Last month, a piece in NEJM Catalyst from healthcare researcher Julia Adler-Milstein, Ph.D., caught the attention of many health IT contingents. In the commentary, titled, “Moving Past the EHR Interoperability Blame Game,” Adler-Milstein, an associate professor at the University of Michigan’s School of Information with a joint appointment in the School of Public Health (Health Management and Policy), argued that while most people tend to focus on who’s at fault for the lack of healthcare interoperability progress, it’s far more useful to learn from missed opportunities and create a path forward that all sides can agree to.
Perhaps the most important takeaway from Adler-Milstein’s piece was her last line in which she wrote, “Once the business case for interoperability unambiguously outweighs the business case against it, both vendors and providers can pursue it without undermining their best interests.” This is where much of the discourse in health IT circles lies—both vendors and providers want to “do right” for the sake of better healthcare, but there also needs to be a business incentive for them to go down that path. To discuss her take further, Adler-Milstein recently spoke with Healthcare Informatics Managing Editor Rajiv Leventhal, digging deeper into one of health IT’s core issues. Below are excerpts from that interview.
You have done a heck of a lot of work when it comes to health IT research over the years. When you look at where the industry started, and where it is today, how would you characterize the rate of progress?
I think there has been a lot of progress and a lot of it has to do with where expectations were set, relative to reality. A lot of the discussion has been around HITECH [the Health Information Technology for Economic and Clinical Health Act], the money spent [by the government], and the benefits of EHRs [electronic health records] as they would [soon become] interoperable. The discussion was almost that they would be basically be one and the same—you put in EHRs and they are interoperable, and they can share data, but anyone who has seen interoperability up close knows that couldn’t be farther from the truth. It was clear from the start that these would be different problems to solve. For EHR adoption broadly, there are a narrow and a specific set of barriers that needed to be addressed, and those were essentially money and making sure that the systems were of good quality. But on the other hand, the barriers to interoperability are broad, diverse, and all difficult. So you are solving different problems and we didn’t put enough muscle behind the harder problems. That being said, we have made progress on interoperability; there is a lot more data exchange going on today than in 2009. So it is frustration for people who work on this every day to be told that no progress is being made, but can we be further along today had we done things differently? I think the answer is yes.
Julia Adler-Milstein, Ph.D.
You mentioned in your piece that most observers blame vendors for the lack of interoperability, but you seem to have come to their defense more than others have. Is that fair to say?
I think we have to distinguish what feels ethically right from what a rational business decision is. It clearly feels that vendors should be doing more, and we all wish they should have done more. They talk about their mission being to improve healthcare, but if that is your mission, you should have thought about interoperability differently. But at the end of the day, these are for-profit companies, and that’s how they’re behaving. If we were the heads and CEOs of these companies, would we have done anything differently? So I don’t think it’s fair to ask these vendors to act like for-profit companies, but also to act like mission-driven not-for-profit companies. You can’t be both. I wish they had made different decisions in the past, but it’s not fair to point fingers at them for these decisions they have made.
For me, the key takeaway of your piece was that the business incentives for vendors and providers need to change. That’s something that I think most people would agree with. But the vital question is, how can this happen?
I think what I tried to set up in the piece is that it’s a cost-benefit equation. Right now it’s the worst of both worlds where the benefits of being interoperable are perhaps ambiguous at best, and the costs are very high. So if you are going to solve that problem, you can do it either by making the benefits less ambiguous, or by bringing the costs down, or ideally by doing both.
If you look at the Epic example, they have made interoperability much less costly and complex. So it’s not as hard for a provider say, “Well I’m not sure if this is good or bad but if you are making it cheap and easy for me, I will take the risk. But I am not willing to take that risk if what you are asking me to do is costly and expensive.” I do think policy efforts have tried to emphasize making interoperability less costly and complex, and that has been helpful. But if we want to make progress we have to really work on the benefits side and we have to make it financially beneficial to engage in interoperability.
There are a few ways to do this. One way is paying for interoperability, in the same way that I think of as meaningful use criteria, so we will literally pay you to exchange data interoperably. I think we can do that, but it can feel less valuable that way. So I am more in favor of paying for outcomes that either can only be achieved or can be much easier achieved if you are exchanging information interoperably. If providers had really strong incentives tied to things like less duplicative testing, so if they knew they wouldn’t get paid for avoidable duplicative tests or avoidable ED visits, they would certainly work hard to avoid that.
The other option is for providers to report quality measures with dominators that come from the community. So I will report immunization data, but I am not just on the hook if hospital A delivers that data; I am also on the hook for reporting if the patient got that immunization, no matter where he or she got it. This would be another way to say that we will create incentives that are tied to things that are so hard to do without interoperability. And I am not saying that’s easy, but it might be the only way to really make this benefit much less ambiguous.
When it comes to data blocking, I’ve talked to experts who are right in the middle of everything, and they have given me completely opposite viewpoints. What are you seeing and hearing about this?
I think information blocking is a very misunderstood topic, and if you look at the definition of it, there are a lot of words that are heavily subject to interpretation. For example, who defines “what is reasonable?” A vendor might think something is perfectly reasonable from their perspective and a policymaker might think that same thing is unreasonable. If you gave those people the same exact scenario, and said this is what happens, and asked if it’s information blocking, they wouldn’t be far apart. But people are understanding the concepts differently, so they are looking at the same behavior and labeling it differently. I have seen people interpret information blocking as charging you for interoperability, but that is not the definition if it, and no vendor will define information blocking as anytime you charge for interoperability. The people who are working to make interoperability happen were nervous when these concepts came out, because they thought that all of their efforts that were resulting in slow progress were going to be labeled as information blocking when in fact they are doing everything they can to make information move, but are facing very hard challenges. They were fearful that it would be a witch hunt.
How bullish are you on the progress of health IT going forward? Do you think the shakeup in Washington, D.C. could affect things at all?
I wouldn’t put money down on this, but if you look at [HHS Secretary] Tom Price, he is someone who is a frontline clinician. So what are the pain points of those frontline clinicians? Interoperability is one of them. I think this issue has the potential to stay on the radar, since you have someone who is advocating for the average doctor. Whether there is enough understanding in HHS and Congress as to what’s really going on with interoperability, how it’s developed, and pursuing new policies to address it, well, that’s another question. Doctors in that frontline position would say that it’s the vendors’ fault, and they may not appreciate all the complex dynamics that are in play. So while the issue may get more attention, whether we will be able to make progress and put in new solutions or not is what I am more concerned about. It’s a challenging topic and there isn’t really a deep understanding as to how complicated healthcare is. There tends to be a desire to simplify things, but that’s not always the most productive approach to solving issues.