Our readers might get the erroneous impression that we are not advocates of CPOE from reading our many columns outlining some of the untoward consequences and other problems with CPOE and healthcare IT in general. Nothing could be further from the truth. We remain huge supporters of CPOE and clinical decision support and IT applications in healthcare. However, once again, we find ourselves commenting on yet another example of problematic CPOE.
In the new study, researchers culled the MEDMARX database for medication errors which included some indication that CPOE may have played a role (Schiff 2015). They then developed a taxonomy and coding for the types of errors, why they occurred, and what preventive strategies might be employed. From their list they then developed multiple scenarios and tested on a representative sample of CPOE systems at multiple facilities whether these error-prone scenarios might pass through.
From over 1 million reports available in MEDMARX they found 6.1% reported as CPOE-related. The reviewers actually found only about half of these were truly CPOE-related. The taxonomy was then developed on a sample of these reports and 21 scenarios were developed for vulnerability testing on representative systems.
Overall, 79.5% of the erroneous order scenarios were able to be placed (28% “easily” placed with no additional steps or warnings and 28.3% placed with only “minor workarounds”). Specific computer-generated warnings occurred for 26.6% of the erroneous orders but the vast majority of these (69%) were passive alerts and another 29% required workarounds that could be done. Interestingly, they encountered one system in which no alerts were triggered at all. It was found that alerts in that system had been turned off several months earlier during an upgrade and were never turned back on!
This process, of course, is quite similar in concept to that developed and used by The Leapfrog Group to simulate problematic medication orders (see our previous columns for July 27, 2010 “EMR’s Still Have a Long Way to Go” and June 2012 “Leapfrog CPOE Simulation: Improvement But Still Shortfalls”). That tool was to be used by hospitals to assess the vulnerabilities of their CPOE systems to medication errors. Voluntary testing at over 200 hospitals (Leapfrog 2010) then revealed that about half of errors on “routine” medication orders were missed and almost a third of potentially fatal medication errors were also missed. Those hospitals adjusted their CPOE systems and protocols and nearly all showed improvement when retested. A similar test was conducted in 2011 by over 250 hospitals (Leapfrog 2012). The rate of missed potentially fatal errors dropped to just over 1% but the rate of missed “routine” medication errors was still on average about one third.
Tools like these to identify potential vulnerabilities are really needed to help prevent such errors from occurring elsewhere. Just like the problems we identify in our RCA’s and other case reviews, issues that occur at one facility or organization are likely to also occur at others. It is only through sharing experiences and lessons learned that we are likely to make progress in reducing errors and their consequences.
This is good work. Schiff and colleagues are to be commended for this considerable undertaking.
See some of our other Patient Safety Tip of the Week columns dealing with unintended consequences of technology and other healthcare IT issues:
Schiff GD, Amato MG, Eguale T, et al. Computerised physician order entry-related medication errors: analysis of reported errors and vulnerability testing of current systems. BMJ Qual Saf 2015; Published Online First 16 January 2015
The Leapfrog Group. Leapfrog Group Report on CPOE Evaluation Tool Results
June 2008 to January 2010. June 2010
The Leapfrog Group. Lack of Testing and Monitoring of Health IT by Hospitals and Vendors Potentially Jeopardizes Patients -- Hospital Performance Is Improving, but More Must Be Done. PR Newswire April 27, 2012
Print “PDF version”