics. And as I said those aren't the only ones.

n some people, COPD is caused by a genetic condition known as alpha-1 antitrypsin, or AAT, deficiency. While very few people know they have AAT deficiency, it is estimated that close to 100,000 Americans have it. People with AAT deficiency can get COPD even if they have never smoked or had long-term exposure to harmful pollutants.

http://www.nhlbi.nih.gov/health/educational/copd/am-i-at-risk/

This deficiency is already known and detectable by DNA testing, so the question is irrelevant. 

My suspicion is that 23andme is partnered up with the Health and Life Insurance industry, probably AHIP. And they share the megadata, at least, with them for underwriting purposes.

, 

---

William, I guess you missed the place where I said my COPD-and-smoking example was an example. I could have said the same thing about lung cancer or any of several other conditions for which smoking is a known contributory factor, but there may also be a genetic link (or multiple genetic links).

The point is that scientific research requires measures to control for the effect of potential confounding factors. If you know that a particular factor (such as a life of cigarette smoking) predisposes a person to a particular health condition, when you are researching a possible genetic association with that health condition by looking at population genetic data, to control for the effect of cigarette smoking on your research findings, it's best to divide the research population into two groups: (1) people who smoke(d) and (2) people who didn't smoke.  23andMe couldn't possibly ask enough questions to find all of the possible confounding variables, but smoking is a causal factor in so many conditions that it wouldn't be possible to do meaningful research on the genetics of those conditions without controlling for the effect of smoking.

And COPD is not a single disease, but rather a complex of different conditions. The inherited condition you mentioned is not the only possible factor in susceptibility to COPD. My archived health risk report from 23andMe prior to FDA restrictions also describes two different genes that preliminary research (that is, not a proven association, but a topic requiring more research) had linked to a person's odds for getting COPD. One marker, rs8034191 appeared to be associated with a smoker's risk of getting COPD. Another marker, at rs7671167 in the FAM13A gene, appeared to be associated with the risk of COPD in persons of European descent, without reference to smoking. I think it's likely that other possible genetic associations with COPD are being researched.

And living near a farm is very definitely a confounding factor in studies of possible genetic factors in various health conditions. Health activists list many conditions linked to living near farming -- see http://chem-tox.com/agriculture/index.htm . I can't vouch for the integrity of that article, but a  quick online search found research documenting that people who have lived near farms have greater exposure to herbicides -- see https://www.ncbi.nlm.nih.gov/pmc/articles/PMC1480526/ -- and other pesticides, resulting in higher rates of certain cancers -- see https://www.ncbi.nlm.nih.gov/pubmed/25038451 .

---

I am not going into a long reply.  But let me take one example.

Living near a farm. Yes, living near a farm subjects a person to a variety of chemicals, pollutants, toxins and diseases, especially an animal farm.

No argument there.

But these are not genetic. I.e. unless they are epigenetic, they are not passed on to children, if for instance, one moves away from the farm before the child is born. My father for instance left a farm, joined the Marines and I was born in Phiadlelphia, spent my childhood between cities like NOLA and Atlanta, and rural Arkansas. Late childhood and most of my adolescence was spent in Philadelphia. I am quite sure that none of this was passed on genetically to my children. Ergo the questions asked by 23andme are irrelevant to genetics. (Same with smoking and drinking, unless there is such a thing as epigenetics).

But those questions are relevant to insurance companies and useful in developing underwriting tables.

---

I guess you must feel you are too old to learn the basic concepts of the scientific method.

---

Au contraire, and this is not the place for put downs and insults.Just because I don't agree with you.

---

You are entitled to not want to try to understand the science that 23andMe is trying to do, and to decide not to share your information. I wouldn't argue with you on that, but I don't want to allow potentially slanderous comments that are based on misunderstanding to be propagated without rebuttal.

---

My  comments were not potentially slanderous. But they are valid.

And my concerns were based on usage of information, which has nothing to do with science and everything to do with business practice.

I did not question the science. And you comment about me not understanding science is a misdirection. Changes the subject.

