The promise of genomics and personalized medicine is immense. But right now health information technology is seriously unprepared to capitalize on the medical advancements that could be just around the corner.

"We're really not in near as good a position as we should be to take advantage of the data that's here right now," said Nephi Walton, MD, biomedical informaticist and genetics fellow at Washington University School of Medicine, speaking Wednesday at the HIMSS Big Data and Healthcare Analytics Forum in San Francisco.

There's been a lot of hype about personalized medicine these past few years, and much of it is justified. But the misunderstandings and barriers are still real.

Part of that has to to with the science itself. "Clinical genetics is a relatively new and evolving science," said Walton. "The technology is based on probabilities and varies in accuracy. You can actually send a genetic test to two different labs and you will get different results in some case. It's not 100 percent."

Consequently, certain perceptions about what it can accomplish are out of proportion to current realities: "There is a huge lack of understanding about genetic testing and there's inflated expectations, a lot of misunderstanding about genetic testing," he said.

"This is not an exact science," Walton explained. "'Whole exome' sequencing doesn't cover the entire exome, 'whole genome' sequencing doesn't cover the entire genome. For some diseases you can't just test the blood, you have to test different regions of the body. There's still a lot we don't understand."

But a lot of the hurdles are related to limitations of clinical technology that, with some retooling, could better enable genomic data to make an impact at the point of care.

[Also: HIT unprepared for 'omics' onslaught]

"Our technology to produce data from genetic medicine is far more advanced than our ability to use it in a clinical environment," said Walton. "Our IT infrastructure is grossly inadequate to meet the demands of precision medicine today."
 
Consider the simple fact of how the data is stored.

"We currently store genetic data in a very robust, complicated, standard format that was developed nearby here, in Silicon Valley," he said. "The problem is, it was developed in 1993 and it's called a PDF."

That format isn't particularly suited to support clinical decision support, or to inform prescribing practices.

"We have Cerner and Allscripts and are moving to Epic," said Walton of Washington University School of Medicine. "None of those vendors can actually use genetic data for decision support, or even store it."

Moreover, "they've been slow to respond to the demand for precision medicine," he said. "although I think that people are demanding it enough now that there has been some change."

That's a good thing because "things are moving really fast," said Walton. "Whole genome sequencing is not used that much presently, but it won't be long. We've got to be able to use this data in our EHRs."

[Also: Genomics pose 'daunting' test for EHRs]

Typically, the exome generates about six to eight gigabytes of data, he said. Whole genome sequencing generates about 100-200 gigs. This data is then distilled down by the lab, which creates a list of specific genes of certain significance.

What happens to all the rest of that sequencing data? "It's not totally thrown away, it's still at the lab," said Walton. "But we can not use it in a clinical environment the way it is. This is valuable information. It can show how you'll respond to medication, what you're susceptible to. But essentially we're throwing it away – in large part based on the fact that we don't have a place to put it or an easy way to use it."

He offered the analogy of an MRI: What if they only gave you a report, but you couldn't see the image?

Likewise, the genetic report is very limited. "It only shows me a number of genes. But I see the patient. The lab doesn't see the patient. The ability to see this data in relation to the person you're taking care of is very powerful. We, as clinicians, need to see the data. We need access to all the data so we can make clinical decisions."

There are many other big issues that could hamper wider deployment of genomics in healthcare, said Walton – from legal and ethical concerns (informed consent/information disclosure; how and when to test, and what sort of test) to questions of insurance coverage.

But the more immediate hurdle is related to technology.

"I think precision medicine is great, it's going to change the world," he said. "But we need to change our healthcare system to accommodate it."

Walton added: "When I can get more genetic information from my iPhone than my hospital, something needs to change."

This article is part of our reporting on the Big Data and Healthcare Analytics Forum taking place this week in San Francisco. Other stories include: Atul Butte says 'precision medicine makes doctors nervous | Mayo Clinic physician engineer says healthcare analytics need a system-based approach | Best practices for healthcare data visualization | Penn Medicine chief data scientist: Analytics should begin at the executive level and Data scientists, execs share advice for assembling a big data analytics team.

Twitter: @MikeMiliardHITN
Email the writer: mike.miliard@himssmedia.com
Like Healthcare IT News on Facebook and LinkedIn