More Recent Comments

Thursday, November 08, 2018

DNA Is Not Destiny by Steven J. Heine

DNA Is Not Destiny: The Remarkable Completely Misunderstood Relationship between You and Your Genes
by Steven J. Heine
W.W. Norton & Company, New York/London (2017)
ISBN: 978-0-393-24408-3

Steven Heine is a Professor in the Department of Psychology at the University of British Columbia (Vancouver, B.C., Canada). He has written a book about the perils of DNA testing and his main thesis is that the results of such tests are bound to make you fell unhappy because you will learn that you have a higher risk of several nasty diseases. He warns us that the science behind these predictions is not nearly as solid as the testing companies would have you believe but his main point is the psychological impact of the test results. He claims that we are not conditioned to put the results into the proper perspective because we are pre-conditioned to adopt a very fatalist view of our genetic makeup.

He had his DNA analyzed by a number of companies. Here's some of what he learned from 23 and Me.
23andMe provided me with a gripping set of predictions about my health with real concrete numbers—I learned that I have a 2.1 percent chance of developing Parkinson's disease, and this is 32 percent higher than the average person. The 23andMe experience "felt" satisfying because it provided a wealth of highly specific and personal information about my health. But then, so would the fortune-teller down the street, and at least she isn't claiming any scientific foundation to her predictions.
How many of us are prepared to ignore results like this because the difference between a 1.6% chance of developing Parkinson's is not much different that a 2.1% chance? Probably very few of us would pass that off as insignificant; instead, we would likely start to worry about when we will die of Parkinson's.

The science behind the prediction isn't anywhere near as accurate as 23andMe would have you believe but the potential psychological damage is very real. This book contains many examples of published genetic associations that have turned out to be unrepeatable and probably false. Those are obvious examples of bad science if genetic testing companies use those associations and ignore the failures to confirm the association. However, as a pschologist, Steven Heine is much more concerned with other issues such as how we interpret the results of scientifically valid conclusions.

He describes something that he calls "switch-thinking." This is the belief that a gene causes a disease and genes are like switches that are either on or off. If you have the disease gene then the switch is "on" and you are very likely to get the disease. Here's the problem: if you send your DNA for testing, then you can be certain that the results will be scary because it's guaranteed that you will have many "genes for" a number of diseases and defects. You cannot avoid thinking that your genes will determine your fate and that fate doesn't look good.
When I received the results from my own genotyping from 23andMe, I made sure to mix myself a stiff cocktail, take some deep breaths, and steel myself as I opened up my results for genetic health risks. And I'm glad I had that cocktail. Because if genes actually were the ultimate causes of diseases, I'd be in big trouble. According to 23andMe, I have "the genes for" psoriasis, prostate cancer, chronic kidney disease, melanoma, Parkinson's disease, ulcerative colitis, multiple sclerosis, narcolepsy, stomach cancer, Hodgkin lymphoma, scoliosis, restless leg syndrome, male breast cancer, sclerderma, hypothyroidism, esophageal squamous cell carcinoma, hay fever, chronic lymphocyte leukemia, and high blood pressure. And these are just the elevated genetic disease risks for conditions that 23andMe rates as being grounded in solid scientific evidence—the list is much longer when 23andMe also includes conditions that scientific consensus rates as being more iffy. I had to keep reminding myself to avoid switch-thinking, but these heightened risks still made me anxious.
There's a lot of interesting stuff in this book; for example, there's a discussion about genetics and race and another about the misleading use of "genes for disease X." These topics alone make the book worth reading but the main point is the best reason. Steven Heine reminds us that while there might be some positive benefits of genetic testing for health reasons there's also a cost. That cost might be your psychological well-being. Although he doesn't say so explicitly, the general impression one gets is that the cost isn't worth it.

I highly recommend this book to anyone interesting in genetic testing and genetic counseling.


  1. "That cost might be your psychological well-being."

    I think there are two different arguments going on here:

    1) The rational argument (true to some degree) that going from genotype to probability of disease isn't always that accurate at this point in time.

    2) The emotional argument that it is "better not to know such things" even if they are true. I strongly disagree with this. This reminds of the horribly unethical way physicians used to hide diagnoses of terminal diseases from patients using the argument that it would just make the patients depressed. So you had people thinking that they were getting better and not doing things like making wills, reconciling with estranged relatives at the like, that they would do if they knew they were dying.

    1. I assume you have already had your DNA analyzed since you "strongly disagree" with the idea that other people might find the emotional distress upsetting.

      However, let's be clear about one thing. The fact that YOU wouldn't be affected by knowing you have a 32% increased risk of Parkinson's does not mean that you can tell everyone else to get their DNA tested. That's not your call and your analogy is completely bogus.