Again  my concern is the partners of 23andme, how they use the info they obtain from health reports and who they share it with, and from their privacy statements they share it with their partners, but will not sell it to third parties.

You seem extraordinarily concerned about the reputation of 23andme. It makes me think that you are an employee, otherwise I don't understand the concern about it's reputation.

I do not discourage people from using it for genealogical purposes, but I have a justifiable suspicion about their uses of these health reports.

I just bought my wife a kit and will help her in her understanding of 23andme.  But I will caution her about the health reports because frankly I don't trust the profit motive, except to fulfill their legal obligation of maximizing shareholders investment.

---

William,I and my father have been tested with 23 and Me. Originally my motivations were about health, rather than genealogy.

I have a condition which  I knew had  a genetic predisposition and that was confirmed by my tests (though I had to check it through the raw data). Many people develop it as children, though it is now known it can be developed much later in life . It's serious, and had I been born 100 years ago I would have died within a couple of years of diagnosis. Medical research in the  1930s meant that it was no longer a death sentence but it still requires daily interventions to prevent nasty complications and a premature death. It is expensive for the health care system where I live. I would hate to live with it in some countries with less comprehensive health care.

There  are more people who have the genetic predisposition that don't  get the condition than do. Many get it in childhood, but we now know that it can develop at any age. What is it that triggers it.? Why do the majority with the 'same genes' avoid it?

 To find out one has to compare people with the same genetics who don't have the condition, and won't have the foggiest idea of their increased risk,  with people who do.  Are there significant differences between them, or significant associations with people who develop it? And hence the questions, as many as possible, exploring every aspect of life. It is then possible to compare  those who develop a condition with those who don't.

The large data base provided by 23 and me makes this sort of research far more feasible and cheaper for researchers. The questions you are asked are generalised because  they may be used to explore environmental and other factors for a myriad of conditions.

The results of an analysis for any particular condition may suggest further avenues of research; perhaps towards, prevention, a cure , a vaccine or a better treatment.  Surely that is good, something that may improve outcomes in the future and hopefully save money (cheaper to prevent than pay out for a lifetime of medical care)

  I tested with 23 and me in the hope my anonymised data would be used for such research . It was actually illegal for me to send the test from here , so I had to do it through the UK rather than where I live. The irony is that  at the moment it seems that  data from samples from outside the US aren't used.

---

I understand.

---

http://www.nationalmssociety.org/What-is-MS/What-Causes-MSour place of birth, or that of your parents', may be irrelevent to your test results but for many this is of particular importance. Epidemeologists studying multiple sclerosis note that place of birth and residence for the first 15 years of life have effect on incidence rate. Also, predictive risk of the disease reduces the closer these early life places are to the equator. The big question is WHY? Is it genetic? Enviornmental? Some combination in which a genetic predisposition is triggered by an enviornmental catalyst?

Interesting article (at least to me) about this via the National MS Society:
http://www.nationalmssociety.org/What-is-MS/What-Causes-MS

---

Does this info go to research or does it go to partners in the insurance industry or both.

I guess we see what we want to see, and what we want to see is determined by our own personal experiences and expectations.

I don't see medical research coming out of these Lifestyle choices

I do see genetic disposition and lifestyle data being used for insurance underwriting..

By the way, about 2 hours ago I had an in home exam for a Life Insurance policy, for which I am applying.

The Nurse had a questionnaire and asked similar lifestyle and health questions.

---

William, if you are worried that some insurance conglomerate is going to get your DNA data, why don't you set up a free email account to use for the results and do the test under a pseudonym? They don't make you present ID and fingerprints to buy a 23 and me kit, do they?

---

I am not worried about anything. If you read all of my statements I am 77.

But I believe in caveat emptor, and just did my bit.

---

Yes. "Be not afraid." is the command to follow at any age.  :-)

---

Be not afraid? I have a rattlesnake to introduce to you, kidding.Fear is what keeps us alive, it keeps us aware and cautious.

---

You could actually just read all their legal documents online.  They have very good information about how and when they use your data.

