healthHealth and Medicine

There's One Christmas Gift Idea That Could Tear Your Entire Family Apart

Gio.tto and Yuganov Konstantin/Shutterstock

Sure, it sounds like a nice idea. Give your loved ones a present that will provide insight into their heritage, potentially connect them with long-lost relatives, and inform them if they carry risk genes for lactose intolerance, weight gain, or serious diseases.  

But we would advise you to rethink doling out direct-to-consumer (DTC) DNA test kits, like those offered by and 23andMe, to everyone on your holiday list. That is, unless the intention of your gifting is to terrify friends with inaccurate warnings about the state of their health or tear your family apart by digging up the type of secrets that should stay buried. It’s your call.


See, as fun as these services might be – who doesn’t want to know how much Neanderthal DNA they have? – the industry behind them is claiming to offer insights that overstep our current understanding of human genetics, and at the same time, creating unprecedented privacy issues that our legal systems have yet to catch up with.

Why They’re Unreliable

Far removed from the costly and time-intensive process of sequencing an entire genome, DTC tests analyze ancestry and health traits by assessing the presence or absence of known DNA sequence variations in certain stretches of the genome. These variants, often different from one another by just a single nucleotide (hence why they are called SNPs: single nucleotide polymorphisms), have become associated with particular ethnic groups or the development of certain diseases based on the results of more far-reaching genetic analyses that were performed in reference populations. For example, scientists who looked at the genomes of Ashkenazi Jews found variants that a percentage of these people carried some of the same variants, and thus, if a new subject has some of these variants, it’s a good guess that some of their relatives belonged to this ethic group.

Similarly, investigations involving people with cancer or dementia have revealed numerous mutations or gene variants that appear to increase the odds that such illnesses will develop.


After determining which of the predetermined variants of interest each customer has, the companies use statistical programs to calculate the likelihood that this unique genetic profile belongs to someone from Scandinavia vs someone from Southeast Asia, or someone with a good metabolism for carbohydrates vs someone who should lay off grains. It’s an estimation at best and misleading nonsense at worst.


Ancestry estimations based on such data are prone to error simply because inheritance is convoluted and the reference populations used as benchmarks for each major ethnicity are still quite small (though accuracy has improved as more people do the tests and add their DNA to the databases).

Creating health reports with these methods is much more problematic. Most conditions are mediated by a complex interplay of many genes as well as epigenetic changes caused by environmental factors. However, no DTC analyses are capable of taking this into account. None of the current tests look at the status of all the genes believed to be involved with each disease reported on, and in fact, none even sequence the entirety of individual genes.

For these reasons, claiming a DTC test holds answers to one’s medical destiny is a bit like a quack doctor claiming they can diagnose exactly what’s wrong with you based on your runny nose and sore throat. We need to know a lot more about the genetic underpinnings of these diseases before we can determine the consequences of certain variations.


What Could Go Wrong?

Well, for starters, these genetic services have been proven to flag harmless gene variants – and those whose significance is not yet known – as pathogenic, meaning that customers are receiving worrying false alarms about their disease risk. This, in turn, can trigger anxiety and drive people to seek expensive yet unnecessary medical attention and/or products.

A recent study showed that customer-oriented reports issued by DTC test providers and some of the third-party companies that evaluate the raw sequence reports misidentify risk factor genes as much as 40 percent of the time.   

And now that we have covered the technical limitations of the tests and how they can mislead people, let’s look at the social disasters. All the major DTC companies allow you to opt-in for relative matching, as in they compare your genetic profile to those of other customers and alert you if they find a significantly close family member. This is wonderful for people seeking more connections, but as several recent high-profile cases demonstrate, may also reveal some hard truths.


Back in spring, a 36-year-old woman from Idaho discovered that her dad was not her biological father after taking an test. Instead, Kelli Rowlette’s results showed that she was a closely related to someone named Gerald Mortimer. It turned out that Mortimer, a fertility doctor who had treated her parents when they struggled to conceive, had secretly used his own sperm to inseminate Rowlette’s mother. Obviously, the revelation shook Ms Rowlette and her parents, who are in the process of suing Dr Mortimer for fraud, medical negligence, battery, emotional distress, and breach of contract.

And just earlier this month, Buzzfeed posted a video series about a young woman who was shocked when her parents informed her that her and her brother’s biological father was an unknown sperm donor. After keeping the information secret, the parents were reportedly forced to break the news when the daughter took a home DNA test.

Another bizarre, unanticipated outcome of the DTC DNA test industry has been criminal identification. Earlier this year, news broke that California police had tracked down the elusive Golden State Killer by comparing DNA left at crime scenes to profiles in an open-access family-finding site called GEDmatch. Using the database's 'familial searching' tools, the detectives found that several distant relatives of the suspect had submitted their DNA, allowing them to ultimately zero in on 72-year-old Joseph James DeAngelo, who was arrested in April.

According to Harvard University, this was the first case of law enforcement utilizing a DTC genetic database, and it was most certainly a paradigm shift. Since DeAngelo’s arrest in April, at least 13 other cold cases have been solved with this approach.


And though this use of the databases has promise for bringing dangerous offenders to justice, it has raised very real privacy concerns. As of now, in the US, DTC companies claim that customer's results will not be shared with third parties such as police, insurance companies, or employers without express permission from the customer themselves. And and 23andMe, undoubtedly the industry leaders, give the option to opt-out of their relative-identifying service and have also stated that they do not allow law enforcement to conduct familial searching using their databases. 

Yet, according to lawyers at the Electronic Frontier Foundation, there are no clear legal protections for our genetic data once it has been put out there. Senior attorney Jennifer Lynch notes that the only safeguard against improper access to and misuse of customer's genetic information is the vague privacy policy of each company.  

Moreover, the privacy policies for and 23andMe both state that customer's information will be shared with law enforcement under the circumstances of a valid court order, subpoena, or search warrant. 

Is It Worth It?


If neither you nor anyone in your family are on the lam for a serious crime and you just want a rough approximation of your lineage (that you will take with a grain of salt), then by all means, buy one of these kits. But if you are not willing to run the risk of wreaking havoc in people's lives and interfering with their fundamental rights to privacy and self-protection, maybe get them something else... Like an Amazon Echo or a Google Home. Oh wait, never mind.


healthHealth and Medicine
  • tag
  • DNA,

  • genome,

  • 23andMe,

  • gene,

  • allele,

  • variant,


  • golden state killer,

  • GEDmatch