Survey Tracks Slow Pace of EHR Adoption

By Peggy Peck
BOSTON, 19 june 2008-- A survey of more than 2,700 office-based physicians confirms that few have entered the digital medical world and that the biggest barrier to switching to electronic health records (EHRs) is cost. Only about 17% of respondents have EHRs, but the handful who had Cadillac systems credited the system with improving patient care by reducing medication errors and reminding them to order tests when needed, Catherine M. DesRoches, Dr.P.H., of Harvard Medical School, and colleagues reported online today in the New England Journal of Medicine. Just 117 physicians (4%) had high-end systems that included a full menu of functionality ranging from electronic charting capabilities to e-prescribing features and decision support tools.
Nonetheless, cost is a major stumbling block for physicians who are considering EHR systems.
The researchers said 47% of the physicians using EHRs said that the cost was a major barrier to expanding use of the system and 66% of those who don't have systems said the capital expense was a stopper for them (P<0.001).
But she said it wasn't clear if physicians had sticker shock from "the real cost of the systems, or if it is just their perception of the cost, what they have heard about cost."
Dr. DesRoches said systems often cost more than $100,000, but the price varies based on practice size and functionality.
She said that a good rule of thumb was to estimate a cost of about $30,000 per physician in the practice, "and that doesn't include the cost of maintaining the system, training to use the system, or system updates."
The cost might seem less daunting, she said, if physicians were sure they would have a good return on their investment, but 50% of those who don't have an EHR said they were uncertain that the investment would pay off, and 33% of those who had already installed systems still weren't sure (P<0.001).
Dr. DesRoches and colleagues used the 2007 Physician Masterfile of the American Medical Association to randomly select 4,484 practicing physicians for a mail survey. They excluded osteopathic physicians as well as residents, physicians working in federally owned hospitals, psychiatrists, anesthesiologists, radiologists, and pathologists.
Of the eligible physicians, 2,758 responded to the survey for a response rate of 62%. Responses from 151 physicians were incomplete and not used in calculating percentages.
Among respondents, only 117 had systems the researchers defined as fully functional, meaning systems that not only tracked patient demographics, diagnoses, medications, and clinical notes, but can also send prescriptions and order tests electronically, flag potential drug-drug interactions and out-of-range test results, and send reminders about evidence-based interventions or screenings.
Satisfaction with the EHR was highest among physicians who used these top-of-the-line systems.
Another 330 physicians had basic systems, which the researchers defined as those that allow electronic charting, but don't have an e-prescribing capability and don't provide clinical decision support tools.
In general, the researchers found, physicians were more likely to have access to an EHR if they practiced in large groups, in medical centers, or hospitals and were in the western region of the U.S. Primary care physicians were also more likely to use electronic systems than were doctors in other specialties (P=0.001).
Dr. DesRoches said she and her colleagues use a "home-grown" system developed for clinicians who practice at Massachusetts General Hospital. The system, she said, is typical of systems developed at large institutions in that it has added features in response to users requests.
"We are, for example, missing one of the features that we defined as fully-functional in the survey," she said. The system, she said, is not set up for electronic prescribing but "that will be added shortly."
Asked if the low adoption rate for EHRs suggested that it is an idea whose time will never come, Dr. DesRoches said that would be a misinterpretation of the findings. "Nothing that we've seen in this study suggests that this isn't going to happen. It will happen," she said.
And the stick that could make it happen is likely to be demands for EHRs from major payers, including federal and state governments through Medicare and Medicaid. Minnesota, for example, recently became the first state to mandate e-prescribing and the U.S. Senate is considering similar legislation.
The study has several limitations, its authors noted, including the potential for response bias. Additionally, the "estimates of the effect of these systems on quality of care and satisfaction are based on a small number of respondents with a large margin of error, especially for the fully functional electronic-records systems."
The study was supported by the Office of the National Coordinator for Health Information Technology and a grant from the Robert Wood Johnson Foundation. Dr. DesRoches declared no financial disclosures.
Primary source: New England Journal of MedicineSource reference:DesRoches CM, et al "Electronic health records in ambulatory care -- a national survey of physicians" N Engl J Med 2008; 359: 50-60 Published online June 18, 2008.