In a healthcare regulatory environment that most practice leaders would describe as nothing less than crushing, the federal meaningful use (MU) program provides an interesting case study of how radiology practices have responded to a program of noble intentions but minimal relevance to the specialty—and one that provides radiology and other so-called hospital-based specialties a built-in exemption (or easy out).
Four radiologists—including one whose practice sat on the sidelines these past five years—provided insight into the costs and benefits of participation on December 1, during a refresher course, “Meaningful Use for Radiologists: Pros and Cons,” at the annual meeting of the Radiological Society of North American in Chicago.
Moderated by Brigham and Women’s radiologist Ramin Khorasani, MD, the session made clear that the costs to date have been considerable and will continue even when incentive payments stop in and penalties begin. This year’s unveiling of the Merit-based Incentive Payment System (MIPS) raised the stakes for participation with the announcement that a full 25% of the MIPS score will be attributed to successful meaningful use attestation.
With most of the nation’s radiologists claiming an exemption based on Provider, Enrollment, Chain, and Ownership (PECOS) specialty code 30, no definitive answers were forthcoming for several of their pressing questions: How will the MIPS score will be calculated for exempt physicians? Does the five-year limit on exemptions from the MU program apply to physicians claiming the specialty code 30 exemption?
The presenters’ experiences were salted with a measure of rue, leavened with a certain amount of ambivalence and guided, in some respects, by luck and ingenuity. With so many variables undecided and rulemaking ongoing and in a continuous state of revision, the presenters left the audience feeling like the runners in the circular caucus-race in Alice’s Adventures in Wonderland when they asked, “But who has won?”
University Radiology: The All-in Private Practice
First up was Alberto Goldszal, PhD, CIO, University Radiology, New Brunswick, N.J. One of the leading voices for radiologist private-practice participation in MU, Goldszal provided a big picture summary of the intention of the program: “It’s really about engagement,” he said. “It’s about access to information and ultimately, hopefully improved outcomes. That’s the goal of the meaningful use program.”
Nonetheless, as the program has rolled out, radiology has struggled to find relevance, particularly in stage 1, which is mostly about structure and collecting a great deal of patient data. “For us in radiology, stage 1 was a bit meaningless because we have been driven by information systems for 20 years with the advent of CT into the clinic,” Goldszal acknowledged. “But the rest of medicine—the other departments in the hospital—lagged behind radiology many years and that was still the case in 2011 when MU rolled out.”
Goldszal said the program began to add some value to the practice in stage 2 with its focus on data exchange and is hopeful that stage 3, delayed until 2018, will bear fruit with its analytics and the outcomes focus.
“From the beginning, the goal here is to provide healthcare that is patient-centered, evidence-based, professional, and efficient,” he said. “That is an altruistic goal and you can judge how far we’ve come.”
University Radiology made the decision to participate in the MU program from the outset and experienced first-hand the twists and turns of the program. Stage 1 was intended to last two years before the introduction of stage 2, but lasted four years.
From the start, eligible professionals (EPs) had to meet 15 core measures (with some exclusions), 5 out of 10 menu-set objectives, and 6 clinical-quality measures. “There were a bunch of things we had to do to just qualify, and it was a bit confusing,” he reported. “There was a little bit of overlap between HEDIS, PQRS and e-Prescribing.”
University Radiology went through the motions, collected the data, submitted it, and received incentives for the first four years of the program. In December, the practice was in the processing of collect data for stage 2 MU.
“Radiology, from the beginning, could elect the exclusion for the first five years,” Goldszal noted. “If you elected the exclusion five years ago in 2011, time is up. You are going to get, perhaps, an adjustment in your payment from CMS.”
This year, the Office of the National Coordinator announced the delay of stage 3 (until 2018) and a modified version of stage 2. All EPs must meet the same 9 measures (8 for hospitals), whether they are in year five or just starting out, although there are different thresholds for first-year, stage 1 participants. There’s a 90-day reporting period in 2016 for everybody, Goldszal said.
