So physicians should understand that genetics is very important, but it’s not the answer to all the problems we’re dealing with, by a long shot. Personalized medicine is important, but just being able to bring information to the physician-patient relationship, is very important. And it’s not going to be very long before the kinds of information can be ground out by some really interesting and modern data mining and statistical programs, so that you can say, OK, if you’re this, and you have this record, and you live this way, and you’re this many pounds overweight, and you have this kind of regimen, these are your risks. Doctors will be able to share those kinds of predictive analytics with their patients.
And we’re not really certain how much is environmental and how much is genetic, in the explosion of diabetes right now?
For many, the idea of the increase in diabetes has to do with overweight, obesity, and over-nutrition of various kinds. Now, those will be important; but it can’t explain a lot of the increase in diabetes. It can explain some of it, but certainly not anywhere near all of it.
Might there be some environmental elements, as in the natural environment, at play here?
My favorite idea is that it has to do with a change in the microbial populations in humans, with shifts in the microbiome. We know that those can be causal, in mouse experiments. And we know that diabetes and changes in the microbiome can be causal in humans. We don’t know that changes in the microbiome can be causal in humans; we know it is true in mice. I have a strong suspicion that changes in the microbiome in humans can have a causal effect. And that would be great if that turned out to be true, because at least we’d have something to look very carefully at. We’re learning a lot, but we still don’t understand the balance of the microbiome and the ways in which that shifts things in humans, and how much the history of childhood through adulthood has an effect.
Do you believe that the two streams of information, from population health management, and from pure scientific research, can be complementary?
Absolutely. Just doing really good mining of data from patient records, can provide some absolutely key clues that one can use to try to identify fundamental causes.
I think it will be an interesting time in the next five years, as information will be flowing into physicians and healthcare organizations, from two different points.
Yes, and the shift from paper records to electronic records, and finally, to query-able data, will provide interesting and very powerful information for medical research. And there will be problems with privacy and those things, but things have changed so rapidly. So this is probably the best time ever to do biological research.
Is there anything you’d like to add?
I think that one of the things that we should realize is that the computational and the underlying mathematical approaches, to doing this kind of mining, and finding these clues in both the patient data as well as in research-level data—we need a lot more people to pay attention to developing new ways of doing this kind of thing. So getting computer scientists, mathematicians and other people who are trained, to work with both scientists and physicians, is a major new area that we should emphasize. The value of some of the more abstract and arcane mathematics and computer science will be really important for the future of human medicine. When you analyze all kinds of data from climate and agriculture, or whatever, those may be complex, but there isn’t anything I know that’s more complex than the human body; and that requires new approaches.