Reader survey results

Thanks to everyone who participated in last week’s reader survey, which is now closed. We received 252 responses (excluding Genomes Unzipped members), which is fantastic. I’ve made an anonymised version of the results (with names, email addresses and all free-text fields removed) available here.

We’ll be breaking down the data over the next week or so, but I wanted to draw attention to some of the more interesting immediate findings here. Please bear in mind that this was not a scientific survey, and the respondents are a wildly biased sample from the general population, so the results are absolutely not representative of the broader population.


Readers who responded to the survey were predominantly male (63%), with European ancestry (90%, with only 21% indicating at least some non-European ancestry), highly educated (only 5% had no tertiary education, and 41% had a PhD or equivalent), and mainly from English-speaking countries (63% from the US, 13% from the UK). There was a fairly broad range of ages – roughly 50% of our readers are between 20 and 40, but 29% are over the age of 50. Exactly half of the respondents currently have children, with a further 18% planning to have kids at some stage.

Our readers generally have a strong background in genetics: 37% of them described themselves as “experts” in genetics, and a further 38% said they would be able to “explain key genetic concepts to others”. Nearly 40% of the respondents described themselves as research scientists working in genetics/genomics.

Genetic testing

A remarkable 65% of our readers have had at least one genetic test.  The most popular test by far was 23andMe‘s complete scan – over 50% of the survey respondents had done this test. Personal genomics competitors deCODEme, Navigenics and Pathway Genomics were well behind, capturing 7%, 4% and 3% of respondents, respectively. Genetic genealogy tests were more popular: 17% and 20% of readers had looked at Y chromosome and mitochondrial DNA markers, respectively, and 12% had taken Family Tree DNA’s Family Finder test. Six respondents reported that they had complete genome sequences: 1 from Knome, 2 from Illumina and 3 from “other sources”.

In terms of motivation for having tests done, the most popular response was simple intellectual curiosity (89%). Interest in health-relevant information (58%), interest in deep ancestry (54%) and interest in genealogy (49% for “tracing my historical family tree” and 33% for “finding new living relatives”) were also reasonably popular.

Of those individuals who had taken a direct-to-consumer genetic test, over 50% discussed the implications with their family beforehand and 93% discussed their results with their families after taking the test, but only 34% discussed their results with a clinician. Around 70% of people who had taken a genetic test had not changed their behaviour as a result, but 14% reported altering their diet, 11% reported increasing their exercise levels, and 8% said they had taken a medical test or increased screening as a result.

We asked if people had experienced direct harm from a direct-to-consumer genetic test or knew of someone else who had, and 10 people (4%) replied that they had – we’ll be getting in touch with some of these people for more details and writing that up as a separate post. [Added in edit 10/12/10: On closer inspection of the text provided by respondents (see comments) only two respondents out of 166 (1.2%) who had taken genetic tests reported negative experiences affecting them directly, one referring to inadequate interpretation of a pharmacogenomic test and a second reporting never receiving results after having paid for a test with a Canadian company. There were no reports of serious medical harm resulting from receiving results from a DTC test. Again, we’ll discuss these responses in more detail next week.]

Inevitably, we asked people how much they would pay for a complete genome sequence. Only 5% of respondents said they planned to never get their genome sequenced. Nearly 90% of respondents would pay $100, 58% would pay $500, and 26% would pay the “arbimagical” $1000 value often touted as the tipping point for widespread adoption of genome sequencing. A single respondent said he would pay more than $10,000 (the same individual said they had already been sequenced by Knome, so this is apparently not just idle speculation).

We also asked about people’s willingness to donate their genetic data to a project like Genomes Unzipped under various conditions, to get a sense of public interest in this possibility (the project is currently exploring options for expanding to include other individuals, but I should emphasise that we have not yet decided whether or not this is an ethically, legally or logistically feasible option). Surprisingly, the profile of responses was nearly identical when we asked about genome scan data (e.g. from 23andMe) and complete genome sequence data, suggesting that most people view the release of both types of genetic data in much the same way: around 10% said they wouldn’t consider donation under any circumstances, 36% would consider it if there was a guarantee of anonymity, 33% would donate without a guarantee of anonymity so long as their name wasn’t publicly linked to the data, and around 20% of respondents (about 50 people) would donate their data with no strings attached.