“In my view, the problems have been streamlined,” he added. “It is easier now to understand what they are trying to do. That doesn’t mean it’s easier to collect the data to get to meaningful use, but it is easier to understand.”
From Goldszal’s vantage point, the foremost MU challenge for radiologists is late adoption. “I thought we’d do stages 1, 2, and 3, and then it would take a back seat, but it’s quite the opposite,” Goldszal said in reference to the MIPS implications. “Meaningful use has been re-energized, and it has more relevance than ever.”
Other significant challenges University Radiology has encountered include exchanging data with outside organizations. “Health systems are very good in communicating information within their information systems vertically, but when you are crossing borders, they are very poor at exchanging data,” he said.
Clinical decision support, integration with other health information exchanges, and submitting data to a registry were all new to University Radiology, and meeting those objectives presented challenges for the practice.
Another common challenge has been electronic view, download, and transfer (VDT) in the context of the patient portal, he added. “If you did not have a relationship with your patients before, now you do, and that takes time, money and effort,” he said.
In 2014, 20% of the individual participating physicians had to return money, and a total of $33 million was returned to the government. “The No. 1 issue was the HIPAA Security assessment,” Goldszal reported. “I think people did not understand what that was all about, and a lot of people are still in the dark about it.”
Challenges and irrelevancies aside, the MU program has enabled University Radiology to accomplish some long-held goals, not the least of which is data sharing with patients. “We can now have these engagements with the patient, provide our results to the patient, empower the patient to be a part of their own care, so that’s also a big plus that has been derived from the meaningful use program,” he said.
Additionally, the practice’s RIS has become far more capable than before, with many more interfaces as a result of MU. Previously, the RIS integrated only with the EMR, upstream with the hospitals, with PACS, and with electronic billing. As a result of MU, it now features the following interfaces: VDT with the portal, e-prescribing, e-links with lab results.
The RIS also displays a problems list, aiding radiologists with exam interpretations. “All of this is incorporated into our system as a consequence of meaningful use,” Goldszal said. “What’s important is that all of these follow standards, HL-7, the e-links, these are all standards based interfaces.
In conclusion, Goldszal reviewed the elements of the MIPS grade, which will provide the basis of value-based reimbursement: 30% will be based on quality measure performance(PQRS), 30% on efficiency (cost data from PQRS quality data), 25% on MU adoption, 15% process improvement (e.g., providing access to data, opening late for patients).
“Everybody has a grade today, it’s called the VBM (value-based modifier), and only CMS knows the grade,” Goldszal said. “The difference here is that MIPS was developed for public consumption. There is going to be a grade that’s going to be posted out there, and its going to compare radiologists across different practices. Those that perform well are going to get incentives, on the order of 27% by 2022, 2019 it’s 12%, and the penalties go up to about 9%. When we have margins of a single digit, if you have a penalty of 9%, that wipes out all of your profit.”
He urged practices to develop a solid foundation for value-based reimbursement, including attention to PQRS, VBM and MU. “It is my personal recommendation that you try to adopt it, even if it’s late,” he said.
Scottsdale Diagnostic Imaging: Meaningful Use Lite
Southwest Diagnostic Imaging, Scottsdale, Ariz., is the fourth largest private practice in the nation, according to this magazine’s list of the 100 Largest Private Radiology Practices. It also is one of the leading operators of outpatient imaging centers with a total of 28 sites.
The practice reads about 1.2 million exams per year, about half of which originate from its imaging centers. Jim Whitfill, chief medical officer, Scottsdale Health Partners, and CIO for SDI, oversaw the implementation of MU at SDI, and his account was much less sanguine than Goldszal’s.
Whitfill, who is currently working with an accountable care organization, spent enough years in radiology to know that applying data collection for MU stage 1 within the radiology workflow would be like trying to fit a square peg into a round hole. “I will relay my experience at Southwest Diagnostic Imaging on how to take a radiology RIS-based workflow and use a standalone community-based EMR to achieve meaningful use,” he began, somewhat sardonically.
