Indication-based prescribing prevents wrong-patient medication errors in computerized provider order entry (CPOE)

From Clinfowiki
Revision as of 15:00, 19 February 2015 by Rorr (Talk | contribs)

(diff) ← Older revision | Latest revision (diff) | Newer revision → (diff)
Jump to: navigation, search

These are the reviews for Galanter, Falck, Burns, Laragh, and Lambert's 2013 article, Indication-Based Prescribing Prevents Wrong-Patient Medication Errors in Computerized Provider Order Entry.[1]

First Review

Computerized provider order entry (CPOE) systems allow providers to electronically enter patient services and medication orders. With the introduction of the American Recovery and Reinvestment Act (ARRA) and Health Information Technology for Economic and Clinical Health (HITECH) Act in 2009, financial incentives were provided to individual physicians and hospitals to increase the health IT implementation,including CPOE systems in hospitals which makes them more capable of reducing medical errors and adverse drug reactions, especially when linked to clinical decision support systems (CDSS). The article, "Indication-based prescribing prevents wrong-patient medication errors in computerized provider order entry (CPOE)”, from the Journal of the American Medical Informatics Association, demonstrates whether indication-based computer order entry alerts can intercept wrong-patient medication errors. [1]

Background and Method

The Institute of Medicine(IOM)([1]) estimates that ,on average, hospitalized patients are subject to at least one medication error per day.[2] The IOM estimates that at least a quarter of all medication errors are preventable, and recommends electronic prescribing (e-prescribing) through a CPOE system to reduce medication errors and patient harm.[3] CPOE systems are known for their cost savings, decrease of medication errors and ability to identify adverse drug reactions for the healthcare systems as part of federal regulations. However, CPOE systems can also increase the risk of opportunities for providers to accidentally enter orders on wrong patient charts if they do not have visual cues . Wrong-patient selection occurs if a system has features that allows providers to search for patients charts rapidly, through "recent" patient list selections or through the ability to have multiple charts open during one session. The article describes how the UI-Health participated in a study, in which they created a group of CDS alerts to "prompt prescribers (providers) to add problems to the problem list when they were prescribing certain medications in the absence of certain documented problems." The alerts were aimed to improve the provider's situation awareness and patient safety. In the study, not only did the medication trigger an alert if the problem list was missing an active problem, it also offered a list of problems that were linked to the medication and gave the provider the ability to add them at their discretion.


Over the 6 year period of the study, the system fired 127,320 alerts for 79,304 encounters from 54,608 unique patients.[1] After further review only 32 intercepted wrong chart errors were identified within the 127, 320 alerts.[1] Overall the study illustrated the concept of medication-problem list mismatches which aims to improve problem list documentation within the electronic medical record (EMR). The article also indicates that although the alerts prevent providers in entering wrong-patient errors, one of the better solutions is to restrict the providers the ability to have more than one patient chart open at a time. In addition, including a photo file of the patient on their chart may be beneficial as well to helping them identify and preventing an error.


Although, the authors did a good representation of demonstrating the functionality of CDS rules when linked to CPOE systems and the effect of preventing wrong-patient errors, the study was limited to only one single medical center facility and used only a group of certain medications (without variation) over a 6 year period. If there had been a rotation of medications within the years, the interception rate could have differed than what was reported in the article. Although it is evident that CPOE systems can increase the risk of wrong-patient errors within EMRs, these type of errors are more evident when entered directly by the user rather than when prompted by the system. In conclusion, the indication-based alerts not only assisted in preventing wrong-patient errors but also in the improvement in the documentation of the problem list more so than expected.

Second Review


The University of Illinois Hospital and Health Sciences System has a computerized provider order entry (CPOE) system with clinical decision support which could detect whether a medication being prescribed is appropriate for the patient depending on his/her problems list. If a medication being ordered does not correspond to the patient’s listed problems, the CPOE would trigger an alert, asking the provider to either: a) enter a corresponding diagnosis, b) proceed with the order without entering a diagnosis, or c) cancel the order. The authors wanted to see whether the cancelled orders were near-miss events of wrong-patient order entry prevented by the CPOE alert.


The authors analyzed almost 6 years worth of patient charts and looked for patterns where: a) a clinician cancelled a medication order after an alert, and b) the same clinician prescribed the same medication for a different patient within 10 minutes of cancelling the order. If a patient chart met these two criteria, two clinicians would then review the chart. If they see that the cancelled order was: a) not a duplicate of a medication the patient is currently taking, and b) not relevant to the patient’s documented condition, they counted that as a near-miss wrong-patient order entry.


The authors found that out of 127,320 alerts triggered in the six-year period, there were 32 near-miss wrong-patient order entries.[1] They noted that in 60% of the events, the clinician had more than two charts open at a time. None of the alerts involved patients with the same last name.


This study shows that CPOE, especially when coupled with CDS, is an effective method of preventing prescription errors. It was not only able to ensure that the patient’s problems list was updated, it also triggered the prescriber into verifying that he was ordering the right medication for the right patient.

I agree with the authors on their comment about the need to study the EMR feature of allowing users to have more than one chart opened at a time. One thing I noticed is that more near-misses were attributed to housestaff (77%) than to attending physicians (18%). Is this because housestaff put in more orders than attending physicians, or could it be attributed to their differing levels of medical practice and workflow mastery?

I agree that usage of CPOE reduces medication errors in hospitals.

The benefits of CPOE can reduce medical errors and adverse drug effects especially when CPOE systems are bundled with clinical decision support systems designed to alert physicians and other healthcare providers of pending lab or medical erros. However, CPOE systems face major barriers associated with adoption in a hospital system, mainly high implementation costs and physician resistance to change. [4]


  1. 1.0 1.1 1.2 1.3 1.4 Galanter, W., Falck, S., Burns, M., Laragh, M., & Lambert, B. L. Indication-based prescribing prevents wrong-patient medication errors in computerized provider order entry. Journal of the American Medical Informatics Association doi: 10.1136/amiajnl-2012-001555.
  2. Aspden P, Wolcott J, Bootman J, et al. Preventing medication errors. Washington, DC: National Academic Press, 2007
  3. Committee on Quality of Health Care in America Crossing the quality chasm: a new health system for the 21st century. Washington, DC: Institute of Medicine, National Academy Press, 2001
  4. Charles, K., Cannon M., Hall R., Coustasse A. Can Utilizing a Computerized Provider Order Entry (CPOE)System Prevent Hospital Medical Errors and Adverse Drug Events.