Ethical challenges of genomics

We asked two very crude and preliminary questions about attitudes towards genomics, as a kind of pilot for a more detailed survey down the line. Briefly, around 50% of respondents thought that newborn babies should routinely have their genomes sequenced once the technology to do so becomes affordable, and 70% of readers felt that parents/guardians “should be able to consent minors for direct-to-consumer genetic testing”. These figures alone are interesting, but they become more interesting when we break them down by demographic parameters: for instance, 59% of females felt that newborns shouldn’t be sequenced vs only 40% of males. We’ll dig into this more in a later post.

Finally, we asked a free text question about which public misconceptions about genetics readers were most concerned about, and how they think those misconceptions should be addressed. We’ve removed responses to this question from the public data for data privacy reasons, but we’ll blog about them soon.

  • Digg
  • StumbleUpon
  • Facebook
  • Twitter
  • Google Bookmarks
  • FriendFeed
  • Reddit

10 Responses to “Reader survey results”

  • very interesting esp that people were mostly having tests, out of ‘intellectual curiosity’! (nb that maybe represents the sort of sample of people who look at this site- which is quite a specific slice of the population!). health wise i was v interested to note that 70% of people hadn’t changed their behaviour post results. i wonder of those that did, whether they had been identified as predisposed to specific health risks,(if so, which?) and whether they felt that there was clearly an environmental factor which could positively influence their risk rates (such as being at increased risk of a multifactorial illness such as heart disease- so they felt increased excercise/ better diet could help?)
    i wonder if many of those that didn’t change their behaviour, had also been identified as having similar sorts of risks, and risk levels to the ones that did? i bet the answer is yes.
    around health particularly i think interventions and impacts and expectations are important. this was something i thought the “baby barcode” question threw up.
    clearly one of the very practical concerns with managing expectations (and concerns!) is genetic counsellors /the medical profession being swamped with people worrying about their essentially perfectly healthy babies- how to process the information, and negotiate risk? who would pay for all the counsellors? etc etc.
    beyond that, there is a broader question of what sort of interventions might follow (not a lot of changed behavior from your sample!)- of course postnatal interventions for CF diagnosis or duchennes etc are ‘special cases’ – but generally for the majority of ‘healthy but possibly predisposed to something or maybe not, who knows’ babies-well what would be the end game here in terms of genomically- informed interventions for broadly healthy babies? it is worth making a broader public health point, that there is plenty of data to identify that being born in a specific postcode area significantly affects life/health expectation. if a baby is born in such a postcode zone, what is the intervention?

  • Very interesting – especially the responses to the more ethics-based questions!

    re: changing lifestyle or behavior in response to genetic tests, I think it’s important to control both for a person’s lifestyle and habits prior to taking the test, and for the person’s test results. Since the sample is not representative of the broader population, it’s possible that the sample is healthier or more health-conscious to begin with and did not need to make significant changes to behavior, or did not receive significant indications to change behavior (no significant increased risks). Now that I think about it, those would have been good questions to suggest for the survey! :p

    “How would you describe your diet, prior to taking the test?”
    “How would you describe your level of physical activity, prior to taking the test?”
    “How would you describe your overall health, prior to taking the test?”
    “Did you receive results indicating significant increased risk for one or more diseases for which preventive measures exist?”


    And motivations may play a factor as well – e.g. the main impetus was intellectual curiosity rather than a medical exercise. Are we interested in knowing whether people change their behaviors simply by virtue of receiving health-related risk information, or whether they change behaviors in the context of seeking out health-related risk information?

  • I tested with 23andMe out of curiosity, and after discussion with my Doctor, I am now taking medication to hopefully reduce the risk of AMD, and have had a scan for Thyroid cancer as I showed increased risk for both these problems, and my mother had both, dying last year aged 97 from the effects of the Thyroid cancer.

    I see 23andMe testing as an opportunity to take action to change diet, behavior or monitor possible problem areas.

  • Daniel MacArthur

    Good points, Alex and Shirley. Shirley, I think your point about dissecting out the effects of receiving vs seeking health-related information is particularly important – we’ll need to think carefully about how best to address that in future surveys.

    I suspect a major factor here is that most healthy personal genomics customers don’t actually receive information sufficiently worrying to warrant behaviour change – and our genetics-savvy sample is capable of recognising that! However, I would hope that those who did receive important information would also be savvy enough to act on it (and the fact that 34% discussed their results with a clinician is perhaps heartening in this respect).

    Hi Dennis – glad to hear the information was useful in your case.