SDI’s client hospitals had a certified ambulatory EMR, but that ambulatory EMR was not used in the hospital environment where most hospital exams were read. SDI decided to focus its MU efforts on the outpatients in its freestanding outpatient imaging centers.
In performing due diligence, SDI’s leadership closely read the Federal Register and even corresponded with CMS on what constitutes “seen by” for a radiologist, a question that engendered much debate in the radiology community during the early days of MU. Was it the patient seen by the mammographer performing a biopsy or the virtual patient for whom the radiologist read the CT scan?
You decide, was the CMS response, but be consistent. SDI elected to collect data only on those patients seen in a face-to-face interaction by the radiologist. “If we are reading a head CT on a patient and there is no face-to-face contact, we do not track meaningful use on that patient,” Whitfill said. “However, if we are doing a thyroid biopsy and the physician was in the room with the patient, we track MU data on that patient.”
The practice also faced a technology challenge when its RIS vendor decided not to pursue complete certification. SDI didn’t want the disruption associated with RIS replacement, so it decided to move forward with an ambulatory EMR and designed a workflow for data collection under that scenario.
When the PECOS specialty code exemption was announced late in 2012, SDI was forced to make another decision: Abandon its plan and claim the exemption or get into the MU game. “We didn’t know what was going to happen with the exemption,” Whitfill said. “We didn’t want to be in the place where we had to buy this more expensive technology down the pike when there would be no stimulus money left.”
SDI began data collection in the fall of 2013 and attested to stage 1 MU in February 2014 for 51 providers. It attested to stage 1 again in 2014 with a few more providers and in December of 2015 was in the midst of its stage 2 data collection, with plans to attest in January of 2016.
Back when it appeared that radiology would be required to collect height, weight, and blood pressure on all patients, SDI did a time-and-motion study to understand the costs involved. “The bottom line was on a practice that was doing 300,000 exams a year, we thought it was going to cost an extra $160,000 in labor just to collect height, weight and blood pressure, as well as some of these other data points,” Whitfill said. “For radiology, you have to consider the total cost of ownership.”
For SDI, that included the cost of an ambulatory EMR that was so tangential to the core delivery of care to patients that when the vendor called to do software upgrades as MU evolved, SDI’s response was, “whenever”—despite the fact that the practice participates in an ACO, a private health information exchange, and the PQRS program.
“Our vendor colleagues really have been struggling just to get their platforms stable,” Whitfill said. “My colleagues in primary care, every specialty is dealing with these problems, so its not just radiology.“
The program requires a good deal of monitoring. Some partners have moved in and out of the program. As new radiologists enter the practice, some are in stage 1 while others are in stage 2. Individual patient encounters are monitored and when a physician has what SDI calls “very low unique patients seen by the EP,” the radiologist gets a call to do a fluoroscopy shift in one of the outpatient centers.
Another cost to consider: Several pre-payment audits in the summer of 2014. “We had two different auditors who were auditing with our 2013 data, on our 2014 data we had no audits,” Whitfill shared. “These are difficult processes. The take-home point is the audits are non-trivial and you need to be ready for them.”
The process was legalistic, but the auditors were not technical people, so SDI provided lengthy responses. Being able to provide supporting documentation for all of your MU modules is critical, particularly in cases where the EMR reports appears to be off base.
“I know what the auditors care about: What the EMR says, not what is true,” he said. ”We packaged all of our audit responses into a giant PDF in a zip file. Don’t make it hard for the auditor, make it as easy as possible with a guide on how to do things.”
One challenge that required considerable back-and-forth was the core measure requirement to disseminate eletronic information to patients. SDI claimed that zero patients had gone to its patient portal in stage 1, but added that it had distributed many CDs that came from non-certified technology.
