Gerald Higgins, PhD., MBA, MS, is not your typical CIO — that’s because Higgins is a chief innovation officer. Getting paid not to carry out EMR implementations or persuade physicians to embrace CPOE, Higgins is remunerated to “see what’s happening one or two hills over the horizon.” With a dizzying array of jobs and accomplishments already behind him, Higgins has a non-traditional perspective for chief information officers, and a hearty dose of sympathy for the pressures they currently face. Recently, healthsystemCIO.com editor Anthony Guerra talked with Higgins about what makes a good innovation officer, meaningful use, and much more.
BOLD STATEMENTS
A good analogy might be the Defense Advanced Research Projects Agency (DARPA), so think of my job as a mini-DARPA — they look five hills over, and I look one or two hills over, in terms of what is approaching the hospital network.
I think it’s important that I tell hospital clinicians they must understand personal genomics and pharmacogenomics, because they are going to be required to use it in their practice. They haven’t a clue about it right now, and the typical 2-hour CME course is not going to cut it.
If you look at the meaningful use objectives that have been suggested for comment, every single one of these things Azyxxi could do eight years ago. The ones that it can’t are kind of ridiculous.
GUERRA: When we connected on LinkedIn, the first thing I noticed was your interesting title. You’re not the usual CIO, you’re a chief innovation officer. Can you tell me more about that?
HIGGINS: I work in a hospital system, MedStar Health, that has about 40,000 employees. We have Georgetown University Hospital and Washington Hospital Center and eight other hospitals. I am the first person to have this kind of job in this hospital system, and what I think impressed my boss — Mark Smith, MD, FACEP, who was head of all the emergency departments — is that he found somebody who thought in an intersectional way. I would see a lot of similarities between domains, whereas others, especially if you’re a medical specialist or software engineer, might be very challenged to do that.
The term chief innovation officer is subject to many different interpretations, but it’s usually the person who has the ear of the directors and the CEO, who is trying to bring change, whether in technology or in the structure of the organization. Typically these jobs go to people who, throughout their careers, have not stayed at jobs more than three or four years. They tend to get easily bored and tend to be people who will remain as the CIO at a place for only three or four years and then leave. If you are supposed to be a change agent of some kind, bringing in a new domain or trying to restructure a hospital network or give advice about it, you probably need someone who is thinking a lot about a lot of different things and tends to get bored easily.
If you look at what I’ve done, I’ve been more of an entrepreneurial person, a senior manager more than in project management or even CIO-type positions (chief information officer). My job at Medstar is to bring more new ideas in about technology and not as much about how to structure the organization. I’m not here to make things more efficient or keep us up to speed on HITECH or ARRA. A good analogy might be the Defense Advanced Research Projects Agency (DARPA), so think of my job as a mini-DARPA — they look five hills over, and I look one or two hills over, in terms of what is approaching the hospital network.
My focus at the moment is to intervene between the bioinformatics community — who are typically folks dealing with the genome — and the clinical practice environment. Those two worlds have not come together, and so the whole notion of translational medicine (being funded by HHS and NIH), bridging scientific discovery and bioinformatics to foster implementation in clinical practice, has not happened because there aren’t enough people who can span those two domains.
The reality is that unless a hospital or clinician is involved in oncology or obstetrics, they don’t know much about the new genomics, and right now I’m focused on trying to educate the hospital clinician who only wants to know what is relevant to his or her practice and patients.
I have also spent much time looking at our EHR and how that could be improved. MedStar, as you know, developed Azyxxi to reduce the ED waiting times. They hired a bunch of residents who followed ER docs around and noted that 50 percent of their time ( in 1995) was spent just gathering information about the patients they had, finding reports and getting faxes, and that’s why it was built. It was bought in 2006 by Microsoft along with 60 of our developers. I came in after that.
We currently have a large group that has a CIO and does clinical informatics. They are segregated in a very high security building at Washington Hospital Center and they manage Azyxxi/Amalga, but other hospitals that we have in our network use other products, like GE Centricity or Cerner, depending on what they are doing, some even use specialty-specific EHRs.
GUERRA: Do you think chief innovation officers have to come from other industries?
HIGGINS: Not necessarily. My whole background is in healthcare, whether it be medical research or clinical practice.
GUERRA: So what makes you a good chief innovation officer, since you have spent your whole career in healthcare?
HIGGINS: Well, because what I have done in healthcare is gone from being a professor of anatomy and neurobiology, to being chief of molecular neurobiology at NIH, to being vice president of Roche, to being a consultant for the Office of the Army Surgeon General.
GUERRA: So to be a chief innovation officer, you can have spent your career in healthcare, but you need to have moved around a bit?
HIGGINS: Yes. I’ve moved around to all these different sectors, and then I’ve moved more into bioinformatics. I have a strong molecular biology background. I think it’s important that I tell hospital clinicians they must understand personal genomics and pharmacogenomics, because they are going to be required to use it in their practice. They haven’t a clue about it right now, and the typical 2-hour CME course is not going to cut it.
Right now I’m developing a genomic clinical decision support system around Warafin, also known as Coumadin, it’s actually rat poison, used for patients that have problems with bleeding coagulation. The problem is there is such a disparity in genetic variability between patients that are given this drug and how they metabolize it in their blood, you can actually give two people the same dose, and one will die while the other will have no response at all. So the FDA is going to require physicians to know when to prescribe it and what dose. You can use clinical decision support systems to help determine that but, on the other hand, there also has to be some underpinning knowledge of what the new genomics is bringing.
GUERRA: I’ve read that current EMRs are not set up to handle genomic information.
HIGGINS: That’s the view of a lot of people, but I think they are wrong because EMRs deal with all kinds of data already from labs, blood tests, radiologic images, EKGs, physicians charts, all this stuff, they handle all different kinds of information. I think the EMR of the future will be an XML-based application supporting any kind of data, tied to a cloud computing environment where there are no limitations for how much you can store and use.
GUERRA: You’ve been there for two years. Tell me about your relationship with Medstar CIO Catherine Szenczy.
HIGGINS: We have a good relationship, but I have more interaction with my boss, Mark Smith, who is co-creator of Azyxxi and a director of MedStar. It’s probably because I am paid to think and get HHS money when I can. Because we are in DC, I am on various committees and work with a lot of federal agencies. A lot of that has to do with predicting what is coming down the pike rather than direct implementation of the EHR.
If you look at the meaningful use objectives that have been suggested for comment, every single one of these things Azyxxi could do eight years ago. The ones that it can’t are kind of ridiculous.
GUERRA: Tell me what you mean.
HIGGINS: Well, let me focus on medication, because that’s what I’ve been working on most recently. Yes, it’s a good idea to be able to assemble a unified list of what a person in the hospital is getting, in terms of drugs. There was a recent study that showed the average chronic care patient is taking at least 12 medications at the same time. To my mind, as it says here, if the person is taking more than one drug they should be listed in one place in the EHR according to the recommended ONC “meaningful use” criteria, but the reality is that the biology of this is crazy, we have no idea how these drugs interact with each other. It doesn’t seem that will provide a lot of insight into the person’s pathological state or physiology. They are emphasizing ICD-9 and SNOMED. I think it should be on ICD-11 right now, which is where things are going in places like Europe.
Share Your Thoughts
You must be logged in to post a comment.