About 6 months ago, I received an email from a close relative, telling me they had done a 23andme analysis, in the UK, and found they had a single copy of e4. I was able to reassure my relative that a single copy of e4 isn't that dangerous, and that one could easily reduce the chances of Alzheimer's back to average with some lifestyle changes. So it's useful information to have. I was however worried; both my mother and her mother died after prolonged and agonizing (for the family) dementia, and I could conceivably have double-e4. So I decided to get tested. And found out -- nothing.
This is because two years ago the Food and Drug Administration, an agency that is supposed to protect us from noxious foods and drugs, but in practice protects us from access to health information and cheap pharmaceuticals, told 23andme they couldn't give out health information without approval by the agency. So now 23andme will tell you if you carry genes for various rare conditions, but they aren't allowed to tell you if you have APOE-e4. Fortunately, I have a Ph.D. in biophysics, considered doing my Ph.D. with Wally Gilbert, who invented gene sequencing, and have taught courses on molecular phylogenetics. It wasn't hard to look up the chromosomal location of the APOE gene and simply read the sequence. And, after about 20 minutes of tedium, I found out what 23andme were forbidden to tell me; that I have two copies of APOE-e3 and thus have a pretty low risk of Alzheimer's. Good news, no? In fact, a friend then told me I could simply send the raw DNA sequence information to a site called promethease.com, and for $5 they'd deliver a report about a large number of polymorphisms that affect human health, though none as directly as APOE-e3. So I've found out I probably have a somewhat elevated risk of heart disease and cancer (no surprise, given my personal and family history) and a low risk of diabetes. Most of this I evaluated by doing a great deal of reading about the significance of these various SNPs. 23andme could have told me the same things, and saved me a lot of effort.
Morals? First, in the information age, the FDA is standing squarely in the path of progress, yelling 'STOP'. How can it be justified that the government can prevent a person from contracting with a company to find out information about their own genome? Second, it's pointless, because (as we used to say back at the dawn of the 'net) information interprets censorship as damage, and routes around it. A cumbersome bureaucracy like the FDA is simply not nimble enough to prevent small enterprising companies from selling interpretation of genomic information (or for that matter, from selling nootropics). All they can do is be an expensive nuisance.
Burgeoning access to information is likely to be the doom of the Mommy State, but in the interim, more and more people are going to be finding out government is usually not your friend. It's the friend of the companies who want to sell you drugs at 100 times their cost in other countries, and doctors who want to maintain their monopoly on health care.