Monday, March 2, 2009

Personalized Medicine at the Point of Care?

I'm concerned that the speeding bullet of genomic discovery is going to outpace our ability to keep up with the demand for tests and our ability to fold the results and knowledge back into the EHR for consumption at the point of care and the process of delivering care. In fact, we already know enough about the effects of genetics on certain healthcare-related conditions, such as warfarin metabolism, that should be presented at the point of care, but largely are not.

This quest for "personalized medicine" seems like a supply chain of data-- i.e., Are we generating the data content in the machinery of healthcare to enable the production of personalized medicine? In my limited brain, the supply chain of data includes these data types and sources:

• Patient demographic data (Registration, patient reported, etc.)
• Clinical treatment data (Epic, Cerner, pharmacy, claims data, etc.)
• Clinical outcomes data (Epic, Cerner, pathology labs, diagnostic labs, other diagnostics, patient reported outcomes etc.)
• Familial data (Patient reported)
• Biological data (Labs)

Yesterday, at a demonstration of Xenobase at Childrens Memorial Research Center, as I was walking through this mental checklist of supply chain data content and sources, a couple of things struck me: (1) If I wanted my genome analyzed for inherited disease tendencies, I wouldn't know who to ask at Northwestern for that analysis; and (2) I didn't know if our hospital labs were running those tests or not and whether that data was somehow making its way back into our Cerner or Epic EHRs, or our Enterprise Data Warehouse. However, I do know that the Northwestern Core Genetics lab can run the tests... but how does that data make its way into my EHR? I can have commercial organizations like run a personal DNA analysis... but I wouldn't know what to do with the results if they indicated a genetic risk of disease, nor would I know if I should trust the results, in any case.

It seems to me that we need to do a few things in the near term: (1) Engage EHR vendors in this discussion and ask, "What's your plan for incorporating the knowledge from genetic medicine into your decision support environment?; and (2) Move the collection and discovery of "genetic medicine" from the background of research into the foreground of healthcare operations. I think the term for this now is "translational" medicine. Genetic tests should be offered to those patients who are interested, but those patients should not necessarily be required to participate in a clinical trial in order to have their DNA analyzed for healthcare risks. Commercial ventures like 23andme are gearing-up in a free-market way to democratize the availability of these tests, and the knowledge associated with them.

Some geneticists argue that the emerging commercial genetic ventures are irresponsible and dangerous-- that they are releasing information that's "unsafe" to patients because the knowledge is not fully vetted and patients won't be informed enough to make informed decisions. Every time I hear someone espouse something akin to the general motive of controlling information to protect the consumer from their own ignorance, I can't help but reflect on the same motives for repressing knowledge, literacy, writing, language translation, etc. that come and go through human history, yet always are eventually doomed to failure. Rather than repressing knowledge, we should do everything we can to release the knowledge we have now as quickly as possible, and thus achieve greater knowledge, sooner.

No comments:

Nuclear and Healthcare Decision Making

Nuclear warfare operations was my data-driven decision making environment before the healthcare phase of my career. It was all about recogni...