Wayne Kubick joined the standards development organization HL7 in February 2016 as chief technology officer. He said he saw opportunities to improve processes and technologies by capitalizing on all the excitement around the FHIR (Fast Healthcare Interoperability Resources) standard, which he said “captivated me quite thoroughly.”
Kubick came to HL7 after a career in the biopharma industry, including working for the Clinical Data Interchange Standards Consortium (CDISC), a data standards organization focused on clinical research studies. Healthcare Informatics recently spoke to him about his new role and developments around FHIR.
1. So how did the move to HL7 happen? Was there a learning curve?
Kubick: HL7 CEO Chuck Jaffe had served on the CDISC board with me. He called when he heard I had left CDISC. I did have some concerns about taking the position here. The world of HL7 is extremely complex and different from what I know, but the process of developing standards has a lot of overlap. If you look a little more deeply, it is the same kind of data. We are collecting things about patient data and using it to make judgments about better treatments. I have loved it at HL7. I knew enough about the basics of what people do, but I get to learn everyday about things I didn’t know so much about. It is really stimulating and the work is really important.
2. What does the CTO job at HL7 involve?
Kubick: My principal role within the organization is centered on trying to manage the processes and technologies associated with the core function of standards development and group collaboration. I am trying to articulate a common technical vision of where HL7 wants to be in the next three to five years and converge all the work they have done in the last 30 years toward more of a unified vision. That would incorporate a migration path from the old V2 world, the C-CDA work, which has become a very significant part of meaningful use in the present, and the future, where FHIR is taking us, and there is an amazing array of opportunities.
We want to capitalize on what people have done in the past, but also recognize that the way we develop standards has to change. We have to develop more of an agile framework and be able to move more quickly as the world changes, because it is moving at an incredible pace. The old way of standards development — where you spent three years and published a PDF document and people spent five years trying to understand what it means — those days are long gone. FHIR encapsulates that nicely by moving very quickly and adapting and representing the real world as people are using it.
3. Could you talk about developments and a timeline for FHIR? I read that there is going to be a “normative edition” of FHIR next year. Is that right and what would that include?
Kubick: It has been recognized that as the specification adapts and evolves, there needs to be a point where we establish a more solid foundation. The plans for next year are not exactly a normative edition. It will be a new, balloted release that will have normative content. Graham Grieve was very pleased that in the most recent ballot for trial use version 3, which just underwent balloting, there weren’t a lot of comments on the core or infrastructure underpinnings of the FHIR specification. Those things should be solidified in the normative release. Also some of the most commonly used resources such as “patient,” “observation” and “diagnostic report” will also be brought to the level 5 maturity that the FHIR model has. But many other things will still be non-normative, evolving and adapting. The vision for FHIR is that over time, more and more of these individual components get rolled over into the normative piece, yet it still maintains its ability to adapt and extend to new use cases. It is going to be an interesting world because that is not the traditional world that HL7 operates in. The important message is that a solid, core foundation will be in place. The reality is that, even without that, people have been putting lots of production applications into use throughout the health industry.
4. At an upcoming meeting you are going to give a talk about FHIR directed at C-suite executives. What do they need to understand about FHIR that they don’t?
Kubick: A whole lot! I think it is one of those things floating around in the background but they may not have a real understanding of what it means. We want to talk about what FHIR means for the future of healthcare from the perspective of many different constituencies. We will have a 10-minute talk from experts in several different segments about what FHIR means for patient care, for clinical research, for the health IT industry, for payers. How does it contribute toward the future of what we want to achieve through interoperability? It will be pretty high level about what an enabling, transformative technology can do to a business operating in each of these spheres, and get people more engaged and interested in why they should pay attention and get involved, and what it can mean to the strategic direction of their own businesses. We also have a need to get more engagement to ensure the sustainability of HL7.
5. You also have a presentation coming up about how FHIR might be related to biopharma topics, such as prioritizing use cases for repurposing EHR data for use in research and drug safety. Is that bringing you full circle back to your interests in data standards in that realm?
Kubick: Yes. Once you start with the assumption of using data acquired at the point of care to learn more about where business opportunities lie, as well as potential risks, you can start looking at all kinds of things. The clinical research business in the pharmaceutical industry is pretty backwards technologically. It takes organizations a long time to change. There is strong regulatory fear and there is a lot of effort involved in even making a small technological upgrade in a big pharmaceutical company. A new version of a system can take a year to implement. FHIR makes it possible to tap directly into EHR systems and reach out directly to patients, as well as to provide a capability for putting data back into EHRs. You have the possibility of tapping into data, not just for randomized clinical trials, but also for an analytical view of what is really going on in terms of patient populations for these targets being explored for various drug products. One early potential use case involves being able to look at what types of patients are in an EHR to see who would be eligible for a clinical trial and whether they would be interested in participating directly. Another is being able to pull in EHR data to pre-populate a case report form. There are so many opportunities to use healthcare data more effectively. It’s a great time to be at HL7.
