Impact of a clinical decision support system for high-alert medications on the prevention of prescription errors
This is a review of the paper by Lee, et al. 
The authors evaluated the introduction of a CDS for high-risk medications in the setting of a very large (2700 bed) tertiary medical center. The goal was to improve the safe administration of five medications: regular insulin, potassium chloride, and the anticoagulants warfarin, heparin and urokinase.
Prescribing data, and intervention log data was analyzed for the six months prior to, and following deployment of the CDS.
Improvements were noted in ordering dilutents for IV medications and maximum-dose monitoring. Orders that omitted dilution fluids for insulin and KCl dropped from 12,878 to 0 cases. Orders that exceeded the maximum dose of a drug dropped from 214 to 9 cases. Following adjustments in the program, no orders exceeded the maximum-dose recommendation.
Implementing the CDS did improve safe ordering of several high-risk medications. Some minor errors in the program had to adjusted after deployment, and there were some unusual errors noted that appeared to be user workarounds.
Data on effectiveness of CDS continues to accumulate. This paper demonstrates the achievement of a reasonable goal. It is perhaps more interesting to look at the unexpected consequences of CDS implementation to help other entities plan such a deployment.
- http://dx.doi.org/10.1016/j.ijmedinf.2014.08.006 Lee, J., Han, H., Ock, M., Lee, S., Lee, S., & Jo, M.-W. (2014). Impact of a clinical decision support system for high-alert medications on the prevention of prescription errors. International Journal of Medical Informatics, 83(12), 929–940.