Only 4% of U.S. physicians in ambulatory care have access to an advanced, "fully functional" electronic health-record system, but even those top-tier systems may not be fully featured enough to qualify for maximum payments under the new CMS pilot program to boost EHR adoption.
Still, most healthcare information technology experts contacted for this story reacted favorably to the release of the executive summary of what may be the most authoritative and methodologically solid study of EHR use to date.
The summary was published in the New England Journal of Medicine. The survey work was conducted under two $600,000 grants from the Robert Wood Johnson Foundation and another $3.6 million grant from the Office of the National Coordinator for Health Information Technology at HHS, the latter of which paid for both the ambulatory-care EHR survey and a separate hospital IT survey that is yet to be completed. A final report on the ambulatory survey is due July 2.
The survey of 2,758 physicians was conducted between September 2007 and March 2008 by the Institute for Health Policy at 902-bed Massachusetts General Hospital, Boston, the Harvard School of Public Health, George Washington University and RTI International.
"This is the best data that there has ever been on the adoption of electronic health records by physicians," said William Jessee, the physician president and chief executive officer of the Medical Group Management Association, Englewood, Colo. That's high praise, given that the MGMA conducted its own IT physician adoption survey of more than 3,000 medical groups in 2005.
"This was a really expensive study," Jessee said. "They spent a bunch of money on it, but that's what you've got to do."
The study also earned praise from physician Blackford Middleton, chairman of the Center for Health Information Technology Leadership at 10-hospital Partners HealthCare System, Boston. The center in a 2003 study estimated that high-end ambulatory EHRs with computerized order entry could prevent 130,000 life-threatening events a year.
"I think it is a good study with a national sample and good sampling methodology," Middleton said. "Their findings are similar to findings we've had before from less rigorous studies. It really confirms a lot of well-known market perspectives."
The survey defined EHRs by functions. The EHR systems fell into two categories. Those with seven functions were dubbed "basic" EHRs, while those with 16 functions, including electronic prescribing and decision-support features such as alerts for contraindicated drugs, were deemed to be "fully functional."
Basic EHRs, which are available to 13% of physician respondents to the survey, could not pass muster with the federally supported Certification Commission for Healthcare Information Technology under current criteria, according to commission Chairman Mark Leavitt. And even the fully functional EHRs used by the remaining 4% of physicians may not pass the CCHIT test under the expanded 2008 criteria unless they have at least one interoperability feature not measured by researchersthe ability to send and receive care summaries using the Continuity of Care Document/Continuity of Care Record data standards.
"They would meet the 2007 requirements, it sounds like," Leavitt said. "They wouldn't with the 2008 criteria."
CCHIT will take applications from July 1-14 for testing ambulatory EHRs under the tougher, 2008 criteria with announcements of the first batch of EHRs certified under the new strictures expected by early October, Leavitt said.
That should be just about in time for adoption by physicians interested in participating in the proposed new CMS pilot program. Under the CMS program, office-based physicians are to receive bonus payments for reporting and meeting certain quality measures using EHRs.
At a recent meeting of the American Health Information Community, an IT advisory panel appointed by HHS Secretary Mike Leavitt, acting CMS Administrator Kerry Weems questioned the CCHIT chairman about its testing program and its relevancy toward the CMS pilot. According to a fact sheet on the CMS Web site, all physicians participating in the pilot must have a CCHIT-certified EHR by the end of the second year.
As long as the CCHIT certification is still valid, it will be acceptable for the demonstration, according to the CMS Web site. CMS does not necessarily expect practices to update their software every year based on the latest certification standards. However, in determining the financial incentive for use of an EHR, practices that have EHRs with more recent certifications and therefore meeting higher standards will be eligible for a higher level of incentive payments.
Mark Leavitt said, "Mr. Weems was asking questions as to that, so I think they are still making a decision. I think it is fairly likely that they'll require (2008) certification because it has that step forward with patient summaries."
If CMS goes forward with requiring systems to qualify with the most recent certification criteria to be eligible for the highest payments, the policy would at least be consistent with other HHS/CMS healthcare IT requirements. In 2006, the federal authorities granted hospitals wanting to subsidize EHR systems to physicians relief from Stark and antitrust laws if the EHRs being offered have been certified for "interoperability" within a year of adoption.
