“For example, the life-time risk for an individual in the United States to develop Crohn’s disease is about 1/1000. How helpful is it for clinicians and patients if that risk shifts to 1⁄500 or 1/2000?”
It may be hard to tell without the context, but they are suggesting that these revised risk assessments would not be useful. My initial thought is: “If having an estimate is helpful, having a more accurate estimate would be better, and there seems to be a big difference between 1⁄500 and 1/1000.
There are common diseases you should worry about and rare diseases you shouldn’t worry about. A factor of 2 does not move Crohn’s from rare to common. The difference between a 70% chance of dying of heart disease and a 30% chance sounds pretty big, but what would you do differently? Either way, it is a big chunk of likely mortality. A factor of 2 is unlikely to change the cost-benefit analysis of actions that might protect you from heart disease. If such an action is useful, it is useful for most people.
Some rare genes do move diseases from rare to common. A broken BRCA (1 in 10k) moves a woman from a 10% chance of dying of breast cancer to an 80% chance of dying of breast cancer, and dying at a young age. Mammograms are valuable for the second woman and not for the first. Some women have prophylactic mastectomies. But if you ask Myriad to test your BRCA, in addition to this useful information, it will also talk about minor variations with useless effects on the risk.
Do you think it would be fair to say that for rare diseases (that are not determined by single loci mutations, like Huntington’s or BRCA, as you described) it’s silly to get a test because a small movement in your risk profile is meaningless in that it wouldn’t impact your treatment or behavior in a meaningful way?
Could you explain what you mean by:
Either way, it is a big chunk of likely mortality
Do you work in a related field? You explained this rather concisely, thanks.
Whether they are about rare diseases or common diseases, almost all results that you get out of 23andMe are silly because they don’t have rational effects on potential behavior. (They may have irrational effects—if you can use it motivate actions that you ought to be doing anyway, that’s great. But there are also bad irrational reactions.)
Depending on your genetics, your chance of dying of heart disease might be as low as 30% or as high as 70%. (I made up those numbers; I suspect the real range evaluable with current genetics is much narrower.) Even the low number, 30% is very high. If you have a 30% of dying of something, you should think about it and react to it. Even in the best case, you still have to think about heart disease.
From an article I’m reading:
It may be hard to tell without the context, but they are suggesting that these revised risk assessments would not be useful. My initial thought is: “If having an estimate is helpful, having a more accurate estimate would be better, and there seems to be a big difference between 1⁄500 and 1/1000.
Any thoughts?
Full article: https://d396qusza40orc.cloudfront.net/ethicalsocialgenomic/DeflatingTheGenomicBubble.pdf
There are common diseases you should worry about and rare diseases you shouldn’t worry about. A factor of 2 does not move Crohn’s from rare to common. The difference between a 70% chance of dying of heart disease and a 30% chance sounds pretty big, but what would you do differently? Either way, it is a big chunk of likely mortality. A factor of 2 is unlikely to change the cost-benefit analysis of actions that might protect you from heart disease. If such an action is useful, it is useful for most people.
Some rare genes do move diseases from rare to common. A broken BRCA (1 in 10k) moves a woman from a 10% chance of dying of breast cancer to an 80% chance of dying of breast cancer, and dying at a young age. Mammograms are valuable for the second woman and not for the first. Some women have prophylactic mastectomies. But if you ask Myriad to test your BRCA, in addition to this useful information, it will also talk about minor variations with useless effects on the risk.
Thanks for your reply.
Do you think it would be fair to say that for rare diseases (that are not determined by single loci mutations, like Huntington’s or BRCA, as you described) it’s silly to get a test because a small movement in your risk profile is meaningless in that it wouldn’t impact your treatment or behavior in a meaningful way?
Could you explain what you mean by:
Do you work in a related field? You explained this rather concisely, thanks.
Whether they are about rare diseases or common diseases, almost all results that you get out of 23andMe are silly because they don’t have rational effects on potential behavior. (They may have irrational effects—if you can use it motivate actions that you ought to be doing anyway, that’s great. But there are also bad irrational reactions.)
Depending on your genetics, your chance of dying of heart disease might be as low as 30% or as high as 70%. (I made up those numbers; I suspect the real range evaluable with current genetics is much narrower.) Even the low number, 30% is very high. If you have a 30% of dying of something, you should think about it and react to it. Even in the best case, you still have to think about heart disease.