With the healthcare landscape changing so rapidly along so many dimensions, it should surprise no one that the landscape of the worlds of radiology, diagnostic imaging, and imaging informatics, should also be changing very rapidly, across the policy, operational, clinical, and information technology fronts. And it is. As HCI Editor-in-Chief Mark Hagland reported in his November/December cover story on imaging informatics, healthcare leaders and industry experts are urging hospital and medical group leaders and physicians in practice to develop strategies to manage the changes sweeping the industry these days. Among those he interviewed was Joe Marion, principal of the Waukesha, Wis.-based Healthcare Integration Strategies, and one of the leading industry experts in the imaging and imaging informatics areas. Below are excerpts from Hagland’s interview this autumn with Marion, published here just days before the start of the annual RSNA Conference, to be held Nov. 29-Dec. 4 at Chicago’s vast McCormick Place Convention Center.
Let’s walk through the current landscape around diagnostic imaging and informatics together. What do healthcare IT leaders in particular need to understand about how the rapidly shifting policy landscape is affecting how they should think about technology right now?
It’s a forest-versus-trees situation. Healthcare IT leaders have got to think about what they’ve got now [with regard to technology], and how they’ll tie into ACO [accountable care organizations], bundled-payment contracts, population health, everything, and what they’ll bring to those arrangements. And with hospitals and larger physician groups buying up medical practices and such, that is affecting everything, too. Large vendors are in this mode now where they’re saying, from a sales perspective, how do we structure, how do we address the client? It’s no longer, I go to Hospital A, Hospital B, Hospital C, and sell them each a CT scanner. Instead, imaging and imaging informatics vendors are working towards signing strategic contracts, wherever possible.
So the large vendors are looking to build longer-term contracts with large integrated health systems that will encompass both modality and imaging informatics technology and services?
Yes, absolutely. And it may even go to the point of onsite equipment support, and even to the point of [contracting around technologists and radiologists.
Certainly, Merge/IBM and other mergers and acquisitions, are significant. I spoke with someone at Merge last week, and a person there says they’ll operate as an independent subsidiary with an influx of people to assist with Watson integration, but otherwise will beat their own drum. But I’m wondering what the benefit of that arrangement is. I’m still wrestling with that idea, because, looking at past acquisitions of the sort, how long will IBM allow the Merge folks to act in an independent way? And Dell acquiring EMC could prove significant, too,” in altering the competitive and contracting landscape around vendors in imaging informatics.
There are two ways to look at all this activity. On the one hand, the fact is that there’s been an overabundance of vendors. Somebody once counted at least 100 PACS vendors out there—and that’s still true, there are all kinds of little players that can offer you a PACS solution. But people are looking at it from a larger perspective, and the whole emphasis on the enterprise is starting to spill over. So that’s one factor. The other is the sustainability of people making the investment. Can they afford to make multiple investments? As well as the “decomposition” of PACS [picture archiving and communications systems]—the Visage people use that term.
And what it means is, let’s say I had been using Vendor A as my PACS vendor. Years ago, if I needed an upgrade, most people would have replaced the entire system. But now, people are taking new pieces of functionality, and adding them in. And in fact, if you go to a vendor-neutral archive architecture, and a universal viewer on an enterprise level, what’s left for PACS to do? The acquisition of images, workstation display and workflow. And so if you see it that way, you might only want certain new functionalities—and most importantly, you may no longer want to purchase an entire dedicated radiology PACS. And given that, how do these little guys survive in the market, when people are trying to take a bigger-picture view?
How might what happens in Stage 3 of meaningful use intersect with imaging informatics?
I’m not exactly sure yet, but some of the premises and measurements within Stage 3 will have more of an impact on patient access, and I think that the issue there is that if patients want to see their images, they don’t need DICOM to do that. So there’s this philosophy that for some purposes, you don’t need DICOM. If you were at RSNA, I could pull out my cell phone and pull down from the Microsoft One Drive cloud and show you my CT images from this summer; they’re just jpg images. And you don’t need DICOM-level viewing for those purposes. And that’s the kind of thing it’ll be. I will want to show someone an ultrasound image from a pregnancy ultrasound, for example. Or my daughter breaks her bone and I want to be able to show people the fracture, for example. But it’s not going to be on paper, and it doesn’t have to be DICOM. So if imaging is going to play a role in the context of Stage 3, it will be at that level, not at the diagnostic DICOM level. And all the ability is there today to do it, it’s just a question of the accessibility of that,
So the meaningful use mandate around giving patients access to their data will shift focus to giving patients access to their images?
Yes, and, per Stage 3, part of the challenge or interest will be to find meaningful ways to deliver imaging information, so there’s some question around an API structure versus integration into an EMR; there’s the whole question of, it’s more than just the patient having the access to it, it’s also going to be around sharing images with other clinicians. So how will I do that?
And some of that relates to HIE (health information exchange)?
Yes. I think there’s huge potential for that, and the infrastructure is probably there to add images to that. I think that XDS, cross-document sharing, will be huge in that context. And it’s more than just radiology and cardiology images, think of dermatology. And I’m totally frustrated with my provider hospital because they haven’t interfaced devices to put a PDF directly into Epic, for ophthalmological images. These laser scanners produce a PDF file, and they have a color printer on the table and stuff them into my folder; they eventually get scanned into Epic. Well, they could do a direct import of PDF files into Epic, if Epic will support it. Same thing for pulmonology. I had exams done in that area, and the output was a PDF. Again, with my health system, I can access all of my lab results, and about 3-4 weeks after the fact, I can access the radiology result, because they’re literally printing those out and scanning them in, because they can’t get them into Epic. And they have no idea when they’re going to do images. When they found this thing on my in January, the tumor on my thymus gland.
