Genomics vs Insurance

Share on FacebookShare on Google+Email this to someoneTweet about this on Twitter

Free Exchange has a comment on a comment on an article about the ability of genomics-informed medicine to break the game of chance at the very core of the insurance industry. The analogy to learning not to build houses in natural disaster areas is made, but with certain diseases no prevention (post-natally) is possible.

An interesting point though re: insuring the poor. Probably you’ve all thought of it this way before:

But what about the poor? It is hard to see any reason why insurance companies should subsidize them. If society thinks that poor families should have insurance, then society should pay for it through the tax code, not slap regulations on insurance companies to keep information from reaching the market.

In a related note, there is a Google Techtalk available featuring Russ B. Altman, the guy heading PharmGKB, an ambitious project to link SNPs to pharmacological outcomes. As a for instance, something like 7% of people have SNPs that disable the enzyme that turns codeine into morphine, so they get no pain relief from Tylenol 3. So let’s not waste the insurance money buying them that medication.

If you like watching videos of people being cool on the internet, you might also want to check out TEDTalks. In one of my favorites, Dr. Robert Fischell shoots magnet guns at peoples’ heads and disrupts migraines before they start.

5 Comments

  1.  
    shoots magnet guns at peoples’ heads and disrupts migraines before they start 
     
     
    I bet that must hurt for a short while though.

  2. Amnestic said: 
    something like 7% of people have SNPs that disable the enzyme that turns codeine into morphine, so they get no pain relief from Tylenol 3 
     
    That’s fascinating, as I had surgery 2 weeks ago and when I came to I was in severe pain – so much so that I was loosing consciousness – and they gave me an IV drip of Percocet and asked me to describe the pain on a scale of 1 to 10, 10 being highest. I said 10. So they gave me another IV of Percocet and again the pain was still 10. It was only 2 hours later after they gave me an IV of Demerol that the pain went down to about 3. So does that mean that I am one of these 7%?? 
     
    I also carry the gene for red hair – MC1R…

  3. pcconroy 
     
    likely yes

  4. PConroy — NO. Percocet is oxycodone + acetominaphen. Codeine is actually metabolized in the liver to morphine, and then can relieve pain. Oxycodone directly can relieve pain. There are several different opioid medications, and all have slightly different opioid receptor profiles. Try as people have, no one has really been able to pin down SNPs for which opioid you will respond to. The majority of people do get pain relieve on at least one of the major five. Pain is a pain for medicine. The only way we can tell someone is in pain is by asking them directly — so we really aren’t that far along in terms of diagnostics for it. 
     
    Codeine, although a very popular example for genomic medicine — is a great example of how genomics is hype. The easiest way to tell someone can’t metabolize codeine, is to ask if they get pain relief. If not, just switch drugs. This takes all of 1 day and is infinitely cheaper than a lab test or even if you have a sequenced genome — the information infastructure neccessary to deliver the right info at the right time. 
     
    I think the only reall genomic test in practice is TMPT SNPs for cancer/immuno dzs — as that drug has much nastier side effects.

  5. Jor –  
    Altman’s view on the codeine story is very similar to yours. He is a practicing/practical doctor who doesn’t try to use genomics to determine whether or not to prescribe codeine. 
     
    Thanks for the pain reliever breakdown. What are the major five again?

a