NR 360 Week 4 Wrong-Patient Medication Case Analysis Example

A Correct Scan of the Wrong Order: A Case Analysis of a Wrong-Patient Beta-Blocker Dose

Student Name

College of Nursing, Chamberlain University

NR 360: Information Systems in Healthcare

Instructor Name

Month Day, Year

A Correct Scan of the Wrong Order: A Case Analysis of a Wrong-Patient Beta-Blocker Dose

The Case, Told Once

The case is a composite, assembled for this analysis from patterns described in the literature and in the writer's experience on a medical unit. At 1710 on a busy weekday, a hospitalist on a 30-bed medical unit had the electronic records of two patients open in separate windows. Patient A, in room 12, was in fast atrial fibrillation, and the hospitalist intended to order oral metoprolol 50 milligrams now. Patient B, in room 14, was an 81-year-old woman admitted with dehydration and a heart rate in the high 50s, on no rate-controlling medication. The hospitalist entered the metoprolol order in the window that was active at the moment, which was Patient B's. The system displayed a warning that the patient's most recent heart rate was below 60; the hospitalist, who saw this warning several times a day for patients on established beta-blockers, dismissed it without reading it and signed the order.

At 1740, Patient B's nurse received a notification of a new stat order, pulled the medication from the automated dispensing cabinet under Patient B's profile, scanned Patient B's wristband and the medication at the bedside, and received a green confirmation. She noticed that metoprolol was new for this patient and that the patient's heart rate on the monitor was 58. She paused, but the order was signed, the scan matched and the patient was due for her 1800 medications, so she gave it. At 1930 Patient B's heart rate was 44 and her blood pressure 82/50. The error was discovered when the hospitalist, reviewing Patient A's chart, found no metoprolol order.

What Was Known at Each Point

At 1710, the hospitalist knew which patient he meant to treat but not which window was active; the screen showed the patient's name in a banner at the top, in the same font and color for both patients. He knew the warning existed but, through repetition, no longer read it. At 1740, the nurse knew the order was new and that the heart rate was low, but she did not know that the order had been meant for someone else, and every signal the system gave her said the administration was correct: the order was signed by a physician, the cabinet released the drug under her patient's name and the scanner confirmed a match. It is only in hindsight, with the knowledge that the order was placed on the wrong patient, that the nurse's pause looks like the moment the error should have stopped. At the time, the information she had pointed toward giving the dose.

Technology as a Participant

Four features of the technology are conditions of this error rather than background to it. First, the order entry system allowed two patients' records to be open at once, with the active window identified only by a banner. Research on wrong-patient orders has shown that such errors are common enough to be measured and that simple design changes can reduce them; in a randomized trial, requiring clinicians to confirm the patient's identity before entering orders reduced wrong-patient orders (Adelman et al., 2013). Second, the heart rate warning fired so often for appropriate orders that it no longer carried information. Third, the barcode system did exactly what it verifies, which is that the medication matches an active order for the patient scanned; it does not verify that the order was intended for that patient. Fourth, the green confirmation on the scanner gave the nurse a signal of correctness strong enough to outweigh her own clinical concern. Koppel et al. (2005) described how computerized order entry systems can facilitate errors in exactly this way, through fragmented displays, alert designs and assumptions that the system's checks cover more than they do.

The Chain, Link by Link

For this outcome to occur, every one of these conditions had to hold. Two charts had to be open at once. The active window had to be the wrong one when the order was entered. The warning had to be dismissed without reading. The dispensing cabinet had to release the drug on the order alone, which it is designed to do. The scan had to confirm the match, which it did correctly. And the nurse's pause had to be resolved in favor of the order rather than a call to the physician. Remove any one link and the error would likely have been caught. Reason (2000) describes this as the Swiss cheese model of accidents, in which several layers of defense each have holes, and harm occurs only when the holes line up. The value of the model here is that it points to several places where a hole could be closed, rather than to one person who should have been more careful.

Beyond Human Error

It would be easy to stop at human error: the hospitalist entered the order in the wrong chart, and the nurse gave a drug to a patient with a low heart rate. Both statements are true, and both clinicians would say so. But stopping there leaves every condition in place for the next occurrence. A more useful conclusion is that the system made the wrong action easy and the right action hard. The two-window display made the entry error likely, the overused warning made the prescriber's check useless, and the scanner's green light made the nurse's doubt look unfounded. The nurse's pause is the most important moment in the case, not because she should have known, but because it shows that she noticed a mismatch that the system did not. The practical question is how to make that kind of pause lead to a call rather than to a dose.

What Would Change the Chain

Three changes would each close a link. The order entry system could restrict prescribers to one patient's record at a time for medication orders, or require identity confirmation before signing, as tested by Adelman et al. (2013). The heart rate warning could be redesigned to fire only when a rate-controlling drug is new for the patient, which would make it rare and meaningful again. And the unit could adopt a practice that a new cardiac, anticoagulant or insulin order that surprises the nurse, especially one that does not fit the patient's condition, triggers a short call to the prescriber before administration, with no expectation of blame for the call. Ash et al. (2004) note that information systems reshape how clinicians communicate, and the third change restores a conversation the system had quietly replaced with a green light.

References

Adelman, J. S., Kalkut, G. E., Schechter, C. B., Weiss, J. M., Berger, M. A., Reissman, S. H., Cohen, H. W., Lorenzen, S. J., Burack, D. A., & Southern, W. N. (2013). Understanding and preventing wrong-patient electronic orders: A randomized controlled trial. Journal of the American Medical Informatics Association, 20(2), 305-310. https://doi.org/10.1136/amiajnl-2012-001055

Ash, J. S., Berg, M., & Coiera, E. (2004). Some unintended consequences of information technology in health care: The nature of patient care information system-related errors. Journal of the American Medical Informatics Association, 11(2), 104-112. https://doi.org/10.1197/jamia.M1471

Koppel, R., Metlay, J. P., Cohen, A., Abaluck, B., Localio, A. R., Kimmel, S. E., & Strom, B. L. (2005). Role of computerized physician order entry systems in facilitating medication errors. JAMA, 293(10), 1197-1203. https://doi.org/10.1001/jama.293.10.1197

Reason, J. (2000). Human error: Models and management. BMJ, 320(7237), 768-770. https://doi.org/10.1136/bmj.320.7237.768