NURS 6401 Week 6 decision support review example

Keep It, but Move It: A Four-Criterion Review of the Duplicate Therapy Alert That Stops Nurses Entering Protocol Acetaminophen

Student Name

College of Nursing, Walden University

NURS 6401: Informatics in Nursing and Healthcare

Instructor Name

Month Day, Year

Keep It, but Move It: A Four-Criterion Review of the Duplicate Therapy Alert That Stops Nurses Entering Protocol Acetaminophen

The Alert and the Criteria

Our hospital's neuroscience unit lets nurses enter acetaminophen under a standing fever and pain protocol. When a nurse signs the protocol order, a duplicate therapy alert fires if the patient already has an active acetaminophen order from a prescriber, including combination products. The alert shows the existing order and asks the nurse to cancel the new one or override with a reason. It fires often, and nurses override it often. This review judges the alert against four criteria set here and applied in the same order throughout: whether it reaches the person who can act, whether it arrives while the decision is still open, whether it offers an action rather than only a warning, and what it costs when it is wrong. The first three follow the widely used principle that decision support should get the needed information to the person who can use it, in a usable format, through a suitable channel and at the point in the workflow where it can still change the decision (Osheroff et al., 2012). The fourth is added because an alert that is often wrong teaches its users to ignore it, which is a harm of its own.

Criterion 1: Does It Reach the Person Who Can Act?

Partly. The alert reaches the nurse, who can act on the new protocol order by canceling it. But in many cases the problem is not the new order. A patient has an as-needed oral acetaminophen order from admission and a scheduled combination opioid product from surgery, and the nurse wants to give acetaminophen for a fever. The duplication that matters is between the two prescriber orders, and the nurse cannot change either. The alert interrupts someone who can prevent a third order but not fix the first two, and it offers no route to the person who can.

Criterion 2: Does It Arrive While the Decision Is Open?

No. By the time the alert fires at signing, the nurse has already assessed the patient, checked the protocol, selected the order and entered the dose. The decision to give acetaminophen has been made; the alert only asks the nurse to reverse it. Decision support that arrives after the decision is made invites dismissal, and published guidance recommends fitting support into the user's workflow at the point where the choice is actually made (Bates et al., 2003). Here that point is earlier, when the nurse opens the protocol.

Criterion 3: Does It Offer an Action?

Only a negative one. The alert offers cancel or override. It does not offer to open the existing order, show when the last dose was given across all acetaminophen-containing products or send a message to the prescriber. A nurse who wants to do the right thing, which is usually to give the dose already ordered or ask for the orders to be consolidated, has to leave the alert and do it manually.

Criterion 4: What Does It Cost When It Is Wrong?

Two kinds of cost. When the alert fires appropriately but the nurse overrides it, the patient may receive two doses too close together, and the total daily dose can creep toward the maximum. When the alert fires in a situation that is not a real duplication, such as an order that was already discontinued but still displays as active until pharmacy verifies it, the nurse loses time and the alert loses credibility. Interruptions of this kind carry their own safety cost, since research on interruption links it to errors in the interrupted task (Li et al., 2012).

What the Overrides Tell Us

Across the literature on drug safety alerts in order entry, clinicians override between 49% and 96% of alerts, and many overrides are justified; the review that reported this recommended focusing on error-producing conditions in the software and organization, such as low specificity and unnecessary workflow disruption, rather than on the clinicians who override (van der Sijs et al., 2006). Read the same way, repeated overrides on our unit are information about the alert. Most override reasons that nurses type fall into three groups: the existing order is a combination product they did not intend to duplicate, the existing order is for pain and the new one is for fever, or the other order has been discontinued. Each group points to a design change rather than a training need.

Limits of This Review

This review judges one alert on one unit, and three limits should be stated. First, the override reasons come from free text that nurses type quickly, so they show why nurses say they override, not always why they do. Second, the review has not measured harm: no event reports on the unit in the past year involve an acetaminophen duplication, which may mean the alert works or may mean near misses go unreported. Third, the redesign depends on changes to how prescriber orders are displayed, which belong to a different governance group than nursing documentation. The recommendation below is written so that the nursing parts can proceed even if the prescriber-facing parts take longer.

Recommendation

Keep the alert, because a real duplicate acetaminophen order can cause harm and nurses do sometimes catch one here. Change three things. First, narrow the audience: suppress the interruptive alert for orders that pharmacy has already discontinued, and route a non-interruptive message to the prescriber when two active prescriber orders duplicate each other, since the prescriber is the person who can act. Second, move the firing point earlier: when the nurse opens the protocol, display all active acetaminophen-containing orders and the total given in the past 24 hours before any selection is made. Third, attach actions: from that display, allow the nurse to document a dose against an existing order or send a consolidation request to the prescriber in one step. After three months, the unit should compare the override rate and the number of 24-hour totals above the protocol limit with the three months before the change. If overrides fall and totals do not rise, the redesign has done its job.

References

Bates, D. W., Kuperman, G. J., Wang, S., Gandhi, T., Kittler, A., Volk, L., Spurr, C., Khorasani, R., Tanasijevic, M., & Middleton, B. (2003). Ten commandments for effective clinical decision support: Making the practice of evidence-based medicine a reality. Journal of the American Medical Informatics Association, 10(6), 523-530. https://doi.org/10.1197/jamia.M1370

Li, S. Y. W., Magrabi, F., & Coiera, E. (2012). A systematic review of the psychological literature on interruption and its patient safety implications. Journal of the American Medical Informatics Association, 19(1), 6-12. https://doi.org/10.1136/amiajnl-2010-000024

Osheroff, J. A., Teich, J. M., Levick, D., Saldana, L., Velasco, F. T., Sittig, D. F., Rogers, K. M., & Jenders, R. A. (2012). Improving outcomes with clinical decision support: An implementer's guide (2nd ed.). HIMSS.

van der Sijs, H., Aarts, J., Vulto, A., & Berg, M. (2006). Overriding of drug safety alerts in computerized physician order entry. Journal of the American Medical Informatics Association, 13(2), 138-147. https://doi.org/10.1197/jamia.M1809