While most of the $20 billion allocated to health IT in the American Recovery and Reinvestment Act was designed to help providers adopt electronic health record systems, a big piece of the pie went to catalyst organizations: research agencies, training centers and planning teams whose work will magnify the effect of the incentive plan.
The recently announced Strategic Health IT Advanced Research Projects (SHARP) awards, regional extension center contracts and the state health information exchange grants are all part of this strategic layer of funding.
None of these funds are likely to have a greater impact than the $8 billion that went to the National Institutes of Health to back advanced informatics research as well as more tactical health IT projects. In deciding how to divvy up the funds, NIH says it wants to balance long-term research against more applied projects. Its ultimate goal: to enrich the underlying data that EHRs will access, crunch, store and route to improve overall quality of U.S. healthcare.
"We shouldn't say we don't care how people are struggling today, that we won't do incremental improvements or come up with ways to help them," said Betsy Humphreys, deputy director of the National Library of Medicine.
Instead, NIH has a different perspective, Humphreys said. It's best described in a 2009 report by the National Research Council, which laid out a vision of computer- assisted clinical decision support"fusing biological data such as genomics with information from EMR systems"as a long-term goal for the nationwide health information network.
"I agree with the main conclusion of the report, which is that very clever incremental improvements will not get us where we really want to go," Humphreys said. "We want to continue to find people who are thinking about the next level of problems as well as those who are going to come up with clever ways of making things work better next year."
DNA-EMR fusion
Dr. Dan Masys, chairman of biomedical informatics at Vanderbilt University Medical Center, which received a $6.4 million ARRA-funded "grand challenge" grant from NIH, said the center's work seeks both advanced research and applied health IT pay-offs.
An umbrella term for the fusion of these disciplines is "translational medicine," he said, where discoveries in the lab quickly inform clinical decisions, and clinical data is fed back into databases to improve basic research.
The concept is embodied in Vanderbilt's work building a "bio bank" of de-identified DNA samples from nearly every patient the center treats. The bank, called BioVU, had 78,000 samples as of March, and increases at the rate of 500 per week. With it, DNA data can be combined with de-identified EMR data to help researchers and clinicians devise individualized care strategies.
In their project, called Vanderbilt Electronic Systems for Pharmacogenomic Assessment (VESPA), Masys and his colleagues will try to determine whether responses to certain drugs as detailed in medical records could have been predicted by variations in corresponding DNA samples.
Masys says he and his colleagues will also research whether specific points in the genome, if changed, would signal likelihood of a good or bad outcome to a specific type and dose of medication.
"We would like, in our vision of personalized medicine, to get that [data] on everybody who walks in the door, so it's already in the record before a doctor even sees the patient," he says. "Then a computerized decision support rule can say, "˜this person has this DNA variation to take into account.'"
Masys says the model will widen the reach of health research by including demographic groups that traditionally have shied away from clinical trials. "This allows us to research almost any health condition, no matter which population group might have it, and also at relatively low incremental cost," he said. "To add a sample into a bio bank is tiny compared to the traditional cohort method of populating data banks."
Even so, incorporating the research into practical health IT systems will not be easy. Getting clinicians to take advantage of new data could be another hurdle. Tony Gorry, co-director of Rice University's W.M. Keck Center for Interdisciplinary Bioscience Training, who is overseeing a $1 million ARRA grant, says the bioscience must be carefully integrated into clinical processes.
"It is not sufficient just to be able to tell you your genes are expressed in a certain way," Gorry said. "Someone is going to have to implement that in medical care, and that is a big hurdle. All you have to do is look at evidence-based medicine, and see how much of it is not practiced. That's going to take a while to get absorbed and it's going to take bright and capable people.
"In this particular domain, it's too bad the NIH doesn't have an anthropologic, sociologic function," Gorry added. "The thing preventing the transformation of the healthcare system has to do with organizational change. President Obama says he wants to transform healthcare, but there is no place where the transformers are being trained."
New research loops
However, one scholar of NIH funding patterns sees promise in the new emphasis on having front-line providers in the research loop. Dr. Sean Lucan, lead author of a report on family medicine and NIH grant funding, says the infusion of stimulus money for IT-enabled networks could reverse a trend which saw the level of interest of medical students in entering family practice at "near welldocumented historic lows."
"The IT resources now available are connecting the world," Lucan said. "Particularly, rural family physicians and primary care providers, who are technologically and functionally isolated, if not from the rest of the healthcare world, from the research world. This pretty much breaks down the important barriers that are out there."
In the near term, one might ask whether or not NIH funding of translational research at institutions which already have advanced EMRs is funding an un-level playing field. However, Masys said the HITECH incentives and penalties should work in concert with the bio bank/EMR data either already in existence or being built right now.
"Meaningful use has a strong flavor of interoperability, including the ability to exchange data, so that is our best hope of driving it economically," he said Masys. "The technologies all exist. It's really about incentivization, and social structure within the organization. The pieces already exist. It clearly can be done; we just need to scale it up."