  • I wonder why many people are willing to pay more than $500 for some type of medical diagnosis test or procedure, but would consider genetic testing only if it were cheaper. Why are we willing to pay more for an MRI (cost range $400 to $3500) than a genetic test? Could it be that many think that knowing their complete genome sequence won´t be useful in preventing or diagnosing diseases at an early stage and is therefore not worth the money? Maybe it would be good to start a general-public campaign that will explain how valuable genetic testing can be and how the information will become more relevant in the future as we gain more insight about the molecular basis of various pathologies.

    It would be interesting to investigate what leads to the perception that genetic testing is different than other types of medical tests.

  • I was curious about the “We asked if people had experienced direct harm from a direct-to-consumer genetic test, and 10 people (4%) replied that they had …” statement because that fraction seemed high to me. In looking at the table of data, I learned that the question that was actually asked was, “Have you experienced or has someone you know experienced harm from a direct-to-consumer genetic test?” And looking at the responses I saw that 4 of the 10 people who responded affirmatively had never had a genetic test themselves (if I am reading the table correctly). So at least in those cases the statement about harm was referring, most likely, to the experience of another person. Am curious to know if that was the case for some or all of the other 6 as well.

    Thanks for providing access to much of the raw data.

  • Daniel MacArthur

    Hi Joanna,

    Digging through the text responses (which we stripped out for privacy reasons) has been illuminating. Four were not really harm in the normal sense, or couldn’t be deciphered:

    • * no details given;
    • * uninterpretable (“I”ve been in a media war for a year or so.”);
    • * two separate respondents know customers disappointed by unexpected or boring ancestry results.

    The remaining 6 responses were of more interest:

    • * respondent knows someone affected by 23andMe sample mix-up, noting that if there hadn’t been obviously incorrect ancestry information the customer might have acted on at least one risk report;
    • * respondent knows a customer unduly concerned about reported diabetes risk;
    • * respondent has counselled “several” patients who were falsely reassured by Navigenics reports of low colon cancer risk, despite having a family history of the disease – respondent says “[n]ot so much harm as a lot of confusion and frustration on the part of the patients (“what did I pay for?” type of reaction)”;
    • * respondent has counselled 2 individuals who tested positive for BRCA and APOE-E4 and received inadequate counselling from DTC providers (23andMe and Navigenics, respectively); respondent acknowledges inadequate counselling similar problems would likely have occurred had they received results from primary care physician;
    • * espondent argues that interpretation of respondent’s own CYP2D6 null allele homozygosity by Pathway Genomics was seriously inadequate, referring to tenuous association with Tamoxifen metabolism and not reporting better-established links to well-established, serious drug metabolism problems;
    • * respondent ordered a DTC genetic test from a Canadian company, BioResolve, paid ~US$400-500, and has never received results – company no longer responds to email queries. [I note that the company has been the subject of numerous customer complaints, and their website seems more-or-less defunct].

    So, we have just two respondents who report being directly affected by problems with their own test results (i.e. around 1.2% of the 166 respondents who reported having taken a genetic test), as well as several credible second-hand reports of problems. However, overall I think the picture looks pretty good for DTC companies, consistent with other recent (and far more thorough) surveys of customers.

  • Hi Dan,

    From what you’ve written, it seems like only one person had a direct adverse experience with results from a DTC test (and even that example was not yet an actual experience of harm, but of perceived future harm from inadequate information). Would you consider revising the post itself to reflect the general conclusions of your more detailed analysis of the this question? For example, clarifying the question that was actually asked (“have you experienced harm or do you know someone who has…”) and then discussing the fact that at most 1 (0.4%) person directly experienced “harm” from their results?

    I ask because subtle confusions like this seem to drive a lot of the fear-based rhetoric out there about DTC genetic tests!

  • Daniel MacArthur

    Hi Shirley,

    Sorry, I somehow missed the fact that I had failed to report the question correctly in the post (despite Joanna’s direct hint). I’ve fixed that and added a clarification of the responses received, to try to head off any misinterpretations. I apologise for being unclear in the original post.

    Again, this is a biased sample, so the results can’t be interpreted too broadly, but I think the overall message regarding customer experiences of DTC testing is a pretty positive one, consistent with other recent studies.

  • Very tardy, but I’ve just added a small analysis on my blog:

    Does having children affect views on genetics?

    No time to be more prompt as I’ve been busy with the kids…..

Comments are currently closed.

Page optimized by WP Minify WordPress Plugin