In some cases, researchers found, doctors had to look at 20 different screens of information to determine all the drugs a patient was taking.
Although the system has been replaced, some of the newer versions Koppel has examined "still have many of the same faults," he said. "They are better, but they make other stupid mistakes."
Pharmacist Michael C. Cotungo is testing a computer program at Brigham Women's Hospital that will enable doctors to prescribe medicine on line.
(David Brown - The Washington Post)
|
|
Under the old system, he said, problems were not fixed because "hospital administrators are too often cowed by the IT [information technology] people" and because residents, who are responsible for writing prescriptions in teaching hospitals, are at the bottom of the physician hierarchy. To Suzanne F. Delbanco, executive director of the Leapfrog Group, Koppel's study demonstrates that "like any other technology, CPOE is a tool. Well-designed and well-implemented systems involve clinicians from the beginning," she said.
The Leapfrog Group has championed CPOE systems because studies have found they reduce errors that result from illegible handwriting and blurred copies in paper systems, and can also result in better controls on the dispensation of drugs, such as antibiotics. Sophisticated CPOE systems check for harmful drug interactions, alert clinicians to patients' allergies and cancel redundant drug orders.
Mark Leavitt, medical director of the Healthcare Information and Management Systems Society (HIMSS), the Chicago-based trade association for medical information technology companies, cautioned against overreacting to one study.
"Hopefully no one is regarding CPOE as being the silver bullet to the problem of medical errors," Leavitt said, because it isn't. "CPOE is one of the most difficult applications to create because it's the thing that can slow the physician down the most."
The study, he added, is a useful reminder "that the people [who use a system] are more important than the software."
David W. Bates is a professor at Harvard Medical School and one of the developers of a much-studied and widely admired CPOE system pioneered at Boston's Brigham and Women's Hospital that is now used at other Harvard teaching facilities. Bates said that Koppel's study highlights the need to continuously improve computer systems after they have been installed.
"I think it's very important for hospitals to track feedback and allocate resources," said Bates, who has published studies showing a sharp decrease in the error rate at Brigham after the installation of a CPOE system in 1993.
That view is echoed in a companion editorial to Koppel's study entitled "Still Waiting for Godot" written by physicians Robert L. Wears and Marc Berg. The authors note that the failure rate for large IT projects in health care is about 75 percent.
"Behind the cheers and the high hopes that dominate conference proceedings, vendor information and large parts of the scientific literature," they wrote, "the reality is that systems that are in use in multiple locations, that have satisfied users and that effectively and efficiently contribute to the quality and safety of care are few and far between."•
