Welcome to Healthcare PittsStop. Today I have with me a very good friend, Dr. Eliot Siegel. Dr. Siegel is a radiologist known to many across the country and across the world. He was the first to have a filmless department in the United States. I was hoping we could get his perspectives on why physicians still don’t have film or images when they need to see patients. Nice to talk to you.
It’s great to talk to you, as always, Alan.
I remember being jealous back in the 90s because you produced so much of the literature on a filmless department. Can you tell me how you were able to get a filmless department?
I was finishing my residency at the University of Maryland. I had a few offers to work for a number of places, including the University of Maryland. And one of the things they offered me right off the bat after my residency was to be in charge of radiology for Baltimore VA Hospital, which was associated with the university. As it turned out, they also gave me the opportunity through the VA to do pretty much what I wanted to in terms of planning a brand new hospital that was going to open just a few years later. Having a strong interest in computers and computer technology, and having been to some conferences where they talked about the future of radiology, I made a deal with the University of Maryland and the VA, saying that if they would allow me to become the world’s first filmless radiology department, I would fix up the job. I really didn’t know any better back then what a daunting process it might be. But the timing was just right. I had an opportunity to do a little bit of design and work with the vendors for PACS (picture archiving and communication system), and it got built into the budget as a line item for the entire hospital. So we were able to create the first filmless radiology department. We purposefully didn’t build alternators or view boxes into the new hospital because it significantly accelerated our vendors’ desire and motivation to make us filmless. We kept saying, we understand that this is brand new; we understand that we haven’t done this before and it might be delayed, but we have to open the hospital, we have to have a radiology department, and the only way to do it at this point is with our digital systems. It was super fun and we had to solve lots of unique challenges that seem very quaint nowadays.
That sounds fortuitous, Eliot, but you were the right guy to lead the ship. What year did that hospital open?
We bought the PACS in 1991, and the hospital opened in 1993. Back then there were some really interesting questions. There was a legal requirement to keep film for 5 or 6 years, but since we weren’t making film, was it possible to go filmless as long as we could make film if somebody needed it. Another question was, the resolution on a monitor is significantly less than the resolution you can get on a film, and it hadn’t been demonstrated or proven yet that you could do primary interpretation directly from monitors for conventional radiographs. Lots of medical, legal, social, adoption and other issues that nowadays we take for granted, but back then it was really completely brand-new ground. It made it really exciting, and when the hospital first opened we had lots of press coverage from CNN and many other media that picked up on us being the world’s first filmless hospital. We ended up having visitors from all over the world who came to see the hospital, and since all of the orthopedic surgeons and vascular surgeons and neurosurgeons and others found themselves on TV saying how wonderful and exciting it was, it accelerated their acceptance of the system because all I had to do was play back their last interview on TV to make sure that they were fully on board with the PACS system.
It sounds like you figured out a secret sauce there by getting them to commit in public, in the media. So your hospital opened almost 25 years ago and yet we still have so many patients who show up to their doctor’s office for their appointment, and their doctor, their oncologist, their surgeon does not have access to their images. Now, I know at the Baltimore VA you would, of course, but how about the rest of the US? What do you think has slowed that? Why doesn’t the routine physician have those images when they need to make a decision?
I think that lack of access to the images is really just a subset of the overall lack of access to healthcare information. When I travel to Australia or Europe or Asia, if I put my ATM card in the machine, or my credit card, it reads it and dollars are made available when I need them through the banking system. But I don’t have anything close to that that’s available for healthcare. If I were seen in those other countries or if I’m seen at a hospital across the street, there isn’t that same level of access. And certainly imaging is no exception. So, it used to be we’d have to take film and we’d have to create a copy of the film onto another film. After that, the next level up was being able to actually write the images onto a CD or a DVD. But the problem is that those are pretty insecure, easily lost. Also they often don’t have the patient’s report on them associated with the images. You would think in today’s modern era of the Internet it would be super easy to essentially just pull or push images from one place to another. But unfortunately, because of concerns related to HIPAA and because of the relatively insecure nature of the standard that we use in medical imaging, which is called DICOM, it turns out that it’s a lot more daunting process than one might guess intuitively.
There have been a few solutions that have been proposed. One solution has been for patients to have their own patient health record. So if you or I were a patient, we could sign up for an image-enabled patient health record and then we could ask the facility to send images to that patient health record. The problem is that most facilities don’t have software that allows them to be able to do that easily—or not at all. So the Radiological Society of North America came up with a project that I was able to assist with and get a pilot study, where there was software written by the RSNA and by a number of other institutions, including our institution at the University of Maryland, where we have the capability of being able to use the RSNA’s software on a server that is referred to as an edge server because it sits between the Internet and the PACS system. And we have the capability to push images from the PACS out to that edge server and to a patient’s personal health record. The next step would of course be to take facilities that have the same RSNA imaging software and allow those facilities to be able to push images directly from one to the other as long as we had patient permission to do that. The next step after that would be for commercial entities, which are already starting to spring up, to provide the same type of software that the RSNA has created for its pilot or prototype study.