The auditor insisted that a CD was electronic information and the numerator and denominator need to be reported. SDI maintained that the CDs came from a non-certified system and therefore did not require data collection. On the third go-round, a supervisor stepped in and agreed with SDI.
Another audit revolved around rejected EPs. CMS claimed that SDI had not submitted two of its radiologists, but SDI was certain it had. Upon further review, CMS found the entry for one radiologist but not the other. “We learned our lesson,” Whitfill says. “Now we take a screenshot of every submission and print it out. That was a rookie mistake.”
When SDI went through its audits in 2014, there was no appeals process, but that has changed. You can actually go onto the CMS web site and get your data, get a form for the appeals process,” Whitfill said. “I think CMS is a tough place, I don’t want to give them a hard time, but the lack of feedback can be challenging.”
Whitfill pegs SDI’s total cost of ownership for the MU program at about a quarter of a million dollars a year. “This ended up being a fairly large financial outlay for us,” he said.
SDI encountered many challenges in complying with MU of health IT, including some specific to Arizona. For instance, the state will not allow syndromic surveillance by physician practices and will accept data from hospitals only.
The state cancer registry would not accept data from SDI based on the state belief that radiologists do not diagnose or treat cancer. Instead, SDI elected to submit data to the ACR’s National Mammography Database because that data comes from its RIS, which has modular certification.
Whitfill said that while CMS typically does a good job of providing explanatory information about the MU program, there is currently just one Power Point document available on the ten measures that all EPs must meet moving forward. “We have to get ready for what we think the attestation is going to look like, but we don’t know exactly what will be, he said.
This uncertainty has added to the reservations SDI has about moving forward with the MU program. SDI is an ACO participant, and because ACOs do their own quality reporting, member practices don’t have to participate in PQRS. Participants in an Alternative Payment Model are exempt from MIPS.
“For us, the bottom line is that it is costing us in the neighborhood of $250,000 a year to participate in MU, so we are looking at potentially sun-setting it unless there becomes another financial reason not to,“ he said. “Depending on how the exemptions play out, I think you are going to see fewer and fewer practices participating moving forward.”
NYP/Weill Cornell Medical Center: On the Sidelines
Keith Hentel, MD, executive vice chair, department of radiology, New York-Presbyterian/Weill Cornell Medical Center (NYP/WCMC), provided the perspective of an academic practice that decided not to participate in MU. Just 4,720 unique radiologists to date have participated in the MU program, so Hentel’s experience is likely a truer reflection of the house of radiology.
“I’m not a meaningful user and none of the radiologists in my practice are meaningful users, “ he said. “I’m embarrassed about this, because when you look at MU and the goals of MU, these are like Mom and apple pie. These are things we obviously want to achieve as a practice.”
In 2011, eyeing both the objectives and the incentives, Hentel and his fellow 63 radiologists were optimistic about their ability to attest to meaningful use of health IT. Employed by the larger physician organization, they qualified as EPs, were collecting drug-drug interactions, used CPOE, and believed they could meet the then-15 core and 5 menu-set objectives with relative ease. They worked in two enterprise EHRs, one that served inpatients and one that served outpatients, and neither of which were certified in early 2011.
Nonetheless, there was considerable support in radiology for the project, Hentel recalled. Despite that support and the potential $2.5 million in annual incentives, the radiology department failed to become meaningful users.
One barrier Hentel cited was the ambiguity associated with the MU regulations, including 196 pages in the Federal Register and various summaries from CMS. The bigger barrier, though, was what Hentel called “the Great Divide.”
The great divide
While NYP/WCMC deploys both EHR technology and radiology IT systems at its institutions, it is radiology IT’s job to transverse the gap, typically using HL-7 messaging. As in most radiology departments, radiologists scheduled, protocoled, in-took, and completed patients in its RIS, viewed images within its PACS, reported within its speech module.