    2. Yes, actually -- by two different companies, in fact! And analyzed the data myself in addition to accepting the provided interpretations. The whole scientific worldview is based on the idea that knowing is better than not knowing -- there are no "things Man was not meant to know" as 19th century Romantic philosophy would have it.

  2. I have not read this book, but I will venture to make the following argument anyway: There are quite a few studies that show that this fear of phycological stress from direct to consumer (DTC) testing is unwarranted. Increased anxiety has been commonly used as an argument aagainst DTC, but it is unlikely that it is a valid one. A key reference: "The impact of personal genomics on risk perceptions and medical decision-making" (

    1. Heine is not arguing that DNA testing should not be done. He is just pointing out that there's a possible downside based on our common misunderstanding of cause and effect and probabilities. He summarizes many studies showing that most people don't suffer any stress because the news isn't bad and they don't pay much attention to the details as long as there isn't anything that stands out. (Many of the studies were done by him and his collaborators.)

      On the other hand, there are some people who suffer as he did until he was able to rationalize the results and learn that some of them were incorrect. It's likely that the more knowledgeable you are about the tests and the results, the more likely you are to worry about how it will affect you and your children.

      He quotes from the largest study to date showing that among people who choose to do the test (already a biased sample) about 10% experienced considerable anxiety when they got the results.

      Given that you are unlikely to learn anything useful (i.e. actionable) about your genes if you submit your DNA for analysis, you should be aware of the possible anxiety that the results may cause. If you think that's going to be a problem, then don't do the test. The probability of benefit is very small.

    2. SciPhu, the paper you cite does not mention psychological stress at all. It deals exclusively with how patients evaluate their risk for certain diseases (Alzheimer's disease, Parkinson's disease, breast cancer, lung cancer, colon cancer, prostate cancer, type II diabetes, coronary heart disease, and obesity) before and after getting results from DNA testing.

  3. Have not had this analysis done to me. But I was interested when I came across a YouTube video by Buzzfeed. They took two groups of their employees, ones who identified as of mixed race, and had 23 And Me do their DNA and infer their actual ancestry. In the video this is presented as if they were astonished to find out who their ancestors really were.

    The video is here:

    But if you look at what they said they thought they were, it turns out that they were mostly very accurate in their assessment. What they were astonished by was the small percentages of ancestry from various odd places. Now a figure of, say, 1.4% Turkish, is dubious because the standard error might be high enough that this whole assignment is questionable. What surprised me was how much they overreacted to minor and questionable information. I suspect this occurs with health assessments too.

    1. My DNA results tell me that 2% of my genome is from Russia and 3% is from the Baltic States. I know quite a bit about my ancestors so I know that these estimates are very likely to be correct.

      On the other hand, my initial score from in 2016 said 3% Iberian Peninsula. That didn't seem very likely but it didn't disturb me. The latest update has eliminated that estimate from my score indicating that some of these estimates are not reliable.

    2. Steven Heine tells us that his results from 23andMe indicated some Chinese ancestry. As the company puts it, "Along your mother's line, you have ancestry in Eastern Asia in the past few hundred years."

      His mother and his sister were quite surprised, as was he.

      "However, a closer inspection suggests two key problems with my proposed maternal line of Chinese ancestry. The first problem is that it's probably not true." (p. 142)

      It turns out that his mitochondrial haplogroup is X2c1 which is, indeed, common in China. However, it's the most common haplogroup in the world and people with this haplogroup live throughout the Northern hemisphere. It's very misleading to say that having this haplogroup indicates Chinese ancestors.

      "23andMe again shows its ability to deliver an exciting and engaging product,even if the information is sometimes embellished far beyond what is scientifically justifiable." (p. 142)

  4. I have Marfans syndrome in my family. A connective tissue disease (genetic) that manifests itself in any of numerous symptoms. Myopia, scoliosis, heart valve prolapse, aortic aneurism, spontaneous pneumothorax, recessed chin, elongated extremities, pigeon chest..,. I have myopia, scoliosis and collapsed lung, but none of the others (yet). I have never had the genetic test for it, but there is little doubt that I have it. But, seriously, how would I benefit by knowing for sure?

  5. This is very welcome as a criticism on these DNMA tests.
    I understand he is suggesting they are wrong and not just unwelcome information?!

    i think the concept of our diseases being already inprinted on our dNA is unlikely.
    Something there, but also not there.
    In fact everybody must have cancer risk if we lived long enough.
    Yes the odds would falsely persuade people they are just waiting for a list of diseases and who beats who.
    As a creatyionist i find something fundamentally wrong with DNA concepts at this point in history.
    They are not accurate trails backward or forward.
    Its still entry level.
    I vote NO to doing these tests.