View source for Criteria for assessing high-priority drug-drug interactions for clinical decision support in electronic health records

This is a review of Phansalkar, S., Desai, A., Choksi, A., Yoshida, E., Doole, J., Czochanski, M., Tucker, A. D., Middleton, B., Bell, B., & Bates, D.W. 2013 article, “Criteria for assessing high-priority drug-drug interactions for clinical decision support in electronic health records”.<ref name="Phansalkar, 2013"> Phansalkar, S., Desai, A., Choksi, A., Yoshida, E., Doole, J., Czochanski, M., Tucker, A. D., Middleton, B., Bell, B., & Bates, D.W. (2013). ''Criteria for assessing high-priority drug-drug interactions for clinical decision support in electronic health records''. BMC Med Inform Decis Mak., 13(1), doi: 10.1186/1472-6947-13-65 Retrieved from: http://www.biomedcentral.com/1472-6947/13/65 </ref>

== Abstract ==
The purpose of this article was to shed some more light in regards to the high override rates for [[Drug-drug interaction|drug-drug interaction]] (DDI) alerts in [[EHR|electronic health records]] (EHRs). The authors explained that due to high DDI alert overrides many providers are increasingly ignoring those alerts which could be dangerous when administrating patient care. This is called [[Alert Fatigue]]. There is also a lack of uniformity criteria for determining the severity and validity of alerts. Therefore, the authors of this article explore this area of care and identify a set of criteria for assessing DDIs that can be used for [[CDS|clinical decision support]] (CDS) alerts in EHRs.

==Methods ==
The authors of this article conducted a literature review for the span of 20 years. The dates were from January 1990 to December 2010. They reviewed articles on MEDLINE and EMBASE databases to identify characteristics of high-priority DDI alerts. The method in which the authors sorted through their research was that they looked for the presence of specific keywords and MeSH terms in two categories within the title, abstract and body of every article. Each reviewer extracted criteria from relevant studies on how the evidence on DDIs could be filtered or tailored to identify a high priority DDI. Once the articles were chosen a panel of experts were then used to validate the data gathered from the systematic literature review. To validate their findings a panel of experts were used in which consisted of a wide range of experience within the subject matter. The panel expertise consisted of medication knowledge base vendors, EHR vendors, in-house knowledge base developers from academic medical centers and agencies involved in the regulation of medication use both federal and private state levels.

==Results and Discussion ==
There were 44 articles that were selected and reviewed in which met the inclusion criteria for assessing characteristics of high-priority DDIs. As a result the panel proposed 5 items that were deemed to be the most important when assessing an interaction. They are the following:

*'''Severity:''' More research is needed to understand alert variation for assessing the severity of an interaction.
*'''Probability:''' Probability is harder to determine without knowing the patient context. In the case of an ADE it is suggested to include the concentration response curve for the [[Adverse drug event| Adverse Drug Events]] (ADE) of interest.
*'''Clinical Implications of the interaction:''' It is suggested that consideration of management burden of the interaction, and the awareness of the provider regarding the interaction.
*'''Patient characteristics:''' Better integration of the medication knowledge base and the EHR of the patient.
*'''Evidence supporting the interaction:''' Inclusion of information from the [[U.S. Food and Drug Administration (FDA)|U.S. Food and Drug Administration]] (FDA) regarding medication and treatment guidelines in abbreviated form would improve assessment of the evidence in order to make appropriate decisions regarding DDI alerts.

==Comments ==
This article was a very interesting read in regards to understanding how DDI alerts are becoming a problem in the clinical setting. It also interesting to point out that it is not enough to have a [[CDS| Clinical Decision Support]] (CDS) system and EHR to improve patient care. Investing in personalizing CDS criteria prior to implementation would be most beneficial. Assessing a system with the 5 proposed criteria mentioned above will guide the practice in the right direction. Physicians and clinical providers will not be bombarded with alerts and alert fatigue will decrease.

= References =
<references/>

[[Category:Reviews]]
[[Category:EHR]]
[[Category:CDS]]