“The only thing that we were using our EHR for in our daily embedded workflow was to register the patients,” Hentel said, adding that there was no true integration between the EHRs and radiology IT systems. “We had read-only access to look at the medical records of patients, and we would review that on cases for which we felt we need more information.”
Another barrier was the perceived lack of relevance to radiology, he said. “It would have been collecting information for the purpose of qualifying for MU and would not have improved the quality of care we were providing for our patients,” Hentel said. “There truthfully was a lack of meaning that tempered our enthusiasm to do it.”
Poor return on investment for radiology was another limiting factor, considering that the Weill Cornell Physicians Organization had a total of 830 EPs not including radiologists. “In radiology, we could have generated 7% of the total MU dollars.,” he said. “To get radiology practices anywhere near where we could attest would have cost a lot more and taken a lot more than 7% of the effort, because we were different—we were operating in systems other than what most of our providers were practicing in.”
Besides the resistance Hentel received from other members of the physician’s organization, he had other, competing projects, including the Medicare Imaging Demonstration project and the implementation of clinical decision support, still in use in the practice today.
“We think that had a lot of bang for the buck,” he said. “Meaningful use took a back seat to some of these other projects that we were trying to accomplish.”
Nonetheless, Hentel expressed some concerns with not being meaningful users. Financial implications and perception problems are concerns.
“I worry about it in the context of the greater medical community at my institution,” he said. “Especially as our practice transitions into accountable care, I want to be considered just as much of a clinician as any other practitioner in our enterprise. Obviously, I worry about the potential pay penalties that are slated to come up in the future.“
The jury is out on whether the exemption for PECOS specialty code 30 does indeed expire after five years, Hentel said. “There is a little bit of discussion about whether the time limit applies only to the individual exemption or if it applies to automatic exemptions as well,” he reported. “Even if it does expire in five years—which is what I am assuming right now—that will actually bring us to stage 3.”
Stage 3 is about presentation data, clinical decision support, summary care data, and engaging with registries, all of which are pertinent to radiologists, he said. The exemption enabled the practice to bridge the gap to MIPS, and Hentel is keeping a close eye on MIPS rulemaking.
“In summary, maybe failure wasn’t so bad,” he said. “It gave us the opportunity to take a different approach.”
Brigham and Women’s Hospital: The All-in Academic Practice
Brigham and Women’s was probably the first large academic medical center in the nation to get onboard with the MU program, according to Ramin Khorasani, MD, vice chair, informatics, Brigham and Women’s (B&W) department of radiology. The hospital had a homegrown EHR that had been certified as CEHRT by the federal government and a leadership team for MU was already in place when radiologists determined to become meaningful users.
The threat of penalties was of far greater concern than the potential incentives, Khorasani said. “A 5% reduction in our payment on CMS claims, which is about 30% to 40% of our revenues, would be a massive hit to our practice,” he said.
As national leaders in health IT, Brigham leadership thought it was directionally correct to participate in an initiative that moved the nation forward in health IT adoption. From the outset, radiology enjoyed the efficiencies of attesting with the entire B&W physician community.
For instance, in approaching stage 1, which was about data capture and sharing, Khorasani recognized the objective was not about radiology entering data to become a meaningful user, it was about someone in the institution entering data on a patient seen by a radiologist. “So, if a secretary in internal medicine entered smoking status when they were seeing my patient, that actually counted for me as a MU data entry,” he said.
Attesting as a practice also effectively streamlined the number of objectives radiology needed to satisfy. “We focused on what we had to get the radiologist to do and what we had to get the nonradiologist to do,” Khorasani explained. “When we focused on that, it made it simpler for us to do.”
Khorasani and team identified just two measures that the radiologist could have an impact on: maintaining an active allergy list and an active problems list. “If you have a patient whose got an allergy, you have to be able to document that allergy in the EHR that’s certified, not in the RIS. We thought that was a useful thing to do,” Khorasani said. “The other was the active problem list. If there were no problems on that problem list we could enter a coded problem list into the EHR. “
The EHR view
In order to contribute to the organization’s meaningful use effort by helping to maintain active allergies lists and problems lists, radiologists would have to interact with the EHR. In one of the session’s most elegant—and meaningful—examples of meaningful use, Khorasani and team developed an EHR view within the PACS where radiologists could both contribute to and access problem lists and allergy lists.