C. Peter Waegemann, executive director of the Medical Records Institute, Boston, said it is unclear what impact any CMS requirements or its pilot might have on EHR adoption, either way.
"I know about 60 of the 300 EMR vendors covering about 70% of the market have the CCR/CCD reporting capability, so that's not such a big deal," Waegemann said. "I doubt whether CMS will have such a big impact. CMS is changing all the time. By January, we'll have a new administrator, and who knows what they will determine. All in all, it's interesting, but it's not clear what it means."
William Bria, chairman of the board of the Association of Medical Directors of Information Systems and chief medical information officer for the Shriners Hospitals for Children, said the persistently low penetration numbers were, "to me, not a surprise."
"Nothing's changed in the fundamental business equation, the idea of being able to afford it and what it takes to say there is an unequivacable incentive," Bria said. "Mainstream medicine is taking a cautiously positive posture. They feel it is something to be involved in, so the notion of discussing when are you going to do it or how you're going to do or if you're not going to do it is going mostly to the side of when. But these things are very expensive. It's still a risk.
"I think people, especially with the coming change in regime, are looking for a whole change in the healthcare landscape," Bria said. "If the early adopters are already adopted, people are probably waiting to see what's going to happen."
Steven Waldren, the physician director of the Center for Health Information Technology at the American Academy of Family Physicians, Leawood, Kan., said he would have liked to have seen the data broken down by medical specialties. Two AAFP surveys last year showed an adoption rate of 40% to 45%, Waldren said. Results from a third, recently completed member survey, should be out in a couple of months, he said.
"Regardless of the true number relative to adoption, the healthcare industry has hard work ahead to align incentives, develop best practices for adoption and utilization, and deploy health IT nationally," Waldren said.
Fix the money
The recent study highlighted the high cost of EHR systems as a key barrier to adoption, but Middleton said the summary didn't emphasize two other key impediments: the misalignment of incentives in which the benefits of EHRs go to payers and other healthcare industry players while the costs are borne by physicians; and systems that don't yet provide the latest relevant information about best practices at the point of care.
"We need to have national initiative to fix the problems of misaligned incentives," he said. "We also are finding most folks who adopt an EHR are adopting an empty shell. They don't have the knowledge. Every clinic is expected to translate the knowledge into decision support, and they don't have the time or the ability to do that."
"Hopefully, it will help us make the case that adoption is not proceeding because of these things not being addressed," Middleton said.
Physician Richard Baron, founder of the five-physician Greenhouse Internists, an independent group practice in Philadelphia, spoke at a news conference when the report was released.
Baron's group adopted a full-function EHR in July 2004, which he described as the hardest thing his group has ever undertaken.
"We were all working past midnight day after day to make the switch over," Baron said. Their reward, he said, was a better system, but at a cost of more than $40,000 per doctor and a 2% decrease in revenue the first year.
"It interrupted every system in our office to do work," he said. "It made us dependent on a technology we did not understand and could not support ourselves."
As a result, the group pays an additional $60,000 a year for tech support, Baron said. Confirming Middleton's point about misalignment of incentives, Baron said none of either the initial cost or the ongoing expenses of the system are being reimbursed by payers, pharmacies or patients who are deriving the greatest benefits from the technology, he said.
"In the current reimbursement system in primary care, we are paid by the length of our progress note," he said. "We get paid for seeing people face-to-face in our office; we don't get paid for e-mail consultations."
In addition, "getting that information into the record turns out to be a whole new species of work in primary care, and it is not reimbursed," he said. "I really think new financing strategies are desperately needed."
A grain of salt
According to study report co-author Catherine DesRoches, an assistant in health policy at the Institute for Health Policy at Massachusetts General Hospital, 16% of physicians surveyed reported that they had purchased an EHR system but had yet to implement it while another 26% indicated they plan to buy an EHR in the next two years.
Jessee warned policy wonks and EHR vendors against getting their hopes up. The 2005 MGMA survey found a 14.1% EHRs adoption rate across all groups, defined as three physicians or more, but another 14.2% indicated that they planned to implement a system in the next 12 months and another 19.8% reported they'd have an EHR within 13 to 24 months.
If DesRoches' survey results are accurate, what they told us three years ago and what they did subsequently are two different things," Jessee said. "I think everybody is overly optimistic about how long this is going to happen."
What do you think? Write us with your comments at [email protected]. Please include your name, title and hometown.