This week the FDA warned a google-backed company called
23andMe to stop sales of its DNA testing services. 23andMe provides a self-administered DNA test
that customers send back to 23andMe and receive detailed genetic reports,
ancestral information and health recommendations. The FDA’s concern is focused on customers
opting for unnecessary treatments and surgeries based upon information received
from 23andMe.
This, of course, is a legitimate concern. But when I heard it was google-backed, my
thoughts went to data. And, as expected,
the privacy statement on 23andMe’s website has a large section devoted to the
sharing of “Aggregated Genetic and Self-Reported Information.” The aggregated genetic data, stripped of
individual identification, could be used in peer-reviewed research
studies. Such assurances that individual
identification will not be linked to shared data may seem adequate on the
surface, and sufficient to satisfy many considering signing up for their services.
However, the deeper implication is in what the aggregate
data is being used for. Who is doing the
study and for what purpose? Is it going
to insurance companies? Can they be
searching for health trends and adjusting policy accordingly? Might it be food, alcohol, or tobacco
companies looking for ways to combat criticism of their industries?
This might seem to be a very cynical point of view. I do recognize that there are those who would
take this data and use it to help society.
But, this story demonstrates again the need for people to understand the
implications of data mining, and to think more critically about how their data
is used whether or not their names are attached to it.