---

That's pretty naive to believe that they put anything online that is not complementatry or crafted to salve privacy concerns.

AsI so stated. These companies have shareholders, investors,seldom are such individuals, but are in fact corporations, profit and non profit.

Universities depend on donations and grants. Grantors provide money in return for information and data.

Such ties are not listed on their websites, and since when does a thinking person, take at face value such statements.

By the way, everybody lies (be a natural or artificial person). Artificial persons )corporations) are notorious for their lies, for their lies harm people, even result in their deaths and harm the environment, not to mention other businesses.

If there were no lies, there would be no civilization, and all would be chaos.

 

So believe at your own risk

---

I've been tested by 23and me. I've been tested by Stanford University in Palo Alto. In both cases, the intent was to provide my family and ancestors a bit more information. I do not subscribe to the theory that it's wrong for either group to make a dime by collecting my blood or saliva. I'm just concerned that new information will add to research into cures from either the public or private sector.

---

So you are unconcerned that your genetic information (autosomal DNA) can and will be used by insurance companies to underwrite (that is price or deny) insurance to  society at large, and in the future when this congress overturns GINA, be used to underwrite your personal insurance.

Congratulations.

Case in point.  23andme not only collects your autosomal DNA but they also entice you to answer 653 (or more) so called health questions, many of which have nothing to do with health, but with environmental factors. 

You are aware that your autosomal DNA ia composed of SNP's (single nucleotide polymorphisms) and that the medical/scientific community has already identified SNP's that are markers for diseases, 

Ranging from Lung Cancer to Tays-Sachs.  Insurance companies that can get their hands on such information can then use that information to deny or price out your policy.

At present a law (GINA) passed by congress prohibits the use of your genetic DNA for such purpose, but we have a brand new congress with a brand new ideology and beholden to donors such as AHIP (Association of Health Insurance Providers)  What can congress can do can be undone by congress.

And it is easy to get DNA information, if you apply for an insurance policy they can, and do, require a physical exam, including a blood draw.

 

Then again I have tested via 23andme and FTDNA, but I am already insured and am a 78 year old looking at End of Life cancer.

If I were younger I would be worried.

You can check out SNP's propensities at www.snpedia.com/index

Here is a copy and paste from SNPedia

 

• rs53576 in the oxytocin receptor influences social behavior and personality
• rs1815739 muscle performance
• rs7412 and rs429358 can raise the risk of Alzheimer's disease by more than 10x
• rs6152 can influence baldness
• rs333 resistance to HIV
• rs1800497 in a dopamine receptor may influence the sense of pleasure
• rs1805007 determines red hair and sensitivity to anesthetics
• rs9939609 triggers obesity and type-2 diabetes
• rs662799 prevents weight gain from high fat diets
• rs7495174 green eye color and rs12913832 for blue eye color
• rs7903146 in 3% of the population greatly increases the risk of type-2 diabetes
• rs12255372 linked to type-2 diabetes and breast cancer
• rs1799971 makes alcohol cravings stronger
• rs17822931 determines earwax, sweating and body odor
• rs4680 varied cognitive effects
• rs1333049 coronary heart disease
• rs1051730 and rs3750344 nicotine dependence
• rs4988235 lactose intolerance

View all 107784 snps in SNPedia.

 

That;s just for starters, there are SNP's for propensity for Lung Cancer and Muscular Dystrophy, among others.

By testing family finder type DNA, and answering so called health questions you are adding to a data base, that can and most certainly will in the future be used to underwrite/deny insurance, if not to you then to others, including children and grandchildren,siblings and other relatives.

 

 

 

---

The fact that 23andme tests for genetic health propensities is the reason why I did not test with them and never recommend them. Even if that information is not being used or sold right now, someone has it. Ancestry and FamilyTree DNA do not even test for it, so IMHO they are safer.

---

The chips used for autosmal testing from the different companies test slightly different SNPs but there is a  lot of crossover between them

 Ancesry and DTDNA tests record many thousands of SNPS with variants that may be connected to health. Indeed you can use the data to find out many of these variants   by uploading the raw data to Promethease

see

https://www.snpedia.com/index.php/Testing

---

Like Helen says, the different companies are all collecting pretty much the same data. The difference with 23andMe is that they are reporting health-related interpretations to their customers. Data from the other companies could be interpreted the same way.

