Leapfrog Group Wants Hospitals To Monitor, Not Just Implement, CPOE Systems

August 15, 2010, AJHP News

Cheryl A. Thompson

BETHESDA, MD 28 July 2010—To achieve “true meaningful use” of a computerized prescriber-order-entry (CPOE) system, according to the organization known as the Leapfrog Group, a hospital must routinely test and monitor that system after implementation.

The call for hospitals to do more than “plug and play” their CPOE system comes from an organization that has been pushing for the adoption of this technology for 10 years.

Evaluation tool. In a report released on June 30, Leapfrog announced that hospitals’ CPOE systems did not produce an appropriate warning for one third of the potentially fatal medication orders entered during a simulation.

The CPOE systems also failed to produce an appropriate warning for half of the orders that would be expected to result in a serious adverse event.

But nearly half of the hospitals that adjusted their CPOE system and related protocols on the basis of the 2008 test results performed better in 2009.

To use the simulation tool, a hospital had to complete Leapfrog’s annual survey, which includes indicating whether the facility has a CPOE system functioning in at least one inpatient unit. About 323 of the 1244 hospitals that completed the survey in 2009 had such a system.

Leapfrog said 214 hospitals used the tool to evaluate their CPOE system from June 2008 to January 2010. Each hospital received a scenario of 10 patients and 50 or 51 medication orders.

Results-based recommendations. Based on the results from 2009, said David Knowlton, chairman of Leapfrog’s board of directors, the organization recommends that

• The federal government insert into its definition of “meaningful use” of health information technology a requirement for hospitals to test and monitor all of their adopted technologies,
• Stakeholders share information about the best practices for hospitals’ adoption of health information technology, and
• Those hospitals without a CPOE system adopt one.

“As medicine grows more complex, it will not be adequate to rely on the individual memories of each and every clinician to assure a plethora of medication errors are avoided,” Knowlton said during a June 30 briefing for reporters. “We’ll need to rely on advancing technology to support clinicians, and we’ll need to improve on the performance of that technology over time.”

Earlier in June, the ASHP House of Delegates approved a professional policy concerning the definition of meaningful use of health information technology. That policy calls for the Society to advocate to policymakers for the definition to address continuous improvement, among other things.

On July 13, the Department of Health and Human Services released its initial definition of “meaningful use.”

The definition is actually the set of initial criteria that hospitals must meet to qualify for an incentive payment for adopting and successfully demonstrating meaningful use of certified electronic health record technology.

Those initial criteria do not include a requirement for hospitals to test and monitor all adopted technologies.

The initial criterion on CPOE at hospitals states that this technology be the method used by licensed health care professionals to order at least one medication for more than 30% of the inpatients receiving pharmacologic therapy.

Two years ago, Leapfrog used the results from its first-ever CPOE test to issue a caution on the importance of hospitals ensuring proper implementation of CPOE systems to prevent serious medication errors.

Hospitals using the same CPOE product to process the same orders achieved different scores, the developers of the tool reported in the April 2010 issue of Health Affairs. Some 13 hospitals using the same vendor’s product, for example, achieved scores ranging from roughly 10% to about 74% out of a possible 100%.

Preference for alerts.Northwestern Memorial Hospital, in Chicago, was declared a top urban hospital by Leapfrog in 2008 and 2009.

The organization has not announced the top hospitals for 2010.

Anne M. Bobb, Northwestern’s clinical informatics pharmacist, said she helped set up the hospital’s CPOE system for the 2010 test and sat with the physician as he entered the orders.

“I support the idea of testing your CPOE system to identify medication orders that can lead to patient harm,” Bobb said. “I think it’s a great idea.”

Part of CPOE system maintenance, she said, is the continual addition of clinical decision support, coupled with confirmatory tests, to avoid errors.

But Leapfrog’s tool tests CPOE systems only for decision support in the form of alerts, she said.

Decision support in the form of standard order sets—for example, when a prescriber enters an order for gentamicin therapy the system prompts the prescriber to order a gentamicin serum-level measurement as well—is not evaluated by the tool, Bobb said.

And, unfortunately, she said, the tool has not changed with time.

“There are drugs on there that have been removed from the market in the United States,” she said.

For that situation and others in which the tool’s medication order cannot be entered exactly as specified, Bobb said a hospital reports the drug as not being on the formulary.

The downside, she said, is that the tool does not provide replacements for orders of nonformulary drugs. Thus the hospital has fewer testable orders.

In the 2010 test, Northwestern received three different orders in the medication-checking category “Contraindicated Route of Administration,” but the testing physician could not enter any of them, Bobb said.

Northwestern’s CPOE system, she said, does not permit prescribers to order certain medications for administration by obviously unsafe routes. For example, the system does not accept orders to administer enoxaparin sodium by i.m. injection.

“We’ve engineered that [safety problem] out of the system, which arguably is safer than just giving an alert about it,” Bobb said. “And we got no credit for meeting that category because all those test questions were thrown out, so it actually hurt us to design a safer system.”

One of the criteria for being a top hospital in an urban setting in 2009 was fully meeting the standards for implementing a CPOE system and passing Leapfrog’s test.

To pass the test, a hospital’s CPOE system had to catch at least 50% of the potential problem orders in 6 or more of 11 medication-checking categories, 2 of which had to be allergies and drug–drug interactions. The system also had to have a score of at least 25% on the checks for false positives.

Leapfrog typically announces the top hospitals each year in September–December.