Slashdot Mirror
New vs. Old: a Comparison of 23andMe's Health Reports and the Raw Data (enlis.com)
"With much fanfare," writes an anonymous reader, "last month 23andMe returned to reporting health information to their genetic service customers. How does their new service stack up?" According to the Enlis Genomics Blog, it's a good move but not perfect. The linked post explains that "the raw data from 23andMe contains significantly more health information than they are reporting in their health reports," and says "23andme has a long way to go to get back to reporting the same number of variants they were before the FDA ban. However – both the previous and new 23andMe reports pale in comparison to an analysis of the raw data. 23andMe’s new reports tell you about less that 1% of the health-related variants that are in their raw data." It's an interesting statistical blow-by-blow; the company making the comparison has a vested interest in you letting them run the numbers, but is not the only option.
Says the man who's mother didn't take thalidomide. Thank you FDA!
Sometimes the FDA does good things and sometimes the FDA does bad things.
I've gone to a number of talks by experts in molecular genetics (i.e. experts in interpreting individual genome sequences to understand genetic disease) and I was a bit surprised that there's pretty much universal consensus that the FDA got it wrong when they shut down 23andMe's health reporting.
What you have to understand is that the quality of the underlying tests used by 23andMe was the same as what you'd get if you ordered the test through a medical doctor (i.e. both CLIA certified). And the interpretations that 23andMe were also as good as what you'd get from a medical doctor. The difference was that, in the case of 23andMe, the interpretations were provided by a website - as opposed to being provided in person by a medical doctor.
So there were a small number of genetic counselors who resented being cut out of the loop. The claim was that it was possible to explain the interpretations/results better in person than via a website. Now, there were actually a number of studies done which didn't show any real benefit to in-person explanations. And, as an aside, as someone who spent entire semesters trying to teach introductory biology to community college students, that makes a lot of sense to me: no matter how brilliant you are, there's a limit to how much biology you can explain to an average person in even face-to-face meeting that lasts an hour - and many medical doctors only meet with their patients for 5-10 minutes.
I don't have any problem with regulating things like brain surgery and even drugs where the wrong dose can kill you. And I'm even OK with regulating blatantly false or misleading information in certain limited contexts. But there's also the right to freedom of speech. When the FDA steps in an won't let a company post useful health information that represents the current expert consensus, that seems to me to be crossing a line.
In the ideal world, if the FDA disagreed with 23andMe's interpretation of any of the variants then they would have created their own website that explained the official government interpretation of each variant. A person could take their 23andMe data and upload it to the government website and get the official government interpretation.