I had a situation this summer where I need a referral to a subspecialist, after a diagnostic imaging procedure. But it was incredibly frustrating, because the diagnostic images were not easily accessible to anyone; as an informed consumer, I was able to get CDs that I could share with the appropriate subspecialist, but I shouldn’t have had to intervene; those images should have been appropriately available to that subspecialist in a different specialty from the other specialists I had seen in that situation.
And the other important aspect is the time factor. How much time even today is being lost or wasted? I was able to directly show images to a specialist. But that time factor element will be improved by this whole accessibility.
In January 2017, referring physicians will need to use clinical decision support in order to order imaging. Will that change things?
[Editor’s Note: This interview took place before senior officials at the Centers for Medicare & Medicaid Services had announced that they were delaying this mandate, in early November, without providing a timeframe for its implementation. HCI will update its readers on developments in that area, as new information becomes available.]
Oh, yes, I think so. There are still a lot of situations where the referring physician leaps to specific imaging orders. But the radiologist may say, we don’t need to jump to an MR right away, but we could order a CT instead. And one vendor executive shared a study with me showing that there’s a cost savings in the context of more carefully or judiciously choosing specific types of studies. That’s the front-end side of it. On the back-end side, there’s a huge potential to be had in the communication between the referring physician and the diagnostician, in terms of interacting around the collaborative aspect of this. How many referring physicians actually interact with radiologists in terms of really consulting based on images and studies?
What should CIOs and CMIOs be thinking about and planning for, in terms of the emerging new world?
In the consolidation you’re seeing in terms of going from a company like Accuo, a big VNA vendor, acquired by Lexmark’s subdivision called Perceptive Software, that acquired them, and Perceptive has different capabilities across different industries, and just before HIMSS, they concluded that the Lexmark name has a stronger brand, so it’s no longer Perceptive, it’s now Lexmark. But the key is if you look at that, and look at Highland Software—they have relationships with both Merge Fujifilm/Teramedica, and all of them see the potential in the integration or merging of documents and images, towards this what I refer to as clinical content management.
So from a CIO perspective, imaging by itself may not get the attention; but if you start to look at some of these situations, you’re talking about managing both documents and images. An EMR may have content in itself but also be pulling other types of content in, but why do I support two independent paths? Per HIE, one of the things that’s going to make that more attractive to the CIO is that if I can support one infrastructure with the EMR, that saves a whole bunch of interfaces and a whole bunch of complexity.
So CIOs and CMIOs should be investing in the broadest capabilities, no longer focusing on individual systems, right?
Yes. It’s a forest-versus-trees situation. They’ve got to think about what they’ve got now and how they’ll tie into ACOs, bundled-payment contracts, population health, everything, and what they’ll bring to those arrangements. And with hospitals and larger physician groups buying up medical practices and such, that is affecting everything, too. And a large vendor is in this mode where they’re saying, from a sales perspective, how do we structure, how do we address the client? It’s no longer, I go to Hospital A, Hospital B, Hospital C, and sell them each a CT scanner? It’s going to be the strategic contract. And what Marin General has negotiated with Philips—don’t use names—how much do I pay you for imaging services. And it may even go to the point of onsite equipment support, and even to the point of technologists and radiologists.
Even remote-read radiology services are consolidating these days. Do you have any thoughts on that trend?
Yes, and some of what’s going on now is advancements in technology. So several organizations, including IntelliRad—they’ve got a couple of huge US radiologist practices out selling their infrastructure to other radiology practices. So there’s a hospital in the Northeast contemplating replacing their existing PACS. And so there’s an outfit called Jefferson Radiology in Connecticut. So they’re contemplating outsourcing actual radiology practice. And I think the radiology groups in some of these hospitals are wondering, is that a better solution for me than just buying another PACS system? Sourcing the PACS technology from a group like that, maybe? Center for Diagnostic Imaging, CDI, in Minneapolis, a radiologist at Abbott, Ken Hightoff, originally, it was a couple of centers in Minneapolis, but he expanded to Florida and here in Milwaukee, and here, he formed a partnership with some of the radiologists in the Medical College of Wisconsin, and built essentially a network of expertise. And the claim to fame he had in Minneapolis was that he was an expert in spine; but he also had a TMJ expert; he had a whole slew of people spread out all over, and because he had a network that allowed him to send the images to an expert subspecialist, he was getting that kind of clinical referral. And the technology is continuing to progress to where no matter where the study is at, you can access it and read it. So it’s efficiency and expertise and all those factors that you could bring to some of those small facilities that don’t have any ability to compete on that basis.
For radiologists, should all these changes be both exciting and scary at the same time? The whole idea of locus seems to be changing now, when it comes to what radiology practice is.
Yes, even here in Milwaukee. There’s a group called SmartChoice MRI. They advertise low-cost MRs. But they also reference their access to Cleveland Clinic radiologists. And even Aurora Health Care, what they’ve done, they have literally a doc-in-the-box in Brookfield, where the radiologists read al day long for patients across the metro area. So in one of my fluid-filling episodes, I went to a local doc off 94, and the doctor sees me and sends me down for a digital chest x-ray, and he’s sitting in Brookfield reading images from all over. He’s got referrals from all over.