This article is more than 5 years old. Information may no longer be current.
Preparation may determine whether using computerized prescription software translates into clinical benefit
Three hour training classes for physicians, three months before system computer order entry software implementation may not adequately prepare users and may result in unfavorable clinical outcomes, results of a recently published meta-analysis suggest.
“House staff cannot learn enough in just three hours, and three months later they probably will have forgotten most of what they did learn,” researchers from several sites in the Netherlands wrote.
The researchers pooled data from12 observational studies, including pediatric non-ICU, PICU, NICU and adult-ICU patient populations to determine the effect of computerized prescription software on medication errors, incidence of adverse event and mortality rates.
Although the researchers observed an overall reduction in medication errors across all studies (RR: 0.08, 95% CI: 0.01-0.77), separately pooled data from the studies involving pediatric and neonatal patients revealed an insignificant decrease in the number of potential and actual adverse events (RR: 0.65, 95% CI: 0.40-1.08) and mortality (RR: 1.02, 95% CI: 0.52-1.94).
In a study of critically ill pediatric patients, Han et al observed increases in mortality rates after implementing automated prescription software. However, steady mortality rates were observed by Del Beccaro et al before and after implementation of the same system.
Despite age and illness severity differences among the two study populations, several reasons have been proposed for the increased mortality rates observed by Han et al:
- Inability to register patients during transport.
- Increase in time needed to enter orders.
- Reduced verbal communication.
- Drug supply shift from hospital wards to centralized pharmacies.
- Network connection problems.
“The implementation process of computerized physician order entry systems requires specific attention because this may be associated with adverse outcomes,” the researchers wrote. They called for more multicenter studies to better understand the way that this software can be used to improve hospital care, specifically for pediatric and critically ill patients. – by Nicole Blazek
van Rosse F. Pediatrics. 2009;doi:10.1542/peds.2008-1494.
As medication errors are a significant cause of morbidity and mortality in U.S. health care institutions, additional data on methods to reduce medication errors are welcome. This is especially true in pediatrics, as less data are available on medication errors and methods that may decrease them in this population. Data from the study by van Rosse et al suggests that more clinical studies are needed to evaluate computerized prescription order entry.
– Edward A. Bell, PharmD
Drake University, Des Moines