“We spent our dollars to create an EMR view within our PACS workstation and train radiologists and front-desk staff to be able to use this tool,” Khorasani said. “We can at least have an ambition of improving the quality of care we are offering our patients by having more access to timely information about the patient on every report we generate.”
The view pops up on the radiologist’s voice recognition screen (the fourth monitor), with abstracted information drawn from the EHR. If the data on a patient is not sufficient to meet meaningful use, a banner would turn red and a thumb down logo would appear. The problem list provides the radiologist with useful information that he or she formerly had to chase.
If meaningful use data points are missing in the EHR, a form would print at the front desk, where it is given to the patient to fill out. The form is scanned into the EHR, and a secretary or nurse enters it into the EHR. The system also produces a notification dashboard that alerts referring physicians to provide information on patients who are missing problem lists.
By offering radiologists $1000 to show up for training, Khorasani engaged the entire practice in short order. “For performance improvement, a small amount of money goes a long way,” he said. “People get very competitive in qualifying for the dollars.”
Every scan, a patient
The greatest topic of discussion as B&W entered into meaningful use was how to define “patients seen by” the radiologist. Ultimately, the organization settled on collecting data on every patient that came through the radiology department.
“For radiology, every patient I did a report on goes into my denominator, whether I saw the patient or not,” Khorasani said. “We defined ‘seen by’ as anyone I report on. This was important to us for a couple of reasons.”
As a strategy to meet MU, reporting only on those patients with a face-to-face encounter is perfectly understandable, Khorasani said. Radiology ambitions at B&W, however, were broader.
“We really had a problem with this minimal consultative service FAQ,” he said, referring to the CMS frequently asked question on the subject of determining patients seen by an EP. “For us to take our seen by to only be patients we saw, fluoroscopy for example, would mean that when I read an abdominal CT, it was a similar consultative service to a cardiologist reading an EKG for another physician. That just didn’t sit very well with the leadership team in our practice.”
The practice viewed the issue within the larger context of lobbying Medicare about the value radiology adds to the care process. “Saying we provide minimal consultative service didn’t sit well with us,” Khorasani reiterated. “That’s why we took such an aggressive approach.”
Costs and benefits
With 96% of radiologists at B&W having met stage 1 requirements, the practice has successfully attested every year since 2012. In December 2015, it was in the midst of a 90-day reporting period for stage 2.
The practice has spent about $42,000 of the anticipated $44,000 incentive program payments, so future investments toward meaningful use will fall to the practice’s bottom line. Radiologists received about $3,000 per year from 2012 through 2014, and were expected to receive $2,000 for 2015, addition up to about 33% of the funds. The remainder went to various cost centers, including people, IT, and hospital shared costs.
In 2016, all providers will have to meet modified stage 2 for the entire year, except for people who are just starting to become eligible providers.
“From a structured perspective, institutional meaningful use was crucial for us,” he said. “This provided a multi-disciplinary team, a process for funds flow, and a technology certified EHR. I chaired this team for radiology.”
This program cost radiology about $100,000 of operating expenses per year, plus $100,000 in capital in year one to create a view of our EHR that is relevant to meaningful use at the PACS workstations.
“In summary, radiology practices in large academic institutions should leverage the existing certified EHR and MU programs if they exist,” Khorasani said. “You cannot do this as a radiology practice in a big place without joining what everyone else is doing; that’s the only way to do it.”
While program participation offered modest financial incentives and penalty avoidance, Khorasani described the biggest gain in terms of patient care. “MU in our practice was used to improve clinical performance, because we dragged clinical information into the workflow for every case we read,” he concluded.