Health Information Technology

Children get less safety benefit than adults from hospital computer order entry systems

Commercially available computerized physician order entry (CPOE) systems may need modification to be as effective at preventing serious medication errors among children as they have for adults, according to a new study. Studies of the effect of CPOE systems on medication errors in adult patients have shown reductions of up to 55 percent in serious medication errors that would otherwise have not been intercepted by hospital staff. However, the results for pediatric centers have been much more variable.

In the new study, Christopher P. Landrigan, M.D., of Brigham and Women's Hospital and Children's Hospital Boston, and colleagues compared monthly rates of medication errors 7 months before and 9 months after implementation of a commercial CPOE system, collecting data on 627 children hospitalized at a hospital pediatric surgical or medical unit, a pediatric intensive care unit (PICU), and a neonatal intensive care unit (NICU).

Using comprehensive error surveillance methods, the researchers did not find a statistically significant difference before and after implementation of CPOE for total errors (44.7 vs. 50.9 errors per 1,000 patient-days), serious medical errors (31.7 vs. 33.0 errors per 1,000 patient-days), and nonintercepted serious medical errors (23.1 versus 20.6 errors per 1,000 patient days). They found a downward trend in nonintercepted serious errors from the beginning of the academic year (September or October) to later in the academic year (February or March), both before and after implementation of CPOE, suggesting the learning curve for new residents. However, there was a significant 7 percent drop in the level of rates of nonintercepted serious medication errors after CPOE implementation compared with pre-CPOE rates. No similar significant changes were seen for all medication errors or all serious medication errors, or for injuries caused by medication errors.

The researchers point out that the commercial CPOE system evaluated in the new study was not optimally designed to prevent common pediatric medication errors, such as the use of weight-based dosing calculations to prevent dosage errors. At the time of the study, the CPOE system required the user to select the pediatric version of the medication to engage weight-based dosing. If the unspecified form of the medication was chosen, this calculation was not done.

The study was funded in part by the Agency for Healthcare Research and Quality (HS13333). More details are in "Effect of computer order entry on prevention of serious medication errors in hospitalized children," by Kathleen E. Walsh, M.D., Dr. Landrigan, William G. Adams, M.D., and others, in the March 2008 Pediatrics 121(3), pp. e421-e427.

Return to Contents
Proceed to Next Article