Healthcare IT vendors critical of CPOE study
The authors of "The Role of Computerized Physician Order Entry Systems in Facilitating Medication Errors" found "that a widely used CPOE system (TDS) facilitated 22 types of medication error risks."
Questionnaires, focus groups and one-on-one interviews with staff at the University of Pennsylvania Health System discovered problems with "fragmented CPOE displays that prevent a coherent view of patients' medications, pharmacy inventory displays mistaken for dosage guidelines, ignored antibiotic renewal notices placed on paper charts rather than in the CPOE system, separation of functions that facilitate double dosing and incompatible orders, and inflexible ordering formats generating wrong orders."
The study's results broke into the mainstream press. National Public Radio, for instance, told listeners the study found that "hospital computer systems" might cause more problems than they solve.
"The reporting of it was very slanted," said Hoda Sayed-Friel, vice president of marketing for Meditech. "Uneducated consumers could have been thrown off by that."
"While the article should be viewed as a valuable cautionary tale, it should not be generalized to conclude that electronic physician order entry promotes errors," said Sam Brandt, MD, Siemens chief medical informatics officer. "Each of the root causes of error identified is amenable to remediation."
Some vendors fear the study could have the same braking effect on adoption rates as the 2003 CPOE implementation failure at Cedars-Sinai Medical Center, where officials pulled the plug on the system following a physician rebellion.
The vendors' main objection? The study was based on a first-generation system deployed in a single setting, noted Eclipsys' chief medical information officer, Rick Mansour, MD. That the nearly 20-year-old TDS system had some shortfalls is old news, he suggested in a written statement to the media.
"As the study was being conducted, UPHS had already recognized the need for a state-of-the-art system," he said. The hospital system bought the Eclipsys successor to TDS, Sunrise Clinical Manager, "which has been designed to address the types of human errors described in the study in conjunction with appropriate training and complementary workflows."
"The author has his heart in the right place, but the system he's describing is totally non-integrated," complained Jay Eisenberg, a Cerner physician executive. "It's like going out and buying a 1972 Pinto and reporting that if cars get rear-ended it'll start a fire."
Sayed-Friel agreed. "The way CPOE is rolled out today is very different than the way it was rolled out 20 years ago," she noted. "Today, it's a combination of technology and workflow and cultural change. It's really about adapting technology to workflow and implementing best practices."
The Healthcare Information and Management Systems Society took the unusual step of issuing its own press release to respond to the JAMA study. Its critique noted "the study is based on interviews with ordering physicians and observations of them at work, but it does not provide actual measurement of errors or adverse events." Additionally, "the study does not compare errors from paper ordering to computerized order entry." HIMSS' own analyses show that when compared to paper ordering, CPOE "results in a substantial overall decrease in errors and adverse drug experiences."
Nearly all the vendors interviewed agreed that the silver lining in the CPOE study storm cloud was clarifying that technology alone won't solve problems and that hospitals implementing CPOE should take a hard look at their own workflows. "It will open some eyes to make sure that what they buy lives up to their needs and expectations," said Meditech's Sayed-Friel. "It highlights possible pitfalls and should prompt hospitals to ask vendors tough questions."
At least one vendor said he wasn't worried about the bad publicity.
"I think it will all blow over," said Ed Meehan, director of marketing for Kean's Healthcare Solutions Division. "There are enough obstacles out there – physician resistance and the costs of most CPOE systems – that this won't have much impact on the marketplace."