We all know that reordering imaging often drives up the costs of healthcare with little value for the patient overall. Do you think it’s mainly legislative regulatory issues or do you think it’s the business model that has prevented wide-scale adoption of personal health records (PHRs) or any form of aggressive image sharing?
I really think it’s a combination of all of those. Part of it is that there’s motivation for hospital systems to want to keep patients within their own system, and enthusiasm for making it easy to transfer images or electronic medical records really isn’t there. And until hospitals have a way of being able to transfer the electronic medical record easily from one facility to the other, we’re not really going to have images—which are really seen by the hospital as a subset of the electronic medical record—to transfer. Hospitals often make the excuse of HIPAA or the HITECH Act, saying that it’s difficult to do that transfer. But you know, we’re already transferring images. We’re just doing it in a less secure way using film. So absolutely, you’re right. It’s a combination of policy issues, politics; to some extent it’s economics. And there really is not yet a completely agreed-upon standard mechanism to be able to transfer images from one facility to the other. So everybody is winging it, to know how to work out when one has patient permission to be able to do that.
There’s certainly the temptation—and it’s happening on a small scale—for patients to say, instead of every institution being able to connect with every other institution and transfer and share my images, I’m going to be my own broker. In addition to the RSNA’s image share project, we’ve also looked at the possibility of having patients who have their own Dropbox account or Google Docs account, for example, to be able to ask to have their images pushed up there. And then some of the vendors, such as Box, one of the data storage providers, are setting up relatively private and HIPAA-secure mechanisms and they’re also setting up viewing software.
Do you have a solution for patients with a chronic condition to take control of their own imaging today, one that’s relatively easy for them to pursue?
I think the easiest path currently—unfortunately because we’re about 10 or 15 years behind in healthcare in comparison to other industries—is to have a mechanism to take the CD that one has and be able to upload it to a viewer. One viewer that works pretty well is Microsoft HealthVault. MS has the capability to allow you to upload images to HealthVault and have other people be able to view those images. The other possibility is other pieces of software that allow you to view medical DICOM images. One possibility is to take a CD and to be able to essentially zip that CD into a file, and then you can send that file from one place to another. Those are all mechanisms currently, but I know it’s really frustrating how few choices there are, and we’re still in the old-school mechanism of people still continuing to FedEx their CDs or DVDs from one place to another.
Do you have any suggestion for innovators out there who want to fix this problem? Are there a couple of key gaps that you think, hey, if somebody could build a company or solution that did this, it would really solve the problem today?
What we really need is for somebody to use the standards that exist already and to create an entity that applies those standards and makes it easy for vendors to be able to move images from one place to another. I think it would be great to have a sort of neutral broker or neutral vendor to do that. The American College of Radiology has certainly thought about and looked into the pros and cons associated with being able to provide that type of service, and I’m not sure whether they will or not. But I’d love to see somebody like Google Docs or one of the other storage providers step up to the plate with a medical imaging type of solution. I think that probably will be what’s going to work. Right now it’s not a technology issue; it’s an issue of who you trust with your images. Would you trust Microsoft? Would you trust Google? Would you trust a small startup that you’d never heard of? Would you trust the American College of Radiology? And so who would that broker be? In the finance industry there are well-established mechanisms for us to be able to exchange data in a safe, private, secure way. I just don’t think we have done that yet in the medical community. Somebody really needs to step up to the plate and do that.
Do you think that healthcare reform is going to catalyze a lot of this discussion, move it along faster now that healthcare systems are on the hook? Do you have to go at risk for much of healthcare?
To me it seems as though government initiatives are increasingly pointing in the direction of a single payer system. As time goes on, if there is more of an implementation of healthcare throughout the country, it might look a lot like Medicaid, for example, or Medicare. As that happens, I think there are going to be increasing government rewards and incentives to have networks that are all part of that—essentially coalesce and have similar types of software. So if you’re in the VA, your images can be easily transferred anywhere within the VA. The same is true for Kaiser, and I think as time goes on and the government ends up becoming more and more the insurance payer for more and more places, you’re going to see the facility and capability to be able to move images among all of those sites. I think the other thing that’s going to happen is that, as the CNS and government end up increasing the regulatory environment, they will specify that in order to get reimbursed you have to have a mechanism to be able to share images along that exchange. So even though individual hospitals and healthcare systems might not have incentive to send patients or images out of their system, I think there will be regulatory forces associated with healthcare reform that will incentivize that, or penalize those that don’t do that.
Eliot, I want to thank you very much for your time. It’s crazy that almost 25 years after you created the first filmless department, physicians and patients are still struggling with this. Hopefully the situation changes in the near future as some of these things all align.