Optimizing the acceptance of medication-based alerts by physicians during CPOE implementation in a community hospital.

From Clinfowiki
Jump to: navigation, search

A review of article written by Stutman et al (2007) "Optimizing the acceptance of medication-based alerts by physicians during CPOE implementation in a community hospital." [1]


The article reported that a community hospital with no mandate for system use faces challenges in implanting a CPOE system with CDS such as medication interaction alerts. The article added, CDSs are important in benefit realization; however, clinicians may perceive those support systems as a deterrent on the clinical workflow. Therefore, to achieve maximum acceptance and successful deployment, the authors monitored “frequency of alert presentation, frequency of “positive response to the alert and physician satisfaction with each of those interactions”. In a response to monitoring the system, implementers had to change different aspects of the alert system so that the system can be adopted smoothly.


The article was written as a Memorial Health Service System implemented in patient focused EHR across the organization. Here the goal was to improve patient safety and quality of care through the adoption of CPOE. In the previous system the article described, “medication orders were entered into a legacy order entry system, then printed in the pharmacy and reentered into a disparate pharmacy system.” This process created a great concern for providers considering the amount of error that can be introduced as information transferred from one system to another manually. Besides doctors and nurses didn’t have any medication based alert, pharmacists had only drug- drug interaction alert. The allergy alert was presented to the pharmacist only if that information was reentered manually into the pharmacy system. On the other hand, the new system consists of a variety of medication-based alerts to physicians. Meanwhile, the implantation required a thoughtful approach to increase system’s acceptability and usability by physicians in addition to involving them actively throughout the process.

Strategic Decision

Physician steering committee strategic decision

  • During the implementation the committee decided to deploy a limited number of alerts so that system acceptability increases while alert fatigue decreased.
  • Increase the number and category of alerts as the system becomes familiar to users.
  • Use of the same alert for physicians, pharmacist and other intermediate providers.

The above strategy was to achieve a process that leads to reconsideration or deletion of medication orders at least 25-30% of the time. In an attempt to tackle the target physician steering and medication management team agreed to deploy drug alerts, which are easy to understand and have greater impact on a large number of problematic orders, while holding on a more complicated alerts. Further, the steering committee continued fine-tuning medication orders by removing some alerts that are deemed to be inappropriate or inefficient.

Data Extraction and Analysis

According to the article, data extraction and analysis included, collecting relevant data such as: “ frequency of alert presentation, action taken on alert presentation, and CPOE utilization statistics.” Those information were analyzed and presented in a form of chart for better understanding of system utilization.


In the first six months after deployment of the system, there was a high frequency of alerts and actions such as removal and reconsideration of medication orders ranging from 32.7% to 66%. Based on information derived from system monitoring, adjustments were made to the number, category and display of alerts. However, despite their best efforts, dose alert responses continued to be low, which led to further adjustments of the dosage range for over 1000 drug records. This slightly improved the number of alert responses with regard to dosing. Even so, overall the frequency of alert presentation decreased and the frequency of order modification continued to grow in an acceptable range. On the other hand, the frequency of alert display had been a source of physician frustration. To tackle this challenge, filtering became stringent and was focused on high-risk interactions. As a result of this continuous monitoring and adjustment of alerts, the implementers were successful in managing a steady rate of CPOE adoption despite the insertion of new alert categories and other drug interaction notifications.

On going challenges and concerns

Some of the ongoing concerns the authors reported and continued to work on are the following:


The article concluded “ Achieving and maintain an excellent rate of physician adoption of CPOE while implementing real-time medication order-base alert is a significant but not insurmountable challenge.” However, through continuous monitoring and subsequent adjustment on medication order alert, it was possible to attain a decent amount of success rate in adopting CPOE system by physicians. Finally the article noted that such monitoring and changes should continue, perhaps for an indefinite time in the post-go-live period considering possibilities of having new problems and challenges along the way.


To attain adequate levels of success in CPOE adoption, it is mandatory to maintain continuous monitoring and adjustment of CDS within the system based on feedbacks from monitoring phase.

Second Review


The study was done at a multi-hospital system with volunteer medical staff. The main goal of this article was to maximize the quality and patient safety through the adoption of CPOE. This new system allowed for the physicians to receive medication based CDS.


Data for alert presentation was categorized on the actions taken upon alert presentation. The categories were: alert overridden, order delete, alert cancelled or order further reconsidered or modified.


The study produced a total of 46,000 medication-order alerts displayed to physicians. The modification or removal of orders was 21-66% for the alert categories and CPOE adoptionwas to 75-78% during the length of the study compared to pre existing rates of 25-30%.


Integrating CPOE and CDS is a challenge, attention to how physicians use the system and making the appropriate changes can help keep physician usage. This has to be used during implementation and throughout the use of the system to identify challenges and help fix them.


This article suggest that if CPOE and CDS is continuously reviewed it is likely to improve the usage of the two.


  1. Stutman, H. R., Fineman, R., Meyer, K., & Jones, D. (2007). Optimizing the acceptance of medication-based alerts by physicians during CPOE implementation in a community hospital environment. AMIA Annual Symposium Proceedings, 2007, 701–705. http://www-ncbi-nlm-nih-gov.ezproxyhost.library.tmc.edu/pmc/articles/PMC2655789/