Answer 1

I would suggest that the first thing to do is to check any Legal (and Privacy) notices on their website. See if there is anything related about the data from the test results.

Comments

Thanks Eric: I had already tried to wade through the  five pages of legalize on 23 and me's privacy statement.  Best as I can understand is that after assuring me of privacy, they will provide their partners with any info they want. And one must understand corporate speak to understand what they mean by partners.(could and probably mean insureres)

They also sell "mega" data, stripped of personal I.D's, not a good thing as it prejudices future populations.

Personally not worried, age 77 on medicare, have a medicare supplement and life insurance policy.

---

According to their privacy policy, no, they don't sell personally identifiable information to insurance companies or any other company. 

We will not sell, lease, or rent your individual-level information (i.e., information about a single individual's genotypes, diseases or other traits/characteristics) to any third-party or to a third-party for research purposes without your explicit consent.

---

You do understand that a third party is an unrelated "person" .

However if you are in a partner agreement, or are part of a corporation, then the exchange is not with a third party.

The very first part of their "Privacy Statement"

"We collect information when you register an account, self-report information through surveys, forms, features or applications, use our Services, upload your own content to our Services, use social media connections and features, refer your contacts to us, share information through various interactions with us and our partners,

 

---

There is a difference between Third Parties and partners.

A basic course in business law informs.

http://venturebeat.com/2015/01/14/23andme-has-signed-12-other-genetic-data-partnerships-beyond-pfizer-and-genentech/

23andMe has signed 12 other genetic data partnerships beyond Pfizer and Genentech

 

---

In an agreement between two parties, a third party is anyone other than one of the two parties. Simply because one of the parties has a business relationship with a third entity, does not mean that third entity is not a third party. Therefore, in a privacy policy, which constitutes an agreement between 23andme and a customer, one of 23andme's "partners" is indeed a third party.

Note that while the privacy policy says they won't not sell personally identifiable information, it does not restrict them from selling non-identifiable aggregate information. That is what they are selling to Pfizer and Genentech.

Companies that collect personal information are very aware of the legal distinction between identifiable information and aggregate information. Selling the former is a big red flag because that is what is protected under privacy laws, selling the latter is routine. The aggregate data, which shows what genes are related to what diseases, would be very useful to drug companies and researchers, but not particularly useful to insurance companies.

---

Not being accusatory, but it sounds very much like you are a rep or employee for 23andme.

But I stand. There is a difference between a partner and a third party.

A legal difference. They don't sell info to third parties, but that statement excludes (does not include) partners.

Err on the side of caution.

23andme for genealogical purposes. Fine.

For Health questionnaire purposes. Exercise caution. Buyer beware.

I am also very concerned, or would be if I was younger, that the aggregate info, sans personal ID, would be sold to partners and third partners, that they would then use in insurance underwriting.

---

You can thank the Supreme Court for this... Corporations ARE "people" for legal purposes.

---

But Ashley, providing informtion to an investor or grantor is not "selling information"

Corporations are not fools, they employ lawyers who are experts in crafting ambiguous language.

Answer 2

I recently changed my participation in 23andme research, because this specific question remains clearly unaddressed in their policies. They hint at not doing so, and widely side-step who they share granular or general info with. The fact that insurance companies aren't mentioned ONCE in their privacy statement is enough reason for my concern.

Comments

You mean like this: https://customercare.23andme.com/hc/en-us/articles/202907860-What-is-GINA-and-could-my-genetic-data-make-it-hard-for-me-to-get-insurance-coverage-

---

That statement seems completely general and does not state how genetic data held by 23&me will be handled.  But they do state that "GINA does not extend to genetic information-based discrimination in life, long-term care, or disability insurance providers."

So, still ambiguous.

It's the vagueness over who might be classed as a "partner" that seems to be the main issue.