Effects of computerized physician order entry and clinical decision support systems on medication safety: a systematic review
The effect of CPOE and decision support on medication safety.
Over the past few years it has been noted that medication errors, potential adverse drug events, and adverse drug events (ADE) are a significant problem within the health care system. The actual incidence of ADE is difficult to estimate due to inadequate reporting and varying classifications for drug errors but it has been reported to be in the range of 2 to 7 per 100 admissions. It is also reported that 28% of ADE are due to medication errors and that these errors occur at the ordering stage in 58% of the cases. It has been suggested that computerized physician order entry (CPOE) and computerized decision support systems (CDSS) would significantly reduce errors at the ordering stage and in turn reduce the incidence of ADE.
Kaushal et al. conducted a study to systematically review the evidence on the effect of CPOE and CDSS on medication safety. They searched Medline and the Cochran Library using MeSH terms and title keywords and reviewed the results to identify other primary studies. They selected 11 studies ranging from randomized controlled trials to observational studies with controls. Of the 5 studies looking at CPOE systems with CDSS, two showed statistically significant reduction in medication errors but the reduction in ADE was not statistically significant. The other 3 studies looked at specific types of medication errors and two found a statistically significant improvement in the study area. Seven studies were identified for the CDSS part of the study and they varied on the measured outcome. Three of these studies did show statistically significant results and the other showed non-statistically significant improvements but they were limited by small study size.
Researchers noticed that most evidence supporting CPOE and CDSS comes from institutions with homegrown systems. They also noticed that although there is evidence supporting the reduction in medication errors, the reduction in actual ADE is less clear but it should be expected based on the current available results. The researchers suggested that it would be more beneficial to study the differences between different systems especially that most hospitals will be implementing commercial systems which haven’t been properly evaluated or compared to systems in the literature. It was also noted that the benefits of CPOE and CDSS systems extends beyond reducing medication errors. Many institutions have reported significant reduction in medical care costs and also savings from the reduction of ADE. It should also be kept in mind that as users trust these systems it is very important for these systems to provide accurate information and high quality support to avoid leading to higher rates of ADE.
Comment: This review shows that CPOE and CDSS have beneficial effects in an important area where errors are frequent. It also highlights that most of these results come form few institutes with home grown systems and it is not clear how commercial systems will compare. It is important to identify the key success factors to help evaluate commercial systems and avoid unintended negative consequences.
One important question from the authors was “whether the currently available data were sufficiently compelling that CPOE should be widely adopted or whether further research was required.”
The authors used the U.S. National Library of Medicine, Medline electronic bibliographic database and the electronic Cochran library for their literature search. They identified and evaluated articles that described computerized systems for performing general order entry or CDS systems for guiding physicians in the order-writing process. All articles identified were grouped into two categories. One category evaluated CPOE with CDS systems while the other evaluated CDS systems alone.
The category of articles that evaluated CPOE with CDS systems showed 25 percent improvement in ordering, 55-86 percent decrease in medication errors, 13 percent decrease in inappropriate dose, 24 percent decrease in inappropriate frequency, and improvement in prescribing practices. The category of studies that evaluated isolated CDS systems showed a decrease in ADEs, lower rates of drug toxicity and fewer interventions.
The studies evaluated in this systematic review provide some evidence that medication errors and serious medication error rates are significantly decreased with the use of CPOE and CDS systems. Although the effect of ADE rates has not been adequately tested, considering the strong correlation between medication errors and ADEs, the use of CPOE and CDS systems also appears to have the propensity to decrease ADEs.
A great number of injuries are caused by medication errors and ADEs. The primary driving force behind the adoption of CPOE and CDS systems is patient safety and better clinical outcomes. It has been shown that the use of CPOE and isolated CDS systems significantly decreases medication error rates and provides other important benefits related to medication use. Many barriers to CPOE adoption exist, and these barriers have made organizational adoption to CPOE very limited. The authors believe that further studies are needed to clarify a few unanswered questions in order to promote the widespread adoption of CPOE. In addition, more investigation is needed with regard to the efficacy of individual CDS elements.