PACS Americana: The U.S. radiology IT market is saturated and stable, but change is on the way
The first thing one notices, when looking at HIMSS Analytics' tallies of picture archiving and communication systems (PACS) adoption in hospitals, is how high they are. Unlike CPOE, e-prescribing or even EHRs, the market penetration for PACS systems is pretty robust.
In 2007, some 70 percent of hospitals had at least one radiology PACS modality up and running. As of this year, nearly 91 percent of 5,309 hospitals surveyed can say the same.
"If you've been a [PACS] salesperson, you've had some good years," jokes John Hoyt, executive vice president of HIMSS Analytics.
That said, not all modalities are created equal. While radiology PACS - for CT, ultrasound, MRI, even digital fluoroscopy - are in place in three-quarters of hospitals or more, other specialties have seen slower adoption.
"We make a distinction between radiology PACS and cardiology PACS, and the reason we do that is that there really are different market penetrations," says Hoyt.
When it comes to cardiology, PACS prevalence is below 40 percent.
"Small hospitals" - those with fewer than 100 or 150 beds, say - "are not going to have cardiology PACS because they don't have dedicated cardiologists who are doing interventional exams," says Hoyt.
Of course, adds Jennifer Horowitz, HIMSS' senior director of research, "That trend is true in the radiology market in general: Smaller hospitals don't have radiology PACS either. That's where the penetration rates lag."
When hospitals do spend money on PACS, though, "They generally go across the board of modalities," says Hoyt. "They get all of their machines to write to PACS. Any analog machine that writes to film, they either have to throw it out or have a major renovation of the machine, so it makes a digital file that is then sent to PACS which then can turn it into a digital image on the monitor."
"Going to PACS is not just a matter of getting a PACS system up, but also may mean some of your older equipment has to be renovated or retrofitted or replaced," he adds.
That's a big and expensive undertaking, which explains why the PACS market, as relatively mature and saturated as it is, is nowhere near as dynamic as in some other areas of health IT.
A September Frost & Sullivan report highlighted how that saturation - combined with long replacement cycles and, especially, a focus of capital and manpower on electronic health records - has led to a lackluster market at the moment.
That said, PACS systems are set for a "steady but sluggish growth path" - from a roughly $1.1 billion market share in 2011 to $1.4 billion by 2018 - as hospitals look toward a new wave of next-generation systems.
"As vendors in the consolidated PACS industry strive to revive past growth synergies in the radiology informatics market, imaging providers are only slowly buying into the value proposition of new generation systems," said Frost & Sullivan principal analyst Nadim Daher in a statement. "In a market where PACS penetration is expected to cross the 90 percent mark across all imaging providers in 2014, replacement cycles stabilizing below the current seven year cycle could maintain sustainable - even if slow - revenue growth."
Indeed, "PACS replacement is a costly thing," says Hoyt. For the time being, "There may be consolidation as opposed to replacement."
He explains that, in the early years of PACS adoption, "You would buy an ultrasound machine, and it would come with its own PACS system. It never made film. You got this PACS system for ultrasound. And then you got this PACS system for nuclear cardiology. And then you decided you were going to go to full PACS, so all your plane films, and your CT and MR are now writing to the bigger PACS system - so you had the big one, and you had the ultrasound, and you had the nuclear."
Soon enough, a hospital would say to the ultrasound and nuclear vendors, "'I don't want to have a free-standing PAC system, you guys are going to have to write to this thing over here from Fuji.' And so you pay those guys to do that and basically redirect those images over the new [system]," says Hoyt.
"There are replacements and consolidations," he says. "Actually ripping out Fuji and bringing in Siemens, or ripping out Siemens and bringing in Philips, I don't know how much of that's going on. But I can tell you it's a pain."
There are some other areas poised for growth in the radiology PACS sector, according to HIMSS Analytics: digital mammography and orthopedics.
"Digital mammography has only been approved by the FDA for 10 years or so, so there's still a lot of analog machines out there that make film," says Hoyt. "That does not preclude those hospitals taking that sheet of film, putting it into a digitizer - like a Xerox machine, it reads the film, spits out the zeroes and ones, and puts it out on PACS - but they originally were analog."
Hospitals "are slowly getting rid of their analog machines to get digital ones that never make film but write as a PACS originally," he says.
As for orthopedics, Hoyt is excited about a newer phenomenon called orthopedic surgical templating.
Say an orthopedic surgeon is going to do a hip procedure. "You take all these measurements, and you buy the appropriate parts from Zimmer ... all the appropriate parts for Mary Jones, age 80," says Hoyt. Zimmer also sends a digital file - a template that shows "exactly how these pieces need to be installed" on Mary Jones' bones.
The surgeon has the template in front of him or her in the OR, showing you how he or she has it aligned compared to the template from Zimmer. "That guides you to make sure you've got perfect alignment and no errors," says Hoyt. "You could be off a degree or two, and that would cause some pain."
Beyond the coolness factor of orthopedic templating, there's plenty of exciting technology on the horizon. Hoyt says neuroscience and brain studies are one frontier that's "really fascinating."
But in the meantime - just as it is in your local movie theater and on your hi-def TV - the future of radiology IT is described with one number and one letter.
"Clearly, the next-generation stuff is 3D imaging," says Hoyt. "It's out there now for cardiology." In 2011, the FDA approved 3D imaging for mammography: "It improves imaging, and it reduces false positives. Which scares the living hell out of the patient. So if we can scare fewer, that's good."
One big challenge? Studies show that 3D imaging of mammography needs 8.5 times more disk space per exam.
"Mass storage: You're gonna' need gobs of it," says Hoyt.
Indeed, Frost & Sullivan estimates that picture archiving and communication system (PACS) storage requirements in U.S. hospitals grew at a rate of over 20 percent per year for the past five years and reached a whopping 27,000 terabytes in 2011.
"Cloud solutions are being adopted to address this current and ongoing need to manage storage growth," says Frost & Sullivan analyst Greg Caressi.
The growth is especially acute in pathology PACS. "All those little slides that pathologists have been looking at for 150 years are now being imaged and put out on disc," says Hoyt. "The rumor is that pathology PACS will make radiology PACS look small."
HIMSS Analytics, in fact, just started collecting data on 3D imaging adoption.
"A lot of people ask us, 'Why are you asking that?' Because we're trying to predict the future," says Hoyt.
And that's exactly what will be on the show floor at RSNA later this month. From 3D imaging to CT and MR hybrids with PET functionality that can do brain function analysis and beyond.
"I like going to RSNA, it's futuristic," says Hoyt. "It's like going to Star Wars."